Chapter 3 Southern Province, England

Introduction

Upper Cretaceous rocks in the Southern Province occur as small outliers in south-west England and form the rolling Chalk downland and cliffs of Wessex and the Weald regions (Figure 3.1). The lowest Upper Cretaceous deposits in south-west England are shallow-water, richly fossiliferous glauconitic greensands, and highly condensed, fossiliferous limestones. In contrast, to the east, these pass into deeper water chalks. The chalks thicken eastwards into Sussex, and continue across the Channel as far as the eastern side of the Paris Basin. The whole of this region is known as the Anglo-Paris Basin'. The Chalk successions in the southern English part of this basin belong to the Southern Province ((Figure 1.16), Chapter 1).

Movement of deep-seated faults has uplifted and folded the younger cover rocks, including the Chalk. Folding results in the Chalk dipping north in the North Downs into the London Basin and south in the South Downs into the Channel. There is a general structural plunge to the west, taking the Chalk under the Hampshire Basin. Chalk is brought to the surface again by a number of smaller, fault-controlled anticlines on the Isle of Wight and along the south Dorset coast. One of these folds brings high Chalk to the surface on Portsdown, where unusual sedimentary structures are exposed at Downend Chalk Pit. Another of these folds is the pronounced topographical feature corresponding to the Dean Hill Anticline, south-east of Salisbury, with Pepperbox Hill Quarry towards its western end (Figure 3.1). The general plunge westwards of the folds into Wessex preserves progressively younger chalks in that direction. The youngest chalks are exposed in the Isle of Wight and in Studland Bay, Dorset, where they extend into the Upper Campanian. In contrast, the most complete earliest Upper Cretaceous deposits are found in the east at Lewes, Beachy Head and Folkestone.

Along the south coast, through the Isle of Wight and Handfast Point to Ballard Point, to White Nothe in Dorset, the Chalk forms a long, narrow rib of downland, dipping steeply north (Figure 3.2). The cliff sections along this rib are spectacular, producing the Needles and Old Harry Rocks, and these sections are crucial to understanding how the basin evolved during the Late Cretaceous Epoch.

In central Dorset there are open expanses of downland in younger Campanian Chalk, which here forms several small but conspicuous escarpments related to the lithostratigraphy. These escarpments are readily identifiable on satellite imagery and can be traced through Hampshire into Sussex (Bristow et al., 1997). Interpreting the stratigraphy of these mapped topographical features, geophysical borehole logs and general geology, relies on a number of very small chalk pits, and extrapolation to the better coastal sections or quarries to the south and east (i.e. the GCR sites).

Two major fault-controlled re-entrants into the western Chalk escarpment in Wiltshire, the Vale of Wardour and Vale of Pewsey, bound Salisbury Plain to the south and north respectively. There are few exposures of Chalk in this area and Shillingstone Quarry in north Dorset, on the south side of the Vale of Wardour, and Beggars Knoll, Westbury, on the south side of the Vale of Pewsey provide critical sections in the Cenomanian, Turonian and Coniacian stages. A number of small pits around Salisbury expose the higher Chalk, including an excellent section in the Lower Campanian Newhaven Chalk Formation at West Harnham Chalk Pit. On the north side of the Vale of Pewsey, controlled by the Mere Fault, is found the highly condensed Turonian–Coniacian succession at Charnage Down Chalk Pit.

In the south-west region, chalks and Chalk-equivalent 'greensands' and limestones are found as small erosional remnants on the dissected Upper Greensand plateau that extends westwards from Lyme Regis in Dorset and Chard in Somerset to Salcombe and Wilmington in east Devon (see (Figure 3.19), p. 109). Farther west, capping the Haldon Hills in south Devon, the Upper Greensand is locally overlain by Cenomanian 'greensands' (the Cullum Sands-with-Cherts (Hamblin and Wood, 1976)) and by Palaeocene and Eocene flint-rich gravels (Tower Wood, Buller's Hill and Aller gravels). The characteristically grey flints in these gravels are unusually fossiliferous and contain well-preserved moulds of Turonian to Campanian fossils, notably echinoderms, brachiopods and inoceramid bivalves (Selwood et al., 1984) that indicate the former presence of parts of the Chalk succession that are no longer preserved. Similar faunas are preserved in flint-rich gravels around the edges of the tectonically down-faulted Bovey Basin, and at Orleigh Court, North Devon, on the extension of the Bovey structure (Figure 3.1).

Preservation of the higher part of the Upper Cretaceous succession is largely dictated by the Laramide phase of tectonism at the end of the Cretaceous Period and in the early part of the Palaeogene Period. Erosion of the Chalk beneath the Palaeogene deposits has cut more deeply in the North Downs, typically to levels high in the Micraster coranguinum Zone (Seaford Chalk Formation), but elsewhere only down as far as the top of the Marsupites testudinarius Zone (e.g. M25 Leatherhead–Reigate section and the British Geological Survey Fetcham Mill Borehole near Leatherhead (Gray, 1965; Murray, 1986)). In east Kent, the basal beds of the Offaster pilula Zone are found in Thanet (Shephard-Thorn, 1994) but they are not preserved in the main mass of the Downs. In the South Downs, the preservation level beneath the Palaeogene surface lies in the basal Gonioteuthis quadrata Zone in East Sussex, whereas it lies in the Belemnitella mucronata Zone in West Sussex, Hampshire and central Dorset.

