"Spongiostromata-type stromatolites" has been changed throughout to "thrombolites" which is the correct modern term. A photograph has been added at the beginning. Some diagrams are being completely redrawn. Minor additions or changes are shown in square brackets. The references are the same, but may be in slightly different format. It is intended to add some new ones in a supplement.

PURBECK FORMATION (OR GROUP), UPPER JURASSIC -
LOWER CRETACEOUS, LAGOONAL FACIES, DORSET, SOUTHERN ENGLAND

Purbeck - Formation, - Facies (Jurassic to Lower Cretaceous)
Purbeck - Evaporites (late Jurassic)
Purbeck - Durlston Bay, Swanage - Peveril Point, Upper Purbeck
Purbeck - Durlston Bay, Middle Purbeck Formation
Purbeck - Durlston Bay - Lower Purbeck Formation.
Purbeck - Durlston Bay - Central Zigzag Part & Erosion
Purbeck - Durlston Head - Lower Purbeck & Portland Stone
Purbeck - Formation - Geological Bibliography - General
Purbeck - Formation - Geological Bibliography - Topics, Alphabetic.
Purbeck - Formation - Fossil Forest Exposure, thrombolites, evaporites
Purbeck - Formation - Fossil Forest and Isle of Portland Fossil Trees
Purbeck - Ridgway Railway Cutting, Weymouth - Purbeck Formation
Purbeck - Portesham Quarry, north of Weymouth.
Purbeck - Poxwell Quarry, east of Weymouth.
Purbeck - Formation - Dinosaur Footprints of the Isle of Portland
Purbeck - Formation - Bibliography - Vertebrates
Purbeck - Formation - Analogues

.

.

Abstract:

WEST, I.M. Evaporites and associated sediments of the basal Purbeck Formation (Upper Jurassic) of Dorset. Proceedings of the Geologists' Association, vol. 86, part 2, pp. 205-225 (updated in 2012)

Four facies of limestones, each with particular contents of calcitised evaporites and of skeletal debris were recognised. They are compared with sediments of modern evaporite-depositing environments. The lowermost limestones, "stromatolitic" [correction - thrombolitic, mainly, although there are also some stromatolites] and pelletoid with foraminifera, probably originated in intertidal to shallow subtidal, moderately hypersaline, water. Overlying pelletoid limestone with algal-mats [correction - microbial mats] and some gypsum are products of high-intertidal flats. The main evaporite beds were originally gypsum, probably formed in supratidal to intertidal, very hypersaline, palaeoenvironments. The gypsum was converted to anhydrite and later brecciated in part, forming the Broken Beds [an old name]. Extensive calcitisation produced porous unfossiliferous limestones. Ostracodal limestones above probably originated in shallow, only moderately hypersaline water. All the basal Purbeck strata were formed in and around a large shallow gulf with extensive tidal flats [not of lunar tides, but of wind "tides"] and with water of varying but predominantly high salinities. At times of uplift, thin soils [rendzinas with charcoal] developed on the former margins of tghe gulf. Forests [of coniferous, Cypress-like trees] were able to exist there because, although the area was within the semi-arid zone [with very seasonal winter rains], it was probably very near to the boundary of the warm temperate zone.

[Present address, 2012: Ian West, Benedict Close, Romsey, Hampshire, England, and Visiting Scientist at Faculty of Natural and Environmental Sciences, Southampton University.]

1. INTRODUCTION

The Purbeck Formation of south and south-eastern England consists of carbonate and clay sediments with evidence of formation in shallow waters of various salinities. They represent the products of a major regression [Tithonian-Berriasian] at the end of a phase of Upper Jurassic marine sedimentation.

Throughout much of southeast England gypsum and anhydrite occur at the base of the Purbeck Beds (Howitt, 1964), attaining about 12m. in some parts of the basin (Taitt and Kent, 1958). Evaporites were also originally present in Dorset, the area in which basal Purbeck strata are best exposed, but they have been largely replaced by carbonate and partly removed in solution. This area is most important because a lateral transition can be observed from strata typical of the central part of the basin to beds representing the western marginal facies. This paper describes the former occurrence of evaporitic strata in each of the main Dorset exposures (Fig.1), as determined by petrographic study. Evidence is presented for the palaeoenvironments in which the evaporitic beds originated.

