Pentland Hills

The Reservoir Formation consists principally of green-grey and purple mudstones and siltstones with interbedded sandy siltstone and very fine sandstone. General fossil abundance is low, with exceptional concentrations in particular horizons rather than a uniformly rich succession. The BGS minimum estimate of 1,080 m refers to the North Esk Inlier formation, not one continuous river exposure; its true thickness is uncertain because beds are repeated and cut by small faults.

Lowest river and gorge exposures. Below the reservoir, interbedded siltstone, fine sandstone and laminated shale are tightly to recumbently folded. Robertson’s excursion distinguishes this early folding from later, more open kink folding. Farther downstream, the position of the angular unconformity separating deformed Silurian strata from younger Old Red Sandstone rocks can be mapped, but the contact itself is not exposed. This younger angular unconformity is separate from the conformable formation boundaries within the Silurian sequence.

South-eastern and eastern reservoir-shore beds. Robertson describes individual sandstone and siltstone beds 1 cm to 1 m thick within a roughly 250 m shore section. Barron’s wider formation description records coarser interbeds up to 1.2 m thick. Sharp upper and lower contacts, massive or relatively coarse bases, ripple and convolute lamination and locally burrowed tops record repeated depositional events and fining-upwards beds. Rounded mud clasts and sheets of the enigmatic fossil Dictyocaris occur at the south-eastern reservoir corner; its biological affinity has been debated. Barron’s thin section from this corner contains quartz with abundant detrital muscovite and biotite, minor feldspar and chlorite, and phyllitic and volcanic rock fragments.

Red mudstones and higher northern-shore beds. In the older shore account, the south-eastern siltstones pass across a disturbed interval into fine red mudstone, with occasional sandstone bands higher up. The red mudstone was described as unfossiliferous. Open folding changes the strike along the eastern shore, while the northern shore includes tight fold closures and small faults. These structural changes are important when comparing separated exposures: apparent progress along the shore is not automatically uninterrupted progress up the stratigraphy.

Historical Gutterford subdivisions. The 1962 memoir distinguished basal fine grits and siltstones interbedded with laminated mudstone, the purple Gutterford Burn Mudstones, the Gutterford Burn Flagstones and overlying predominantly argillaceous beds. These are useful historical divisions, not formal members requiring separate coloured bars. The flagstones and lower overlying fine beds supplied many of the early graptolite records. Older accounts placed the Gutterford Limestone, Eurypterid Bed and starfish horizons in Reservoir; the 2007 boundary revision transfers the fossiliferous sequence above the first calcareous sandstone into lower Deerhope. Consequently, the old thickness stack and older undifferentiated crenulata-zone wording cannot be applied unchanged to the revised formation.

Within that historical scheme, the GCR summarises the basal exposed grit–siltstone interval as seen to about 50 m, the Gutterford Burn Mudstones as about 125 m, the Gutterford Burn Flagstones as about 120 m and the succeeding grey-to-olive mudstones as more than 200 m. The mudstone division includes rare flaggy sandstone beds; the flagstone division alternates buff to dark grey coarser beds with grey mudstone and siltstone; the succeeding argillaceous division retains thin flags near its base. These are old, geographically combined estimates with different boundaries from the modern detailed Gutterford log, not a continuous present-day exposure or extra thicknesses to add to the revised formations.

Fairliehope–Gutterford intrusion. A narrow, strongly altered intrusive body is recorded at the north-western reservoir corner and in the Gutterford succession. The excursion describes chloritised sub-basic dioritic rock, baked sedimentary contacts and an exposed width or thickness of about 2–4 m. It records the body as deformed with the Silurian rocks and older than the overlying Old Red Sandstone. This is an intrusion cutting the sedimentary succession, not an additional sedimentary bed or a member of a younger lava formation.

Gutterford local log: 0–3 m micaceous siltstones. The detailed 2007 account is based on about 23 m of strata, using neighbouring lateral exposures where vertical continuity was absent. Its metre values are positions in that local log, not heights above the stream or distances along a path. The overturned bedding places older rocks higher on the slope and younger rocks lower down. The lowest 3 m comprise green-brown micaceous siltstones with thin sandstone ribs, parallel and occasional ripple lamination, muscovite-rich laminae and carbonaceous sheets. Branching Chondrites-type burrows, about 10 mm wide, are concentrated in parts of the first metre.

Gutterford in-situ coral horizon at 2.5 m. Fine green siltstone contains colonies identified in the 2007 study as Favosites multipora preserved in life position, together with lingulid brachiopods, monograptid graptolites, stick bryozoans and a scolecodont. Sparse Glassia occurs in the surrounding micaceous-siltstone unit. This horizon lies below the revised Deerhope base and demonstrates that some corals grew on the sea floor here; it must be distinguished from the transported shell-rich calcareous beds above.

Deerhope is predominantly grey-green, grey-brown, purple and grey laminated or bioturbated mudstone and siltstone, with fossil-concentrating coarser beds. Its type sequence is in Deerhope Burn, with important comparison exposures in the North Esk. The often-quoted 390 m thickness belongs to the older definition used by Robertson and retained in the BGS Lexicon. Anderson and colleagues lowered the local Gutterford boundary in 2007 to the first calcareous sandstone, at 3.00 m in their measured log. Their revision includes calcareous sandstone within lower Deerhope, so an absolute description excluding all sediment coarser than siltstone is no longer appropriate to this revised succession.

Gutterford calcareous sandstone succession, 3.00–12.5 m. Eighteen calcareous beds occur through this 9.5 m local interval, separated by finely laminated green shale that is mostly unfossiliferous. The sandstone beds become thicker upwards and contain shelly lags of brachiopods, bivalves, corals, bryozoans and rare trilobites. They are calcareous sandstones and fossil-rich concentrations rather than eighteen beds of pure limestone. Possible hummocky structures in the 4–5 m interval were interpreted by the authors as evidence for storm influence; this remains an interpretation of subdued structures, not a diagnostic feature of every bed.

Weathered calcareous horizons at 5.2, 7.5, 7.7 and 9.5 m. Each of these recorded levels contains sticky brown clay associated with intense weathering and decalcification of the calcareous succession. Rusty, mouldic textures develop as carbonate is removed, especially towards bed tops and bases. Although volcanic material occurs elsewhere in the section, these four brown-clay horizons should not all be labelled proven primary ash beds.

Gutterford Burn Limestone at 9.90 m. The thickest calcareous bed in the 2007 log is the informally named Gutterford Burn Limestone of earlier accounts. Historical descriptions give about 20–25 cm for this shelly, calcareous-siltstone or sandy-limestone lag. Crinoid ossicles, broken corals and bryozoans, disarticulated brachiopods and bivalves, and less common gastropods, trilobites and ostracods form a concentrated marine assemblage. Its position above the new 3 m boundary places it in lower Deerhope under the 2007 scheme. The named bed is not a separate formal limestone member.

Fossiliferous tops of the calcareous beds. Sheets of Dictyocaris ramsayi, small articulated branching bryozoans and monograptids including Monograptus priodon occur on upper surfaces, interpreted as the ends of fining-upwards depositional cycles. Some black volcanic clasts, 1–2 cm across, are encrusted by bryozoans. The study also reports a conodont identification of Pterospathodus amorphognathoides communicated to the authors; that report is not a precise independent fix on the Llandovery–Wenlock boundary. The calcareous interval ends at the last sandstone at 12.5 m.

Green shale and repeated coral horizons, 12.5–15.4 m. Dark green, massive to locally finely laminated shale succeeds the calcareous sandstone interval. Favosites multipora and Palaeocyclus porpita, as identified in the study, occur at several levels with their calices facing upwards. These are repeated episodes of coral growth, not a single universal coral bed. A separate nodular horizon at about 13.7 m yielded sparse Glassia.

Graptolite flagstones from 15.4 m upwards. Six grey-green micaceous flagstones, typically 10–15 cm thick, alternate with dark green shale in the interval described as 2.5–5.9 m beneath the Eurypterid Bed. They are planar laminated and form resistant ribs; aligned mica, graptolites and sheets of Dictyocaris ramsayi record current influence. Dendroid graptolites, monograptids and Oktavites excentricus occur here, together with small intact starfish and an orthocone. Some individual flags were traced for tens of metres along strike. Lithological comparison places the historical Mirmor andreae material, revised as the octocoral Atractosella andreae, in this interval rather than identifying it as fish armour.

Hard sandstone 1.5 m below the Eurypterid Bed. A particularly hard, “flinty” sandstone contains weathered and decalcified clasts 10–20 mm across, now represented by rusty cavities. The 2007 authors recorded no fossils in this particular bed. Above it, green silts alternate with hard, massive micaceous sandstone before the clay underlying the Eurypterid Bed is reached.

Basal Eurypterid Bed ash. Immediately beneath the fossil-bearing claystone lies about 5 cm of soft, sticky pea-green clay. X-ray diffraction confirmed an altered volcanic ash, or metabentonite, rich in swelling clays. It fills an irregular underlying surface and provides a separate lower boundary marker for the exceptional-preservation unit. This Gutterford ash is much lower in the succession than the Wether Law Linn Main Ash Band.

Eurypterid Bed proper. About 55 cm of very finely laminated light olive-green claystone lies above the basal ash. Its smooth, waxy texture, carbonised arthropod cuticle and locally abundant Dictyocaris distinguish it from the surrounding rocks. Bed thickness varies across the roughly 5 m lateral extent investigated during excavation, and the deposit drapes and fills irregular sea-floor relief. The older GCR account described a 30 cm arthropod-bearing band; that earlier description should not replace the approximately 55 cm whole unit measured in the 2007 investigation. The latter paper places the bed approximately 19 m above the revised Deerhope base; the drawn log is not precise enough to replace that statement with a newly calculated centimetre-level position. Eurypterids, scorpions and other chelicerate arthropods, with articulated echinoderms and associated marine fossils, give the bed its exceptional scientific importance.

Internal and upper Eurypterid Bed horizons. A second thin green clay band occurs 8 cm above the claystone’s base. At 13 cm above the base, a distinct calcareous-rich horizon contains crinoid ossicles, stick bryozoans, monograptids and brachiopods, weathering to a rusty mouldic texture. The authors considered this the likely source of some historical articulated crinoids, rather than proving the bed of origin of every specimen. A third sticky green clay band marks the top of the Eurypterid Bed and drapes its irregular surface.

Post-Eurypterid sandstone and purple siltstone, 21.8–23 m. Lens-shaped, apparently channelised sandstones about 100–150 mm thick overlie the upper clay band. Rusty patches may represent degraded volcanic clasts. Hard blue-purple micaceous siltstone completes the exposed log, with fine lamination on a 3–4 mm scale and occasional sandy lenses. The logged rock sequence ends at the stream-bed exposure; the water-covered part above it in the figure is not an additional measured rock unit.

Preservation and the separate starfish records. Anderson and colleagues interpret the Eurypterid Bed as a normal-marine deposit below storm wave-base, where rapid burial by fine, ash-influenced sediment favoured exceptional preservation. Their investigation found starfish in micaceous siltstone, calcareous beds, graptolite flags and the Eurypterid Bed itself, so no single horizon could uniquely be called the Starfish Bed within their measured section. Older memoirs and excursions separately record two starfish-bearing horizons farther along Gutterford. Those historical records must not be merged into one exclusive marker or assumed to be the same exposed lens as the Eurypterid Bed.

Historical lower starfish horizon. The 1962 memoir describes a roughly 2 mm band of buff siltstone within grey mudstone and siltstone, near the base of the old post-flagstone argillaceous division. It records starfish under the identification Furcaster leptosoma, with sparse accompanying shelly fossils, and describes their preservation as poorer than at the classical upper locality. This is a source-specific historical horizon, not a newly located bed in the 2007 Eurypterid Bed excavation.

Historical upper starfish horizon. The classical higher locality is described in the memoir and excursion as fine micaceous silty shale with a diverse asteroid fauna and relatively sparse brachiopods and other shells. The older accounts include Crepidosoma wenlocki among the recorded starfish. Its significance rests on historical collections and stratigraphic descriptions; the named locality does not establish a single starfish-only bed through the entire Gutterford succession.

The Gutterford coral-bearing sequence and the Deerhope Coral Beds share environmental features, but the 2007 study expressly does not require them to be the same laterally continuous bed: similar conditions recurred at different times. A projected continuation of the Gutterford fossiliferous sequence towards the northern reservoir shore was not confirmed by detailed logging. An isolated starfish-bearing loose slab from the eastern shore likewise does not establish an in-situ correlated horizon.

Separate R199 starfish-bed investigation, 2008. Donovan and colleagues described two trenches through near-vertical beds striking about 020°. Units 1, 3, 5 and 7 were massive to weakly laminated siltstone and sandstone, alternating with grey-clay units 2, 4 and 6 interpreted as weathered bentonite. Neither section limit was exposed. Even targeted examination below the clays yielded only half a brachiopod valve. The authors suggested depletion of a small channel-fill lens, while considering possible mislocation; this did not establish exhaustion of all starfish horizons. These seven units belong to the separate lower-starfish-bed investigation, not the 2007 Eurypterid Bed log. No individual thicknesses are reconstructed here.

Deerhope Burn type succession and Coral Beds. Robertson’s original Deerhope section progresses from mudstone and shale with fine silty laminae into micaceous-siltstone beds about 0.10–0.15 m thick. Bioturbation is common, and shelly faunas become more varied upwards. Several thin fossil-concentrating siltstones in a side-stream are collectively termed the Deerhope Coral Bed or Coral Beds. They contain corals, crinoid debris, brachiopods, bivalves, bryozoans and trilobites; the name does not imply a massive coral reef. The modern brachiopod Isorthis (Ovalella) clarksoni is directly documented here, including the holotype NMS G.2006.40.18.

Coral Bed and higher North Esk bivalve horizon: separate historical positions. Robertson’s 1985 log places the Deerhope Coral Bed at outcrop R247, approximately 265 m up his original 390 m formation. His North Esk comparison places Lamont’s bivalve locality, R55, higher, around 300 m on that same historical scale. These are different fossil concentrations, not one coral bed simply traced across the valley. Neither numerical height should be read as a measurement above the revised 2007 Deerhope base.

Upper Deerhope shell-lag siltstones. The North Esk section contains thin siltstone beds, commonly 5–10 cm in the excursion account, with fossils concentrated at their bases. Bivalves, gastropods, brachiopods, corals, trilobite remains and crinoid or other pelmatozoan debris occur in these lags. Tool marks and lineations record currents broadly from the east, with erosion of cohesive sediment surfaces. Barron’s thin-section study supports primary sedimentary lamination and ripple-like structures rather than the older suggestion that every corrugated surface was produced by tectonic shearing. This is upper Deerhope, despite an inconsistent reference to Cock Rig within the older excursion’s locality description.

R55 in the 2011 crinoid study. Donovan and colleagues separately describe a roughly 12 m-long exposure with vertical bedding striking about 020°. Their account emphasises medium-grained grey sandstone, in 5–100 mm beds with millimetre-scale lamination, and rare purple siltstone. Crinoid-bearing samples came from loose material at its north-western end, not directly from a measured bed. This later sandstone description is retained alongside the older upper-Deerhope siltstone account; neither is silently substituted for the other.

The Cock Rig Formation, historically the Haggis Grit and Conglomerate and included in Tipper’s older Deerhope Formation, marks a conspicuous change to medium- and coarse-grained sandstone with conglomerate and thin mudstone partings. The base is the first coarse sandstone above Deerhope fine sediment; the upper boundary is the change from the highest sandstone to green-grey Wether Law Linn mudstone. The BGS formation range is 80–97 m, with 97 m in Robertson’s original measurement, rather than the height of a single quarry face. The North Esk riverbanks and old quarries and the Deerhope Burn expose different parts and lateral expressions of this interval.

North Esk sandstone and conglomerate beds. Robertson recorded at least fourteen conglomerate beds, mostly about 0.25 m thick with clasts up to 50 mm across, interspersed with sandstone beds 0.1–1 m thick. The conglomerates are clast-supported and poorly sorted or graded, with both fining- and coarsening-upwards intervals. Granite, porphyritic and volcanic rocks, quartzite, chert and fine sedimentary rocks contribute to the mixed clast assemblage. Low-angle cross-bedding, about 5–15° in his measured examples, shows reworking and varied flow directions; basal load structures and tool-marked surfaces are also recorded. These are historical measured observations, not fourteen separately named beds guaranteed to be countable in a present quarry exposure.

South-westward change and the upper sandstone. In Robertson’s Deerhope comparison, only four grit horizons about 0.15 m thick remained, with clasts up to about 15 mm. This strong lateral reduction explains why the river and burn cannot be combined into one invariant stack. Shale interbeds reach about 0.25 m in the original description. Towards the top, low-amplitude symmetrical ripples were recorded at Deerhope R260, while the North Esk upper sandstone at R79 shows trough cross-bedding.

Upper Cock Rig transported shell horizons. Most of the sandstone and conglomerate is sparsely fossiliferous and carries broken or disarticulated material, but shell concentrations increase towards the highest sandstone. Brachiopods, bivalves and pelmatozoan debris, including crinoid ossicles, form transported assemblages. The association formerly named for Isorthis mackenziei is now termed the Isorthis clarksoni–pelmatozoan association following the revised brachiopod work. It is an ecological and sedimentary association, not a time-zone. The older submarine-fan feeder-channel interpretation differs from later offshore-bar and nearshore models.

Green-grey bioturbated mudstone above the Cock Rig sandstone marks the base of this fossiliferous marine formation. Its published 230–265 m thickness is a composite formation range, assembled from discontinuous exposures and allowing for lateral change. The formal ascending members are Grain Heads Siltstone, Lamb Rig Siltstone and Baddinsgill Mudstone. These correspond respectively to Robertson’s Lower, Middle and Upper members. Tipper’s informal units a–c form Grain Heads, unit d corresponds to Lamb Rig and unit e to Baddinsgill. His a–e scheme, also lettered A–E in his figure, is different from the nineteenth-century lettered fossil localities and from Robertson’s numbered sampling intervals.

Red sandstone and conglomerate overlying Wether Law Linn record the establishment of predominantly continental conditions. The formation-wide minimum thickness is 725 m, and its full type succession extends through the Baddinsgill–Lyne Water area. Only part is represented by the North Esk and Henshaw Burn reference exposures. The formal BGS age range is Sheinwoodian–Homerian, within the Wenlock, while the lower succession has early Wenlock support from regional spore evidence. Broad alluvial-fan and river interpretations apply to the continental succession; minor marine influence recorded elsewhere in the formation should not be treated as proof of a marine fossil bed at every Henshaw exposure.

Igneous Conglomerate, informal Unit 1. North Esk exposures include red, clast-supported conglomerate, poorly bedded or arranged in beds up to about 1 m thick. Barron’s composite unit estimate is about 35 m, rather than the height of one river face. Rounded to subrounded cobbles and boulders reach about 0.3 m and include granite, microgranite, quartz-feldspar porphyry, felsic lava, altered sub-basic lava, chert, quartzite and fine-grained sedimentary rocks. Thin-section evidence from the river confirms abundant igneous fragments, including granophyric granite and texturally varied felsite. This Silurian conglomerate is distinct from the younger Greywacke Conglomerate Formation above the angular unconformity.

Lower red sandstone, informal Unit 2. The overlying red micaceous sandstones range from fine to coarse grained, with quartz-granule lenses, pebble strings and mud clasts. Barron’s formation-scale account gives beds commonly 0.1–0.5 m thick within a composite interval of about 250 m. North Esk exposures above the conglomerate contain medium-grained sandstone with occasional trough cross-bedding, recording current transport. The large unexposed gaps in the historical North Esk column prevent these estimates being presented as a continuously accessible 250 m section.

Quartzite Conglomerate in Henshaw Burn. Near the upper part of the lower red-sandstone interval, a distinctive conglomerate is about 1.5 m thick in the Henshaw Burn descriptions of the memoir, excursion and Barron. The excursion estimates roughly 70% quartzite clasts; Barron describes rounded to subrounded pebbles and cobbles, including haematite-veined white quartzite and occasional igneous and fine-grained sedimentary clasts. The 1962 account places the roughly 5 ft conglomerate above about 3 ft 4 in of laminated mudstone and siltstone at this local occurrence. Its composition contrasts with the lower Igneous Conglomerate. A GCR account quotes 7.5 m for a quartzite-conglomerate exposure in the North Esk River; this different reported exposure and measurement should not replace the supported 1.5 m Henshaw value.

Limit of the North Esk–Henshaw record. Robertson’s original North Esk R105–R112 column shows separated conglomerate and sandstone outcrops, large gaps and an uncertain upper limit. His longer comparison column for Baddinsgill and Lyne Water carries the Lyne Water Fish Bed marker; that marker is absent from the North Esk column. The detailed higher Henshaw Units 3–6 and fish-bearing Lyne Water–Lynslie succession therefore belong to a different part of the inlier and are not transferred into this route’s bed log or advertised as Henshaw Burn fish-finding horizons.

References

British Geological Survey. BGS Lexicon of Named Rock Units: North Esk Group; Reservoir, Deerhope, Cock Rig, Wether Law Linn and Henshaw formations; Grain Heads Siltstone, Lamb Rig Siltstone and Baddinsgill Mudstone members. Records accessed 13 October 2026.

Barron, H. F. (1998). Geology of the central Pentland Hills. British Geological Survey Technical Report WA/98/41, particularly pp. 4–9 and Table 2.

Robertson, G. (1985). Palaeoenvironmental interpretation of the Silurian rocks of the Pentland Hills. PhD thesis, University of Edinburgh, particularly the measured sections, sampling profiles and palaeoecological comparisons in Chapters 2, 4, 5 and 10.

Tipper, J. C. (1973). The marine Silurian of the North Esk Inlier. PhD thesis, University of Edinburgh, particularly pp. 41–44 and Figures 13, 16 and 17.

Donovan, S. K., Fearnhead, F. E., Clarkson, E. N. K. & Donovan, M. (2011). Aspects of crinoid palaeontology of the North Esk Inlier, Scotland (Silurian, Llandovery, Telychian). Palaeontology 54, 241–252.

Anderson, L. I., Clarkson, E. N. K., Stewart, S. E. & Mitchell, D. (2007). An Upper Llandovery Konservat-Lagerstätte in a depositional context: the Pentland Hills Eurypterid Bed, Midlothian. Scottish Journal of Geology 43(1), 41–50.

Molyneux, S. G., Barron, H. F. & Smith, R. A. (2008). Upper Llandovery–Wenlock (Silurian) palynology of the Pentland Hills inliers, Midland Valley of Scotland. Scottish Journal of Geology 44(2), 151–168. Accepted manuscript.

Candela, Y. & Crighton, W. R. B. (2019). Synoptic revision of the Silurian fauna from the Pentland Hills, Scotland described by Lamont (1978). Palaeontologia Electronica 22.2.19A, 1–45.

Harper, D. A. T., Candela, Y. & Gallagher, E. E. (published online 2026; issue dated 2025). The Pentlandian: fact or fiction? Earth and Environmental Science Transactions of the Royal Society of Edinburgh 116, 185–197.

Robertson, G. (1996, Earthwise-hosted edition). North Esk Inlier: River North Esk and Gutterford Burn. In McAdam, A. D. & Clarkson, E. N. K. (eds), Lothian geology: an excursion guide.

Aldridge, R. J., Siveter, David J., Siveter, Derek J., Lane, P. D., Palmer, D. & Woodcock, N. H. (2000). British Silurian Stratigraphy. Geological Conservation Review Series 19. JNCC, Peterborough. Gutterford Burn, pp. 173–176; Lyne Water and Lynslie Burn, pp. 319–323.

Mitchell, G. H. & Mykura, W. (1962). The geology of the neighbourhood of Edinburgh. Third edition, Memoirs of the Geological Survey of Great Britain, Scotland, Sheet 32. Silurian chapter by Mykura and Smith.