Tectonic structure and sedimentation history

The structural line along the Vale of Pewsey forms a natural boundary with the Marlborough Downs–Berkshire Downs–Chiltern Hills region to the north. The significance of this structural boundary is demonstrated by the difficulty of correlating successions across it. To the south and south-west are a series of structural lines, the South Downs Axis (Allen, 1975, 1981; Young and Monkhouse, 1980), the Mid-Dorset Swell (Drummond, 1970) along the Fordingbridge–Cranbourne Fault block, the St Valery–Bembridge line and the Cotentin line of Smith and Curry (1975). The Cotentin Line was considered by Curry and Smith (1975) to form a boundary between a Western Approaches Basin and basins to the east in the Channel. Other workers have identified sub-basins, such as the 'Channel' Basin (Chadwick, 1986), but these do not readily correspond to the sedimentary pattern during the Late Cretaceous Epoch. In the Southern Province the pattern of Chalk sedimentation closely relates to the fault-controlled fold lineaments (e.g. the Cranbourne–Fordingbridge fault block), and even to the subtleties of individual folds (Mortimore, 1986b; Mortimore and Pomerol, 1987, 1991a, 1997).

The Late Cretaceous sedimentation history reflects both the broad tectonic setting of shelves and basins, and the local effects of growth tectonics on folds and faults. The broad tectonic structure of the Southern Province is subdivided (e.g. Drummond, 1970) into a 'Wessex Shelf', including south-east Devon, Somerset and west Dorset; and a 'Wessex Basin', which continues eastwards into the Weald Basin, forming one large basin system. The old concept of a 'Cenomanian Transgression' (Suess, 1883–1888), as applied to this region, actually refers to the progressive overstep of the Albian Upper Greensand onto pre-Cretaceous rocks westwards in Dorset and Devon. The Greensand rests on the Middle Jurassic strata at Bridport, the Lower Jurassic succession at Lyme Regis, the Triassic System in east Devon, and on Devonian to Permian rocks in south-west Devon (Figure 3.1).

The most westerly Cenomanian deposits in Dorset, between Eggardon and Litton Cheyney, expose about 30 m of typical Grey Chalk Subgroup lithologies (Wilson et al., 1958), whereas their nearest correlative in east Devon consists of less than 1 m of highly condensed nodular limestone. The Upper Cretaceous deposits in the south-west region are themselves cut by north-south-trending faults (Kellaway and Welch, 1948) that were active in Cenomanian and early Turonian times, resulting in rapid lateral lithological variations in the east Devon successions. This faulted shelf contains a richness and diversity of echinoderm and crustacean faunas (especially crabs) that is unmatched at this stratigraphical level elsewhere in the Southern Province. By the Turonian, the whole of south-west England was probably inundated by the sea, and chalks were deposited throughout the region and beyond.

In contrast to the shelf, the main basin was a complex submarine trough, with its apex in the west in Wiltshire and north Dorset, opening eastwards across the Anglo-Paris Basin (Drummond, 1970; Mortimore, 1979, 1983; Mortimore and Pomerol, 1987). This contrasts with the earlier Mesozoic palaeogeographical model comprising a 'Wessex Trough' and a local 'Wealden' Basin (Whitaker, 1985; Chadwick, 1985, 1986). Within this submarine trough, a relationship between the deep faulting, surface folding and Upper Cretaceous sedimentation has been suggested (Mortimore, 1986b; Mortimore and Pomerol, 1987, 1991b, 1997), including changes in thickness of Chalk strata, and the occurrence of channel scours at particular places and stratigraphical intervals (e.g. the Southerham Grey Pit Channel and Strahan's Hardground in the Southerham Pit). In addition, the Newhaven Chalk Formation contains a particular fracture style, including faults that are confined to that level and do not continue upwards into the overlying Culver Chalk Formation (e.g. at Castle Hill, part of the Newhaven to Brighton GCR site).

The evidence for intra-Chalk tectonic movements, channel scour and slumping events is further supported by the large-scale growth structures and slump folding at Downend Chalk Pit, Portsdown, in the Early Campanian and by the slumping seen on Solent marine seismic lines in Late Turonian–Early Campanian times (Mortimore and Pomerol, 1991a, 1997). Evans and Hopson (2000), have provided the first onshore seismic sections showing the scale of channels in the Chalk in Dorset and Hampshire. Big coastal exposures such as Whitecliff, Isle of Wight, provide evidence for episodic movements closely associated with well-known tectonic lines over which condensed or anomalous sediments recur. Similar evidence is found at White Nothe, Dorset. As a result of deep intra-Chalk channel scour in particular, the stratigraphy preserved at any one locality can be very variable. The presence of large-scale channels helps explain many stratigraphical anomalies in the province.

Stratigraphy

The stratigraphy of the Southern Province contains two distinct developments, one relating to the Chalk Group of the main basin (Figure 3.3) and the other to the condensed Cenomanian deposits of the south-west marginal shelf of Devon, Somerset and west Dorset (Figure 3.4).

Lithostratigraphy of the Chalk in the Southern Province

Evolution of the lithostratigraphy in the Chalk Group towards the present subdivision into two subgroups and nine mappable formations in the Southern Province (Figure 3.3) has been long and tortuous (Bristow et al., 1997; Rawson et al., 2001).

Many 19th century observers (e.g. Evans, 1870) had recognized that marker beds existed that could be used for local correlation and for recording the distribution of fossils. Whitaker (1872) described a 'Three Inch Flint Band' that he could trace around the coast of Kent. Similarly Bedwell (1874), in addition to Whitaker's flint band, identified two marker flint beds on the Thanet Coast. Rowe (1900) coined the names 'Whitaker's 3-inch tabular flint-band', 'Bedwell's columnar band' and the 'Bedwell-line' for these marker beds, but he did not identify them outside the Kent coast sections. Price (1874, 1877) introduced identified horizons such as the 'Cast Bed' in the Lower Chalk, a key marker horizon on top of the Tenuis Limestone (see Folkestone to Kingsdown GCR site report, this volume).

Work on a detailed modern lithostratigraphy for the Southern Province really began with the subdivision of the Plenus Marls into eight beds by Jefferies (1962, 1963), who nominated the Merstham Quarry, Surrey, as the stratotype section. This section no longer exists and the Holywell, Eastbourne section, figured by Jefferies (1963) is taken as the replacement stratotype section because of its completeness. A study of the detailed lithostratigraphy for the White Chalk Subgroup (Middle and Upper Chalk) of the province was undertaken in the 1970s (Mortimore, 1977, 1983, 1986b). Sections were nominated as stratotypes for the lithology, boundary markers were identified, and geographical names applied to members and beds in the areas where the succession was the most complete. The lithostratigraphical concepts continued to evolve through the 1980s, when a second, parallel nomenclature was introduced for the North Downs (Robinson, 1986). Modifications to the existing schemes were required as the correlation of units was tested farther afield (Mortimore and Pomerol, 1987, 1996). It was only when geological mapping was undertaken by the British Geological Survey, beginning in central Dorset (Bristow et al., 1995) and in West Sussex, that the lithostratigraphy could be related over a wider area to mapping units and the nomenclature rationalized (Bristow et al., 1997).

Grey Chalk Subgroup

The Grey Chalk Subgroup includes the traditional Lower Chalk up to the base of the Plenus Marls Member and the Cenomanian Limestone (Bed A and Bed B only) in south-east Devon. For the main Wessex–Weald basin the subgroup is divided into two mapping units, a lower West Melbury Marly Chalk Formation and an upper Zig Zag Chalk Formation ((Figure 2.8), Chapter 2). In south-east Devon these two mapping units are condensed into a few metres of sandy limestones (Cenomanian Limestone Beds A to C of older literature), the Beer Head Limestone Formation of Jarvis and Woodroof (1984). The topmost division of the Beer Head Limestone, the Pinnacles Member (Cenomanian Limestone Bed C of older literature), correlates with the Plenus Marls–Sussex Melbourn Rock of the main basin. To conform with the main basin, the Pinnacles Member is, therefore, separated from the Beer Head Limestone and included in the lowest formation of the White Chalk Subgroup above.

Throughout most of the Southern Province, the base of the Chalk, and the Upper Cretaceous Series, is taken at the base of the Glaucontic Marl ((Figure 2.2)b, Chapter 2). However, the hiatus at the base between the Glauconitic Marl and underlying Lower Cretaceous greensands increases south-westwards from Eastbourne through the Isle of Wight and along the Dorset coast to White Nothe. Farther west (Hooken Cliff), the hiatus is marked by a bioclastic calcarenite (Beer Head Limestone) resting on the Small Cove Hardground at the top of the Upper Greensand.

The Glauconitic Marl is defined as a Member at the base of the West Melbury Marly Chalk Formation (a unit that encompasses much of the traditional Chalk Marl). The upper boundary of the West Melbury Maly Chalk is taken along a mapping feature defined partly by a conspicuous limestone ('Tennis Limestone') and partly by the overlying markedly silty, clayey bed, the 'Cast Bed' that balls up in the plough in fields. Marl–limestone rhythms are a feature of the West Melbury Marly Chalk (Figure 3.5), but the number of rhythms varies, with the most complete successions at Folkestone (Folkestone to Kingsdown GCR Site) and Southerham Grey Pit and reduced successions progressively westwards towards White Nothe. In central Dorset, around Shaftesbury, the rhythmic units are replaced by a broad unit of silty and marly chalk. At Shillingstone Quarry the manly chalk passes south-westwards into conglomerates and calcareous sands on the Mid-Dorset Swell. As this unit is traced into the Chiltern Hills it is truncated by the basal erosion surface of the Totternhoe Stone.

The top unit of the Grey Chalk Subgroup, the Zig Zag Chalk Formation (broadly the former Grey Chalk), can be traced over great distances and marks a change to more calcareous sediments with less marl ((Figure 2.2)a, Chapter 2). There are several distinct lithological units within the Zig Zag Chalk, including the Jukes-Browne Bed 7 with its laminated structures (see Folkestone to Kingsdown GCR site report, this volume), which is a consistent feature throughout the North and South Downs. The overlying Bed 8 or 'White Bed' is consistent in the North Downs, but shows considerable lateral variation in the South Downs, from a rhythmically bedded unit at Beachy Head and Southerham Grey Pit, to a homogeneous pale green coloured unit in parts of West Sussex.

White Chalk Subgroup

The White Chalk Subgroup encompasses all the traditional Middle and Upper Chalk and the Plenus Marls (Figure 3.6). A major erosion surface beneath the Plenus Marls Member marks a change in sedimentation across Europe reflected in a marked colour change from the White Bed beneath to the green hues of the Plenus Marls and overlying nodular beds of the Holywell Nodular Chalk Formation. This sub-Plenus erosion surface is, therefore, taken as the base of the White Chalk Subgroup and the base of the Holywell Nodular Chalk Formation, the lowest mapping unit of the subgroup. The White Chalk Subgroup is marked by the presence of purer chalks and the general entry of flint in formations above the Holywell Nodular Chalk. Seven mapping formations and two mapping members are recognized (Figure 3.3). Numerous marker beds are also recognized (Figure 2.8), (Figure 2.9), (Figure 2.21), (Figure 2.22) and (Figure 2.27), Chapter 2).

Holywell Nodular Chalk Formation and New Pit Chalk Formation

The basal unit of the White Chalk Subgroup and the Holywell Nodular Chalk Formation, the Plenus Marls Member, is recognized everywhere except in south-east Devon. The thickness of this unit also varies greatly, being thickest and most complete at Eastbourne (Beachy Head). Above the Plenus Marls, the Holywell Nodular Chalk Formation is gritty and nodular in contrast to the smoother chalks of the New Pit Chalk Formation, which contains numerous marker marl seams (Mortimore and Pomerol, 1996; Bristow et al., 1997). These two formations change thickness dramatically across the Southern Province (Mortimore, 1986a; Mortimore and Pomerol, 1987, 1996). The Holywell Nodular Chalk thins onto the southern edge of the Anglo-Brabant Platform at Dover and through the North Downs, and thickens along the south-west margin in Dorset and Devon. In contrast, the overlying New Pit Chalk is thicker in the North Downs and almost completely disappears along the south-west margin in Dorset (Mupe Bay) and in parts of south-east Devon.

Marker Beds in the Holywell Nodular and New Pit Chalk formations

Numerous lithostratigraphical marker beds are recognized in the Holywell Nodular Chalk and New Pit Chalk formations (Mortimore, 1983, 1986a, 1990; Mortimore and Pomerol, 1990, 1991b, 1996). These include:

1. beds numbered 1 to 8 in the Plenus Marls Member Oefferies, 1963);
2. the six Meads Marls that span the Cenomanian–Turonian boundary, and;
3. the Holywell Marl and Gun Gardens Marl within the Holywell Nodular Chalk Formation.

The Gun Gardens Main Marl is the boundary between the Holywell Nodular Chalk and New Pit Chalk formations, and has a very distinctive finger-flint horizon above, the Glyndebourne Flints. This marl and overlying flint band has been recognized in Kent at the Hailing Pit; at Southerham Pit, Lewes; on the Dorset coast; and at Beggars Knoll Quarry, Westbury, Wiltshire. Lithological marker beds in the New Pit Chalk include the Mailing Street Marls, the New Pit Marls and the Glynde Marls. These marls have been recognized throughout the main basin axis, in the shelf successions of south-east Devon and have been correlated north through the London Basin and parts of the adjoining Transitional Province (Mortimore, 1986a,b, 1987; Mortimore and Wood, 1986; Mortimore and Pomerol, 1987, 1996). These marl seams, and those in the lower Lewes Nodular Chalk above, are critical markers in determining the amount of condensation and erosion involved in the Chalk Rock.

Gale (1996) studied the cyclostratigraphy of the Turonian Chalk, using the newly introduced Lulworth Marl and Robinson's (1986) Round Down Marl as key markers. These are the correlatives of the Gun Gardens Main Marl and Mailing Street Marl 2, respectively. Hence these extra names are unnecessary.

The level of entry of persistent flint bands in the Southern Province varies. In West Sussex (Duncton Quarry) through Hampshire (M3 Cuttings, Twyford Down), and in Wiltshire (Beggars Knoll Quarry), flint bands are present in the New Pit Chalk. Flints are progressively more common in south Dorset and southeast Devon, even occurring in the basal Holywell Nodular Chalk of Worbarrow Bay. In contrast, in Kent and East Sussex, flint bands generally enter in the Glynde Beds at the base of the Lewes Nodular Chalk Formation. This pattern of flint distribution probably reflects the basin geometry, with flint entering earlier along the west and south-west margin and on the north-east shelf (Mortimore and Pomerol, 1987).

Lewes Nodular Chalk Formation

The base of the Lewes Nodular Chalk Formation (Mortimore, 1983, 1986a) is defined at the entry of persistent nodular chalk beds, which occurs above the Glynde Marls (Bristow et al., 1997) throughout most of the province including south-east Devon. Caburn Pit, Lewes, and Beachy Head, Eastbourne, are the basal boundary stratotype sections. In the western part of the Southern Province the base of the formation is taken at the base of the 'Spurious Chalk Rock' of Rowe (1908), which has been correlated with the Ogbourne Hardground at the base of the standard Chalk Rock stratigraphy (Bromley and Gale, 1982), as seen at Charnage Down Chalk Pit, Mere, Wiltshire. The Spurious Chalk Rock is well developed along the southern margin of the basin at Compton Bay, Isle of Wight, Ballard Head (Handfast Point to Ballard Point GCR site), Mupe Bay, and White Nothe, Dorset. Exactly how the Spurious Chalk Rock relates to the basal Lewes Nodular Chalk of the expanded sections in Sussex and the North Downs is still under discussion.

The Lewes Nodular Chalk is 80 m thick in East Sussex and may be only 20–30 m thick at Shillingstone Quarry and Charnage Down Chalk Pit, Mere. Mortimore and Pomerol (1996), divided the Lewes Nodular Chalk into lower and upper parts at the Lewes Marl, a key basin-wide and inter-basinal marker bed and a vulcanogenic marl. This marl is associated with the conspicuous Lewes Tubular Flints ((Figure 3.9)b, p. 96), which are also present even in the condensed sections where the Lewes Marl may be locally occluded (e.g. Dover, Kent, (Figure 3.7)) or very thin (Chapel Rock, south-east Devon). Within the lower Lewes Nodular Chalk are a number of conspicuous marker marl seams named after localities around Lewes, Sussex. These include, in ascending order, the Southerham, Caburn, Bridgewick and Lewes marls, some of which are derived from decomposed volcanic ash (tuffs, Wray, 1999). Each of these marls is lost in the condensed hardgrounds comprising the Chalk Rock at Charnage Down Chalk Pit, Mere and Cley Hill, Warminster, Wiltshire (Bromley and Gale, 1982; Mortimore, 1983). The lower Lewes Nodular Chalk, 40 m thick around Lewes, East Sussex, probably represents the expanded stratigraphy of the Chalk Rock, which is only 1 m thick at Cley Hill.

Also in the lower Lewes Nodular Chalk is a unit known as the 'Basal Complex', which is particularly clearly seen in the Dover coast sections ((Figure 3.8); Jukes-Browne and Hill, 1904; Mortimore and Wood, 1986). This unit comprises a succession of conspicuous flint bands and marl seams named the Bridgewick Flints, Bridgewick Marls and Bopeep Flints, in ascending order (Mortimore, 1986a, 1997). The Basal Complex is traceable through the North Downs, where it was formerly used by the British Geological Survey to map the base to the traditional Upper Chalk. It is a flint maximum in the Upper Turonian of the Transitional Provinces (Mortimore and Wood, 1986), where it corresponds to the horizon of the Brandon Flints at Grimes Graves Flint Mines, Brandon, Norfolk. This same horizon is taken as the mapping base of the Burnham Chalk Formation (Wood and Smith, 1978) in the Northern Province. This horizon can also be identified in south-east Devon (at Hooken Cliff; (Figure 3.9), (Figure 3.10), (Figure 3.11).

The upper Lewes Nodular Chalk continues the character of the lower Lewes Nodular Chalk, comprising a number of cyclic packages of hardgrounds and nodular, gritty chalks. One of the best known marker beds is the traditional Top Rock, which equates with the Navigation Hardground (Mortimore, 1983, 1986a; Bailey et al., 1983, 1984). Elsewhere in the Southern Province, the Top Rock may represent either the Cliffe or Hope Gap hardgrounds (e.g. Charnage Down Chalk Pit), while in the Transitional Province (e.g. Aston Rowant Cutting and Kensworth Chalk Pit in the Chiltern Hills) it may be an amalgam of the Navigation, Cliffe, Hope Gap hardgrounds and even higher hardgrounds. Below the Navigation Hardground, the Cuilfail Zoophycos (Flints), a unit of chalk with dark, colour-contrasting, millimetre-thin marly streaks representing the trace fossil Zoophycos ((Figure 3.10)a), is a conspicuous, widely correlatable, marker horizon present from Hooken Cliff, south-east Devon, to Dover. At both Lewes and Hooken Cliff ((Figure 3.9)a) these trace fossils are spectacularly silicified.

Near the top of the Lewes Nodular Chalk there is a second, equally distinctive, unit of Zoophycos chalk, the Beachy Head Zoophycos Beds. This marker, which is seen at Seaford Head (Cuckmere to Seaford GCR site) and is particularly well developed in the Beachy Head section, can be traced northwards from Dover (Folkestone to Kingsdown GCR site) through the London Basin, where the unit is conspicuous in borehole cores (Mortimore et al., 1990), and westwards into flaser or griotte marly chalk in south-east Devon (e.g. Chapel Rock and, less accessibly, at Hooken Cliff). The Beachy Head Zoophycos Beds are overlain by the two Shoreham Marls, between which the Shoreham Tubular Flints (Figure 3.10) are conspicuous markers in both borehole core and exposures (e.g. Seaford Head (Cuckmere to Seaford Head GCR site) and Charnage Down Chalk Pit, Mere), as well as forming a distinctive field brash for mapping. Over structural highs, either or both of the Shoreham Marls may be occluded in the development of strong hardgrounds, for example the Bar End Hardgrounds of Winchester (Mortimore, 1986a) and the Rochester Hardground (Robinson, 1986) of Kent and Surrey.

Seaford Chalk Formation

Coarse, gritty Lewes Nodular Chalk changes abruptly to the smooth pure white, homogeneous Seaford Chalk Formation, which produces a quite distinct, slabby-chalk field brash. The basal boundary marker is the Shoreham Marl 2 at Seaford Head, Sussex (Cuckmere to Seaford GCR site) which has been shown to be a vulcanogenic marl (Wray, 1999). Within the Seaford Chalk there are again numerous lithological marker beds that have been widely correlated (Mortimore, 1986a, 1997; Mortimore and Pomerol, 1987). Of these, the most conspicuous are the big flint bands, the Seven Sisters, Michel Dean, Baily's Hill, Flat Hill/Bedwell's Columnar and Whitaker's 3-inch flint bands. There are many more flint bands along the southern margin of the Basin in Sussex, Hampshire and Dorset and Devon, compared to the northern margin in Kent. As the flints are traced westwards through the Isle of Wight and south Dorset coast sections many of the individual flints in the marker bands also increase in size.

Newhaven Chalk Formation

Regular marl seams in white chalk with flint bands characterize the Newhaven Chalk and the basal boundary marker is taken at the lowest of these, Buckle Marl 1 at Seaford Head (Cuckmere to Seaford GCR site). Individual marl seams have been traced through the main axis of the basin to Salisbury at Pepperbox Quarry, East Grimstead Quarry and West Harnham Chalk Pit (Mortimore, 1986a). Marl seams and many flint bands disappear over local tectonically controlled highs such as the Hollingbury Dome near Brighton, Shawford near Winchester (Mortimore and Pomerol, 1997) and on the Dean Hill Anticline at Dean Hill near Salisbury (see (Figure 3.1) and (Figure 3.43), p 157). The marl seams strengthen and thicken into more basinal areas. At Whitecliff, Isle of Wight, there is a special section of condensed Newhaven Chalk with a number of hardgrounds and associated phosphatic chalks. The key marker beds in the Newhaven Chalk Formation are identified and defined in the Newhaven to Brighton and Cuckmere to Seaford GCR sites, Sussex.

The Culver Chalk Formation

The Culver Chalk Formation can locally be subdivided into the Tarrant Member and the Spetisbury Member (Bristow et al., 1995). These two members comprise white chalks with numerous flint bands, and correspond to the Sompting Beds and Whitecliff Beds of Mortimore (1983, 1986a). They were introduced in central Dorset for the mappable topographical features present in that area.

Subsequently the topographical feature representing the Tarrant–Spetisbury boundary was traced through West Sussex to Warningcamp Chalk Pit near Arundel, and was found to correlate with the Warningcamp–Whitecliff Flint of Mortimore (1986a,b). This suggests that the same basal boundary marker chosen for the base of the Whitecliff Beds can be used for the base of the Spetisbury Chalk Member, at least in the South Downs into Hampshire.

The base of the Tarrant Chalk is not so easily defined, as it appears to be at different levels in the South Downs compared to Dorset. In West Sussex, the basal Tarrant Chalk topographical feature has been traced into the Black Rabbit Pit at Arundel, where it corresponds to the entry of the large Castle Hill Flints 4 and 5 of Mortimore (1986a). This is just above the Pepperbox Marls, at the boundary between Mortimore's Newhaven Chalk and Culver Chalk formations. Biostratigraphical evidence from Dorset suggests that the base of the Tarrant topographical feature is lower there, around the Arundel Sponge Bed or even at the level of the Meeching Marls (Bristow et al., 1997). A key problem is finding an accessible, continuous section through the entire Tarrant–Spetisbury interval. Such sections are only to be found on the Isle of Wight at Whitecliff and Scratchell's Bay, in hard, steeply dipping chalks, where the topographic significance of the lithologies is less easily determined (but see (Figure 3.12)). Using Whitecliff as a standard, the one key lithological feature that can be used to characterize the Spetisbury Chalk is the presence of many bands of paramoudra flints. Paramoudra flints also occur in the Tarrant Member, but not as regularly. Marker beds in the Culver Chalk are identified in the Whitecliff, Isle of Wight, sections (Figure 3.13), (Figure 3.14), (Figure 3.15).

Portsdown Chalk Formation

The base of the Portsdown Chalk Formation is marked by the reappearance of conspicuous marl seams, following a unit (Culver Chalk) in which they are virtually absent. This change is a conspicuous feature in the high chalk on Portsdown and along the coast between Whitecliff, Isle of Wight, and Bats Head on the Dorset coast. The basal boundary marker is the Portsdown Marl 1 at Farlington, Portsdown and at Whitecliff. In central Dorset, the marl seams characteristic of this unit are largely absent, except for the Almer Marl, but the chalk still produces a very fine topographical feature that is mappable into Hampshire and West Sussex (Bristow et al., 1997).

Above the Portsdown Chalk with many marl seams, up to the Palaeogene surface on the Isle of Wight and at Studland, is another unit without marls informally named the Alum Bay Beds' (Mortimore, 1979, 1983) but later formally designated the 'Studland Chalk Member' (Gale et al., 1988).

South-west England (south-east Devon, Somerset and west Dorset)

The lowest parts of the Upper Cretaceous deposits of south-west England are so different from those of the main basins to the east that they require separate treatment (Figure 3.4). In particular, most of the Cenomanian deposits consist not of chalks, but instead comprise highly condensed, fossiliferous sands and sandy limestones. These form the Beer Head Limestone Formation of the Grey Chalk Subgroup.

The Holywell Nodular Chalk, New Pit Chalk, Lewes Nodular Chalk, and Seaford Chalk formations of the White Chalk Subgroup can be recognized, albeit the New Pit Chalk Formation is conspicuously flinty (the Beer Roads Member of Jarvis and Woodruff 1984) compared with its development elsewhere in the Southern Province.

The top Pinnacles Member (Bed C) of the Beer Head Limestone is included in the White Chalk Subgroup as the basal member of the Holywell Nodular Chalk Formation. The preserved and exposed succession in south-east Devon goes up only to the basal Seaford Chalk Formation, the highest beds being present in Hooken Cliff (Figure 3.16). The lower part of the White Chalk Subgroup in east Devon shows considerable lateral variation over short distances, and contains numerous minor sedimentary breaks represented by hardground surfaces. It consists of gritty and nodular chalks that can be broadly correlated on lithological and faunal grounds with the Holywell Nodular Chalk of the standard Southern Province succession. These beds include lithologies not seen elsewhere in southern Britain, notably the calcarenitic Beer Stone made up of inoceramid debris. The top of the Holywell Nodular Chalk is marked in Hooken Cliff and in Beer Cliffs by a laterally persistent hardground (the Branscombe Hardground of Jarvis and Woothoof, 1984), which represents a significant sedimentary break.

Biostratigraphy and chronostratigraphy

The current biostratigraphical and chronostratigraphical division of the Late Cretaceous deposits of the Southern Province is a mixture of traditional assemblage zones from the Anglo-Paris Basin and internationally agreed zonal and subzonal divisions of the Late Cretaceous stages ((Figure 1.5), Chapter 1; (Figure 2.8), (Figure 2.9), (Figure 2.21), (Figure 2.22) and (Figure 2.27), Chapter 2). The boundary stratotypes for the stages are mostly based outside the UK (see Appendix, this volume).

The biostratigraphy has largely evolved from the work of many great amateur palaeontologists working in the 19th and early 20th centuries. A Lewes surgeon, Mantell, published his studies in The Fossils of the South Downs or Illustrations of the Geology of Sussex (1822). This work can be compared with the description at the same time of the Geology around Paris (which actually included a huge area of the Paris Basin), by Cuvier and Brongniart (1822). These wonderful books were the first to illustrate the extraordinary range of fossils in the rocks of the region and they must have stimulated others to investigate further. Like Mantel's publications, Dixon's Geology of Sussex (1850) is a source of many type and figured specimens, for example the Micraster corbovis of plana Zone type (Forbes, Stokes, 1975) (the Forbesaster forbesi of Drummond, in manuscript; Drummond, 1983).

The information accumulated by Mantell, Dixon, and others contributed to d'Orbigny's (1847, 1850–1852) worldwide synthesis in lists of fossils contained within Upper Cretaceous stages. This work was further developed by Hebert (1866, 1874) and Barrois (1876), who both deserve a special mention for their application to England of the familiar Chalk zones from the Chalk of the Paris Basin ((Figure 2.8), Chapter 2). These are the zones that were applied by Rowe (1900–1908) to the coast sections and later mapped by both Brydone (1912) and Gaster (1924–1951) in Hampshire and Sussex respectively.

Rowe concentrated on the well-exposed coastal sections (Rowe, 1900–1908), whereas the [British] Geological Survey paid more attention to the inland sections. This led to wide discrepancies in zonal concepts because of the different nature of exposures and preservation of fossils between the coast and inland pits, and the absence of a good lithostratigraphical framework. Rowe (1899) also published the first modern study of a fossil group, the Chalk heart-shaped urchin Micraster. This paper has had a great influence on evolutionary thought, and is still widely used for teaching. In his papers on the coastal sections (but not in his posthumous paper on the Chalk of Lincolnshire, Rowe, 1929), Rowe did not accept the principle of lithostratigraphical correlation in the Chalk. Nevertheless his 'zoological boundaries' invariably conveniently coincided with lithological marker horizons at any one locality, which, despite his stated philosophy, led him to make serious mistakes in correlation.

Brydone (in Griffith and Brydone, 1911; Brydone, 1912, 1914, 1915) not only mapped the Chalk zones in Hampshire but also provided an immense amount of information on the numerous chalk pits in that county Brydone's map illustrated the extent of folding in the Chalk, including the strongly developed Warnford Dome and numerous subsidiary folds in south Hampshire from Winchester to the Sussex border. Brydone was exceptional in providing some measured sections and details of marl seams in the Marsupites testudinarius and Offaster pilula zones on the Sussex coast and in Hampshire. He recognized the two-inch marl that could be traced from Seaford Head to Rottingdean on the Sussex coast, now named the 'Old Nore Marl' (Mortimore, 1986a).

Apart from the vast collection of fossils from Hampshire, now mostly held by the British Geological Survey, Brydone's great contribution to the biostratigraphy of the Chalk of the Southern Province was the description of many of the stratigraphically distinctive shape changes in the echinoid genus Echinocorys ((Figure 2.3), Chapter 2). Brydone (1912, 1939), following previous suggestions by Rowe (1900) and Jukes-Browne (1912), formally separated off the lower part of the existing Actinocamax quadratus Zone as an independent zone of Offaster pilula.

Gaster, a local solicitor, mapped the zones of the Chalk in Sussex (1924–1951). In 1920 he introduced his 'Trochiliopora bed' in the…lower portion of the Micraster coranguinum Zone… ', based on the small bryozoan Trochiliopora gasteri (Gregory), and later indicated (Gaster, 1929) that it could be traced from Beachy Head to the Adur. Gaster had been struck by the differences of opinion between Rowe (1900), Jukes-Browne (1912) and Brydone (1912, 1914) on the zonal divisions of the Chalk above the Marsupites Zone (i.e. where to place the Offaster pilula Zone and whether to introduce other divisions). By washing blocks and pieces of chalk Gaster (1924, 1929) obtained abundant small fossils (mesofossils). He found numerous rostra of the delicate and very fragile small echinoid Hagenowia blackmorei Wright and Wright (then misidentified as H. rostrata (Forbes)) and suggested that this could also be used as a guide fossil for the highest unit of his revised concept of the Offaster pilula Zone. He inferred the position of this zone in other exposures from Hampshire and the Isle of Wight based on published lists of fossils from those exposures.

Gaster (1924) modified Brydone's restricted Zone of Gonioteuthis (then Actinocamax) quadrata with a Subzone of Saccocoma cretacea (now Applinocrinus cretaceus) about 40 m thick, and a horizon of Hagenowia rostrata (now H. blackmorei), 15–20 m thick. In doing so, he added greatly to the list of guide fossils, particularly the mesofossils. In 1929, Gaster revised the biostratigraphy of his 'Offaster pilula Zone' (not that used today), establishing two subzones of Echinocorys scutatus var. depressa and E. scutatus var. cincta, in ascending order. Within the second, he recognized the two horizons of abundant Offaster pilula and Hagenowia rostrata and, in the latter horizon, recorded Saccocoma cretacea in chalk inside the tests of Echinocorys. He also drew attention to the appearance, in the horizon of Hagenowia rostrata, of large — followed by small — forms of Echinocorys. These are generally referred to as the 'large and small forms of Gaster' respectively.

Jefferies (1962, 1963) began the modern approach by describing in detail the lithology and fossils in the Plenus Marls. He was followed by Kennedy (1969, 1970) who gave a detailed account of the Lower Chalk in the region, using the Folkestone sections as a reference. Kennedy (1969) assigned letters to beds in measured sections containing particular groups of fossils that he then used for correlation within the southeast part of the region (1969, fig. 16). He employed the broad French ammonite-based zonation for the Cenomanian Stage introduced to the region by Hancock (1959), to establish a number of ammonite assemblages and subzones capable of being Widely correlated. Kennedy (1970) extended this work to the western margin of the Southern Province, an area studied in detail by Drummond (1967, 1970). Both researchers provided measured sections with descriptions of fossil assemblages, illustrating the age of the deposits and the influence of condensation across the Mid-Dorset Swell and the shelf area of south-east Devon.

Drummond (1970) added to the discussion of fossil assemblages and sedimentology of the south-western margin. Detailed litho- and biostratigraphical work on the Middle and Upper Chalk was not undertaken until later (Mortimore, 1983, 1986a,b; Robinson 1986).

Carter and Hart (1977a) introduced a micro-fossil zonation for the Gault and Lower Chalk comprising benthic foraminiferal zones designated by numbers; these zones were extensively used in investigations connected with the construction of the Channel Tunnel (Harris et al., 1996b). Hart's co-workers, Bailey and Swiecicki, extended the microfossil stratigraphy to the whole of the Chalk (Bailey et al., 1983).

The biostratigraphy of the Upper Cretaceous Chalk for the region is now based upon detailed measured sections. First occurrences, last occurrences and ranges of macro-, meso-, micro-and nannofossils are plotted in relation to lithological marker beds (e.g. Bailey et al., 1983, 1984; Mortimore, 1986a). For this purpose, the most complete sections are essential. Within the region there are several potential basal boundary stratotype sections for the international Stages or Substages. These include the base of the Middle Cenomanian Substage at Southerham Grey Pit, Lewes, the base of the Santonian Stage and the base of the Campanian Stage, both at Seaford Head, Sussex (see Cuckmere to Seaford GCR site report, this volume).

There are two major regional variations in the biostratigraphy in the Southern Province. In the first, fossils that are common on the Thanet Coast, for example Late Santonian belemnites at Margate, are rare at similar horizons in Sussex. These differences may reflect palaeogeographical settings, Kent being on the edge of the Anglo-Brabant Massif and Sussex in a more basinal setting. Local anomalies also occur. The second relates to the marginal facies of southwest England. Detailed studies of the faunas, notably of the ammonites by Kennedy (1971) and Wright and Kennedy (1981, 1984) and of the echinoderms by Smith et al. (1988), have shown that the Cenomanian faunas of south-west England are not only extraordinarily diverse, but that they include species not recorded elsewhere in Britain. This makes detailed correlation between the two facies difficult, but adds to the international importance of the Devon faunas.

The GCR sites in the Southern Province cover the entire stratigraphical range of the preserved Upper Cretaceous Series in the region, beneath the Palaeogene erosion surface. These sites also serve to illustrate the influence of sea level and tectonics on the formation of the Chalk. The localities are arranged in geographical order from Devon to Dorset and Hampshire, and through Sussex to Kent. Many of the sections form the classic white Chalk cliffs of Albion.

References