The basal Purbeck Formation [or Group] overlies the Portland Stone Formation [Portland Freestone Formation of the Portland Group] without a sharp break.

[END OF P. 205]

[START OF P.206]

Comparison of these underlying limestones with modern carbonate sediments, discussed below suggest an origin in sea-water of less than 50 parts per thousand salinity rather than hypersaline water [for more on Portland Stone see Townson (1975)].

The basal Purbeck limestones are poorly fossiliferous, fine-grained and often laminated. They are usually divided into the units listed below ( Strahan, 1898), which although not precisely defined are usually regarded as Members, and are convenient for field use.

(c) 'Cypris' Freestones (mainly ostracodal limestones)
(b) Broken Beds (limestone breccia)
At the base (a) Caps and Dirt Beds [thrombolitic] ("stromatolitic")and pelletoid [peloidal] limestones and carbonaceous marls or rendzina palaeosols.

Stratigraphic information on the Purbeck Formation has be summarised [up to 1975] with references to earlier literature by Strahan, 1898, Arkell, 1947, and Anderson and Bazley (1971). Petrographic features of some Lower Purbeck limestones have been described by Brown (1963; 1964). Diagenesis of Purbeck evaporites has been discussed by West (1964; 1965), Shearman (1966), Holliday (1973) and Holliday and Shephard-Thorn (1974).

The lithology of limestones associated with the evaporites is summarised in the vertical sections of Figs. 2 and 3. Thrombolitic limestones or thrombolites are considered here, only briefly, in terms of the two major categories of Brown (1963) and Pugh (1969).

[For later literature see associated webpage: Bibliography for the Purbeck Formation.]

2. EVIDENCE FOR THE FORMER PRESENCE OF EVAPORITES

West (1964), West and others (1968), Folk and Pittman (1971) and West (1973) have given criteria for detecting the former presence of evaporites, since replaced or removed in solution. The main

[END OF P.206]

[START OF P. 207]
indicators of former calcium sulphate in Purbeck strata are listed below. Search was made for these criteria in each bed of each of the main exposures. Conclusions were usually based on several such features in association.

(1) Pseudomorphs of chalcedony (particularly lutecite) or of calcite after lenticular crystals of gypsum or after tabular or rod-shaped crystals of anhydrite.
(2) Spherulites of lutecite, often with euhedral crystals of authigenic quartz.
(3) Celestite, sometimes with calciostrontianite.
(4) Net-texture (Brown, 1931 ; West, 1964 )
(5) Chicken-wire, nodular structure, or spherical cavities formed by the dissolution of nodules.
(6) Porous, coarsely crystalline limestones without skeletal debris.
(7) Contortions in lithified limestone, other than subaqueous slump structures, suggesting that plastic flow has occurred.

Many of the limestones contain only relics of only small quantities of evaporites, often in a pelletoid [peloid] matrix. Three rock types, however, consist almost entirely of replacements of calcium sulphate.

The first of these is found at most exposures of basal Purbeck strata. It is calcitised secondary anhydrite ('secondary limestone' of West, 1964 ) which has resulted from alteration of calcium sulphate to calcium carbonate. This replacement process has been shown by West, 1964 to have occurred partly before and partly after the tectonic brecciation that formed the Broken Beds [a limestone breccia]. The limestone produced by the calcitisation is white or light grey in colour, coarsely crystalline, often saccharoidal and unfossiliferous. It is frequently, although not in all case, extremely porous. It is often laminated and with nodular or chicken-wire structure or net-texture. Pseudomorphs after anhydrite ofen occur. Euhedral crystals of quartz and spherulites of lutecite are common.

Celestite rock, a less common replacement of secondary anhydrite, occasionally occurs within beds of calcitised anhydrite. It usually consists of euhedral crystals and shows either nodular structure or net texture. Calcite, lutecite, authigenic quartz and sometimes calciostrontianite (Salter and West, 1965) are accessory minerals.

Nodules or thin beds of chert which has replaced calcium sulphate are common. The chert consists of pseudomorphs, usually of lutecite, after either lenticular crystals of primary gypsum or after secondary anhydrite.

3. DISTRIBUTION OF EVAPORITIC STRATA

(a) Carbonate and evaporite facies and dirt beds

The distribution of evaporites in each of the main sections is summarised diagrammatically in Figs. 2 and 3 and details are described in the Appendix. Four main facies (Fig.4) which differ in their quantities of replaced evaporites and of skeletal debris, were distinguished. They are interrupted by thin beds of carbonaceous marl known as 'dirt beds' [rendzina palaeosols].

.

(i) Facies A - Limestones with Foraminifera [and with Thrombolites] [generally without evaporites].
(lowest Purbeck facies above the Portland Freestone)

This, the lowermost facies is characterised by the presence of foraminifera and by the absence of replaced gypsum. It consists largely of thin-bedded limestones with fine-grained pelletoid textures. However, thrombolites are common features of this facies and often form continuous beds. The Lower Dirt Bed (Figs 2 and 3) and other thin dirt beds occur within facies A.
Remains of peculiar faunas, not only of foraminifera, but also of calcispheres, ostracods, small gastropods (Hydrobia and Valvata) and bivalves are frequently present. The isopod

[END OF P. 207]

[DIAGRAM OF P. 208-209]

[The diagrams above is an updated version of Figs. 2 and 3 of West (1975). It has been completely redrawn for clarity.]

Caption for Figs. 2 and 3 above.

Distribution of replaced evaporites in, and lithology of, the basal Purbeck limestones of Dorset as determined by thin-section petrography. The facies A-D of which the boundaries are shown, are discussed in the text. The Portsdown borehole section, shown for comparison is based on: Taitt and Kent (1958). Note that the thicknesses of the thrombolitic limestones vary in short distances. The positions of tree remains are based partly on the author's observations and partly on previous reports.

[The following is a supplementary diagram showing details of the basin margin in the basal Purbecks of the Lulworth Cove area. This was not in the original paper.]

.

. [END OF P.209]

[START OF P. 210 BELOW]

Archaeoniscus and fish have been recorded from these beds (Fisher, 1856). The red alga Solenopora has been found by Pugh (1969).
The base of the facies is recognised in the field by the first appearance of fine lamination, an abundance of ostracods and the disappearance of large marine molluscs.

(ii) Facies B - Limestones with Thrombolites and Some Replaced Gypsum.

Except for occasional remains of ostracods, skeletal debris is almost absent in this facies. Small lenticular crystals of gypsum, since replaced by calcite or chalcedony are common in the limestones and cherts. Large thrombolites occur above dirt beds or palaeosols [often around silicified tree stumps] and remains of microbial mats are particularly characteristic of this facies.

Many of the limestones are pelletoidal or pelloidal but in the Lulworth area there are some unusual, fine-grained oolites (of lagoonal origin). The ooids often contain nuclei of lenticular crystals of gypsum, since replaced by calcite (West, 1964) . Pebbles of this unusual oolite occurring within the Great Dirt Bed palaeosol differ in that although the interstitial calcite cement has survived the oois have been dissolved (they have oomouldic porosity). it is probable that, like most modern examples (Loreau and Purser, 1973), the ooids were aragonitic and thus they were prone to dissolution in the subaerial environment in which the dirt bed, the palaeosol, originated.

(iii) Facies C - Calcitised Evaporites.

This, the major evaporite facies, is unfossiliferous and consists almost entirely of calcitised secondary anhydrite. Chert nodules which have replaced calcium sulphate are abundant and some celestite is occasionally present. Thin beds of calcareous shale and some pelletoid limestone occurs. Details of these evaporitic beds are given in the Appendix. Representative sections are those of Worbarrow Tout and particularly the calcitised anhydrite of Durlston Head.

[This calcium sulphate evaporite facies represents the marginal fringe of the well-known Purbeck anhydrite of southeastern England, an important seismic reflector of importance to the local oil industry. Near surface at Brightling and Mountfield in Sussex the anhydrite is hydrated to gypsum and worked economically on a large scale.] [this paragraph is an additional note, 2012]

(iv) Facies D - Limestones with Ostracods.
(higher in the sequence)

This facies largely corresponds to the 'Cypris' Freestones Member. It contrasts strikingly with the underlying facies partly in its abundance of ostracods ("Cypris purbeckensis", not the Cypridea etc that occurs higher in the Purbecks). It also differs from beds immediately beneath in the relative lack of replaced gypsum. Normal marine faunas do not occur but the small lagoonal gastropods Hydrobia and Valvata are present as in Facies A (Clements, 1969). Ostracodal biosparites, intrasparites and pelletoidal limestones are dominant rock types but calcareous shales also occur and, in the western part of the area, north of Weymouth, some oolites are present. The most common evidence of evaporites consists of occasional casts of halite crystals, which are composed of faecal pellets or other sediment. Calcite pseudomorphs after gypsum are relatively rare.

In the field these limestones have a laminated appearance which is caused by ripple cross-bedding. Surfaces with ripple marks are common features and sediments appear to have consisted of carbonate allochems of sand and silt size. Dirt beds (palaeosols), microbial mats and large thrombolites are not usually present, although Girvanella nodules have been recorded by (Pugh, 1969).

As in facies A, remains of the isopod Archaeoniscus occur (Bristow and Forbes in (Damon, 1884). The fish Ichthyokentema purbeckensis (Davies) was considered by Griffith and Patterson (1963) possibly to have been adapted to hypersaline conditions.

(v) Dirt Beds [Rendzina Palaeosols with charcoal] and Associated Marls.

The "dirt beds" [an old quarrying term] are thin bands of black or dark brown, carbonaceous marls sometimes with well-rounded limestone pebbles. They sometimes contain minute carbonate concretions but are relatively deficient in quartz sand. There is little subsoil and usually no gradual transition upwards from the bedrock. The underlying limestones appear to have been eroded to some extent prior to their formation.

A peculiar feature of the pebbles in the Great Dirt Bed, the highest dirt bed (Figs. 2 and 3) is that many of them are black. The black limestone of which they consist is petrographically similar, except in pigment, to the limestones of the buff-coloured pebbles and the underlying limestone. They occur particularly around the Mupe Bay area of penecontemporaneous movement and also on Portland.

The Great Dirt Bed passes laterally westward in the Weymouth area (see Appendix) into buff- coloured, less carbonaceous, marls containing freshwater fossils and chert nodules which have replaced gypsum. The origin of these marls is discussed below.

.

(b) Thickness variations of the facies

Facies A occurs throughout the region at the base of the Purbeck but is thin in the east (Figs. 2. 3 and 4). Facies B is markedly wedge-shaped and is clearly a marginal facies, also thinning towards the basin centre. The main evaporite facies (C) on the other hand, is thin north of Weymouth and on the Isle of Portland and thickens in an eastward, basinward, direction (Fig 4). It attains more than 8m. at Durlston Head and appears to be the equivalent of the beds of calcium sulphate of the basal Purbeck at Portsdown (Fig. 3) and other localities to the east. The upper limit of facies D, only well-exposed in the eastern area, is not shown in Fig. 4.

There are local variations in thickness of the facies near Lulworth (Fig. 4). At Dungy Head the Lower Dirt Bed rests on Portland Stone. At Bacon Hole, near Mupe Bay, facies A and B suddenly increase in thickness and the Great Dirt Bed is displaced upwards. Facies A there contains some small subaqueous slump structures. These features suggest that the Lulworth and Mupe Bay area was subject to small oscillations during sedimentation, with uplift after deposition of the Portland strata, followed by slight depression during deposition of much of facies A and B.

Pebbles of limestone within and beneath the Great Dirt Bed increase in diameter from the west towards this area. Local uplift, prior to the development of the dirt bed in which the pebbles were later incorporated, perhaps resulted in penecontemporaneous erosion of the underlying limestone.

Further evidence of local uplift is the reduction in total thickness of the Purbeck Formation near Lulworth ( (Arkell, 1938); (Howitt, 1964) and, particularly, of Upper Cretaceous strata (Strahan, 1901) . Drummond, (1901) suggested that an unexposed intra-Cretaceous fault bordered the area of Upper Cretaceous uplift. Since the present evidence reveals that uplift has occurred earlier such a fault may have been in existence at an earlier date. It may thus account for the presence of an unusual conglomeratic oil sand in the Wealden Beds at Mupe Bay (Lees and Cox, 1937). Seepage of oil up the fault plane to the Wealden land surface perhaps resulted in the impregnation of unconsolidated sand sediments, followed by erosion, local transport and deposition of oil-bound clasts.

.

(c) Relationship of the Broken Beds to the evaporitic facies

The Broken Beds, the well-known limestone breccias, are absent at Portesham, the most westerly exposure, and also on the Isle of Portland. Only thin developments occur at Upwey

[START OF P. 212]

[Original caption:}

Fig. 4. East-west sections of the basal Purbeck strata of Dorset. The lower diagram shows strata discussed in the text. The upper diagram shows the distribution of breccia (the 'Broken Beds') which can be compared with the distribution of Facies C (replaced evaporites). [This diagram has been improved by having been redrawn in colour with additional information on facies and palaeosalinities. See text for details.]

[CONTINUING TEXT OF P. 212, BELOW DIAGRAM] and at Holworth House, Ringstead (Fig. 4). At Durdle Door, however, the Broken Beds are 2m. thick and they increase in thickness eastwards. In the east of the Isle of Purbeck the lower limit of brecciation descendsd well below the level of the Great Dirt Bed.

It is obvious that the breccia varies in thickness in a manner remarkably similar to that of the main evaporitic facies (C). The base of this facies, however, is often lower than the base of the breccia while the top of the breccia is often higher than the evaporites.

The effect of this is seen in the Lulworth area where the Broken Beds can be divided into two distinct parts. The upper consists mainly of joint bounded blocks of ostracodal limestone with

[END OF PAGE 212]

[P. 213 BELOW]

----- little or no matrix. The limestone is similar in lithology to the 'Cypris' Freestones above, and this part of the breccia belongs to facies D. In contrast the lower portion shows characteristics of facies C. It consists mainly of blocks of unfossiliferous limestone with a matrix of similar but less lithified limestone. Both blocks and matrix are calcitised secondary anhydrite. Pseudomorphs after anhydrite in fragmented chert show that much conversion of primary gypsum (stage I of West, 1964) to secondary anhydrite (stage III ) had occurred prior to brecciation. Farther east almost the whole thickness of the Broken Beds presents evidence of evaporites and lies within facies C (Fig. 4). The relationship to former evaporites might be regarded as evidence for a collapse origin of the breccia. Arkell (1938), however, has convincingly demonstrated that the brecciation was tectonic in origin rather than merely due to collapse. Tectonic contortions at Durlston Head and other localities provide confirmatory evidence. The originally plastic evaporites apparently formed an incompetent bed that was technically contorted and brecciated. The basal part of the sequence of the brittle ostracodal limestones above was also brecciated at many localities. Such an origin ior the Broken Beds was suggested by Hollingworth in 1938 before the discovery of replaced evaporites in this area.

4. BASAL PURBECK PALAEOENVIRONMENTS

(a) The Palaeoclimate

The Purbeck evaporites have sedimentological and faunal associations which suggest that climatic conditions were unusual for evaporites. The former occurrence of the bedded gypsum and of occasional halite crystals demonstrates that conditions were dry in the southern England area at that time. No evidence exists, however, for desert conditions. There are no red beds or blown sands. The absence of bedded halite, even in deep boreholes, and the occasional presence of freshwater beds suggests that, at most, the climate was only semi-arid. Indeed, the insect and molluscan faunas of the Purbeck Formation indicate a warm-temperate palaeoclimate Arkell (1947). Furthermore, rooted in the Purbeck dirt beds are numerous remains of coniferous trees; forests would not be expected to tolerate conditions of any appreciable aridity.

The evidence is explicable when the world distribution of evaporites of Upper Jurassic and Lower Cretaceous age (Green, 1961; Lotze, 1964); is examined. In Lower Purbeck times southern England appears to have been situated at the northern limit of the northern hemisphere zone of evaporite-deposition. Although just within the dry (semi-arid or steppe) zone the region was so close to the boundary of the warm-temperate (sub-humid) zone that warm-temperate types of fauna and flora could often exist. Studies of ostracods ( Anderson in Allen, Keith, Tan and Deines, 1973; Anderson, 1973) suggests comparison with the modern southern Mediterranean-North African area. The boundary between the warm-temperate and dry zones crosses that region at present.

(b) Depositional environments of the facies

Evidence for the origin of the basal Purbeck facies is provided by comparison with certain recent shallow-water sediments. Difficulties of interpretation could arise, however, if the evaporites had been developed penecontemporaneously beneath the surface in sediments that had originated in different environments. Gypsum or anhydrite is sometimes precipitated within arenaceous tidal- flat or sabkha sediments at the present time (Masson, 1955; Shearman, 1966)

[END OF P. 213]

[START OF P. 214 BELOW]

There is the following evidence that this did not usually take place in the basal Purbeck sediments. Firstly, remains of evaporites are usually absent in beds containing skeletal debris. Indeed, unfossiliferous beds with evaporites often overlie fossiliferous beds without trace of gypsum or anhydrite (Fig. 2). Secondly, the well-developed lateral continuity of thin evaporitic beds suggests a sediment-surface origin. Thirdly, remains of gypsum sand-crystals, the form of gypsum which usually develops beneath the surface, are very rare. The lack of evaporites of this type may result from the Fine particle-sizes of the Purbeck sediments and the presence of laminae of clay. These probably restricted vertical movement of water containing sulphate ions.

(i)Facies A

Hypersalinity during deposition of facies A is suggested by the absence of normal marine faunas and the position o: the beds between marine and evaporitic strata. This is supported by the presence of algal stromatolites which, like modern stromatolites, probably existed in water too hypersaline for destructive grazing and burrowing molluscs (Logan and Cebulski, 1970; Garrett, 1970). Nevertheless, because of the existence at times of some limited molluscan, foraminiferal and ostracodal faunas and the absence of gypsum it is unlikely that palaeosalinities were extremely high. Modern lagoons of Shark Bay and of the Persian Gulf with similarly limited faunas have salinities between about 50 and 70 parts per thousand (Logan & Cebulski, 1970; Hughes-Clarke and Keij, 1973) . Isotope studies by Allen and Keith (1965) of certain limestones belonging to facies A are in accordance with such moderate values.

The carbonate pelletoid silts of the facies A limestones resemble the aragonitic pelletoid sediments of the Persian Gulf lagoons referred to above (Illing and others, 1965; (Evans and Bush, 1969) . Thus, although the Purbeck gulf was, of course, of much greater size than these small modern lagoons, an origin in similarly shallow water is likely. The thin dirt beds of this facies record phases of subaerial exposure. The spongiostromata-type stromatolites which often occur above the dirt beds resemble modern stromatolite heads. These characterise the intertidal zones of hypersaline lagoons (Logan, 1961; Logan and Cebulski, 1970) and the Purbeck stromatolites probably originated in similar environments. Necessity of suitable substrata for the attachment of algal filaments probably accounts for many of them having formed over the stumps of submerged trees. Algal mats also occur.

Thus facies A represents the sediments of shallow, moderately hypersaline water and of intertidal flats. Occasionally there was subaerial exposure and development of soils,

(ii)Facies B

Thrombolites of this facies also suggest hypersaline conditions. Higher palaeosalinities, how ever, are indicated by the general absence of faunal remains and the occasional precipitation of gypsum. