Upper Lias, Jurassic, Bristol and Gloucester region
|Green, G W. 1992. British regional geology: Bristol and Gloucester region (Third edition). (London: HMSO for the British Geological Survey.)|
Sedimentation during the lowest two zones of the Upper Lias was on a modest scale throughout the region, but thereafter the pattern was markedly different in the north and south. In the mid-north Cotswolds the next three to four zones are represented by thick arenaceous and argillaceous sequences, whereas the corresponding strata to the south are thin and much condensed. From dispansum Subzone times the situation is almost exactly reversed, with a condensed sequence in the mid-north Cotswolds and a thick arenaceous and argillaceous sequence to the south. The condensed sequences are typically represented by fine-grained, pale-coloured and often ironshot ‘cephalopod limestones’, in which ammonites may be locally very abundant. These are represented by the Cephalopod Bed in the north and the Junction Bed in the south. Superimposed on this sedimentation pattern, the Upper Lias has suffered a more local attenuation in the areas of the Moreton-in-Marsh and Mendip ‘axes’ due to a combination of sedimentary thinning and strong erosion (P948980). The thickest Upper Lias within the region is present in the Cotswold Basin, whose axis stretches between Bredon Hill and Stowell Park with thicknesses of 110 m and 102 m respectively.
The Junction Bed consists of thinly bedded grey limestone with ferruginous, oolitic and conglomeratic beds, yielding Harpoceras, Hildoceras, Dactylioceras and other mid-Toarcian ammonites. The Junction Bed is known in great detail from the Ilminster area in Somerset, where, within a thickness of 5 to 6 m, the abundance of ammonites, including those of remanié origin, has enabled recognition of all the subzones from the upper part of the tenuicostatum Zone to the lowest part of the late Toarcian levesequei Zone inclusive. The lowest 1.5 m include the well-known paper shales of the ‘Saurian and Fish Bed’, with nodules containing fish and reptilian remains, and the ‘Leptaena Bed’, with a remarkable fauna of very small (micromorphic) forms of brachiopods. The tenuicostatum Zone, which here is represented by only of 0.2 m of strata, is unproven and probably absent elsewhere in the present region, apart from the Cotswold Basin. Although it is rarely more than 3 to 5 m thick, the Junction Bed is widely distributed, being found in the main escarpment from Doulting to Chard and also in the outliers of Pennard, Glastonbury and Dundry.
In the eastern Mendips, the Upper Lias is missing, partly as a result of the Bajocian Denudation, but reappears south of Bath in the form of impersistent thin marly and ferruginous ironshot limestones resting on the clays of the Lower Lias and overlain by the Midford Sands (P948981). The thickness does not exceed 1.5 m and there is much evidence of erosion and reworking. The beds continue as far north as Old Sodbury where 1.5 m of cream-coloured marl, limestone and conglomerate of falciferum Zone age are overlain by a sandy pyritous bed with Hildoceras bifrons, at the base of the Cotteswold Sands. From this point northward the Junction Bed passes into the sands and clays at the base of the thick Cotswold Basin succession and ceases to be a recognisable lithological unit.
Upper Lias Sands, Cephalopod Bed and Upper Lias Clay
The lateral passage of the Junction Bed into sand and clay in the south Cotswolds is followed by even more rapid facies changes in the beds above. Before describing these, however, it is necessary to consider the general succession as proved in the escarpment between Old Sodbury and Stroud (see P948979 and P948980). Here, the sandy limestones of the Scissum Beds (Inferior Oolite) rest upon the Cephalopod Bed. The latter attains its maximum thickness of 4.6 m in the Dursley district, where it consists of ferruginous, oolitic limestones and marls with abundant ammonites and belemnites. The stratigraphical significance of the condensed sequence in the Cephalopod Bed can be interpreted from its fauna which includes many ammonites, e.g. Grammoceras striatulum, Phlyseogrammoceras dispansum, Dumortieria moorei and Pleydellia (see P948980) from which it can be dated as representing the thouarsense and levesquei zones of the late Toarcian.
Below the Cephalopod Bed lie some 60 m of fine yellow sand with doggers forming the Cotteswold Sands. These in turn rest upon the Junction Bed or on the Upper Lias Clay.
As the Cephalopod Bed is traced southwards from Stroud and Dursley towards Old Sodbury the basal layers, with Grammoceras thouarsense and G. striatulum (thouarsense Zone), become sandy and then thicken as they pass into the Midford Sands (P948980). Near Bath the higher parts of the Cephalopod Bed are similarly affected, until in the Midford district, south of Bath, the bulk of the sand falls within the subzone of Phlyseogrammoceras dispansum (early levesquei Zone). In general, therefore, the sandy facies is older in the north than in the south. To the north of Stroud, the Cephalopod Bed and Cotteswold Sands are seen to die away or pass into clay. This change of facies first affects the lower part of the sands and then spreads upwards into the higher zones so that at Leckhampton Hill, the Cotteswold Sands have practically disappeared (P948980). Farther north, in the Cleeve Hill area, the whole succession from the top of the Marlstone Rock Bed to the base of the Scissum Beds is represented by Upper Lias Clay. This formation attains a maximum thickness of 110 m in the Bredon Hill outlier.
The basal beds, comprising parts of the tenuicostatum and falciferum zones, are thickest in the axial region of the Cotswold Basin and have long been known from the Dumbleton–Gretton area. Here paper shales up to 7 m in thickness contain micromorphic brachiopods (‘Leptaena Bed’) and laminated limestone nodules (‘Saurian and Fish Bed’) with well-preserved fish and insect remains. The occurrences are similar to those at Ilminster.
Borehole information downdip shows that the easterly change to a predominantly argillaceous succession takes place along a south-south-east-trending line extending from Leckhampton Hill to a few kilometres east of Malmesbury, and possibly as far as Salisbury Plain. West of Bath, the Midford Sands pass into silty mudstones with Dumortieria as shown in the Dundry Hill outlier (P948996, col. 4; P948981).
Owing to the absence of the Upper Lias over the Mendips area due to erosion, these facies changes cannot be traced continuously into Somerset. A recent deep borehole near Devizes has, however, provided information at depth east of the Mendips. This shows a succession of ‘southern aspect’ with 38 m of Midford Sands overlying 30.5 m of clay and finally 6 m of Junction Bed facies. The older borehole at Westbury (Wilts), much quoted in this context (for instance, Kellaway and Welch 1948, p.56), is now thought, on seismic reflection evidence, to pass through the Vale of Pewsey Fault, thereby cutting out much of the Lias succession.
South of the Mendips, the arenaceous facies, known as the Yeovil Sands (P948980), is well developed along the outcrop throughout the district and continues southwards, as the Bridport Sand, to the Dorset Coast. The thickness averages about 60 m and is closely comparable with the Midford Sands. Typically it consists of fine-grained, buff yellow, friable sandstone with layers of calcareous sand ‘burrs’, which contain comminuted shell debris. Well-preserved fossils are rare in the Yeovil Sands, but evidence of the dispansum Subzone of the levesquei Zone is found at Barrington near Ilminster. The main mass probably belongs to the levesquei and moorei subzones of the levesquei Zone. Most of the best-preserved fossils come from the upper part of the succession and include Dumortieria moorei, Trigonia charlockensis and Rhynchonelloidea cynica. The youngest sands occur in the south-west part of the district, west of Yeovil Junction, where thin representatives of the aalensis Subzone and the succeeding opalinum Zone (Middle Jurassic) are usually present. These form part of the Inferior Oolite in the Cotswolds. Northwards from Yeovil, these higher beds appear to have been removed and the Inferior Oolite rests unconformably on the moorei Subzone. In the Yeovil area, the lower part of the formation consists of small interdigitating lenses of impure limestone, blue-grey micaceous silty mudstone and pale grey silt. These pass upwards into strata which contain increasing quantities of silt and fine sand, while the top 15 to 20 m, the only part to merit the name of Yeovil Sands, are of the typical sand facies described above. At Ham Hill, near Yeovil, a large mass of shelly limestone of the type present as small lenses or beds elsewhere is known as the ‘Ham Hill Stone’. This great lens of shelly debris held together by a ferruginous cement is up to 27 m thick and passes laterally into sand with calcareous lenticles. It was extensively quarried at Ham Hill for use as a building stone.
West of the main outcrop, the two outliers of Glastonbury Tor and Brent Knoll provide evidence for the westwards replacement of sand by clay and silt, already noted at Dundry, north of the Mendips. At Glastonbury, underneath 53 m of sands (top not seen), about 11 m of clays supervene above the Junction Bed. Farther west, at Brent Knoll, the sands have passed into silts, and the underlying beds comprise mudstones with pale grey rubbly and nodular limestone beds in the lower part (P948996, col. 2). Offshore to the west, in the Bristol Channel, the entire Upper Lias sequence is apparently argillaceous.
Downdip from the main outcrop to the south-east, and beyond the confines of the region, deep boreholes drilled in the last decade or so have proved a thickening of the Upper Lias into the Wessex Basin similar to that already noted at Devizes. Both the upper sands and the lower clays share in the thickness increase.
Provenance and distribution of the sands
Following Buckman’s (1889) classic demonstration of the southwards younging of the sands in the Upper Lias, which suggests derivation from the north-east, and Boswell’s study (1924) of the heavy mineral content, which supports derivation from the Armorican metamorphic terrain of the Brittany area to the south-west, the mode of sedimentation of the sands has provided a puzzle. Palaeocurrent analysis based on cross-bedding measurements (Davies, 1969) showed a dominant current flow to the south-west and a subordinate flow to the north-east.
In the most detailed explanation so far advanced, Davies (1969) postulated a separation of the processes of sediment supply and sediment deposition. Supply from the Armorican source area was by means of longshore drift to the north adjacent to a Cornubian land mass, lying some distance to the west of the present sand outcrop. Deposition was by means of a sediment return system activated by the dominant internal basinal current flow which was towards the south-west, as shown by the cross-bedding measurements. He postulated that the return flow was stimulated by a marked change in the orientation of the ‘Cornubian’ coastline west of Cheltenham. The sand body itself represents an east-west-trending bar complex that moved southwards from the vicinity of Cheltenham to Bridport by means of accretion of sediment on its southern side. Forebar as well as backbar sands are claimed to be recognised. The Ham Hill Stone which, unlike the main sand mass, shows cross-bedding indicating a strong counterflow direction to the north-east during moorei subzonal (late levesquei) times is said to represent the deposits of a tidal channel cut through the bar.
This model has been criticised on the grounds that, apart from the Ham Hill Stone, the supposed bar deposits do not show large-scale cross-bedding and that there appears to be no significant difference between the fore- and backbar deposits. In addition, the only surviving deposits to the west are silts and clays (P948981). A more recent suggestion, that the sands entered the region from the north-east via the western margin of the London Platform, encounters a similar difficulty in that the central Cotswold clay basin apparently intervenes (P948981).
- Kellaway, G A, and Welch, F B A. 1948. British regional geology: Bristol and Gloucester district (2nd edition). (London: HMSO for Institute of Geological Sciences.)
- Buckman, S S. 1889. On the Cotteswold, Midford and Yeovil Sands, and the division between the Lias and the Oolite. Quarterly Journal of the Geological Society of London, Vol. 59, 445–458.
- Davies, D K. 1969. Shelf sedimentation: an example from the Jurassic of Britain. Journal of Sedimentary Petrology. Vol. 39, 1344–1370.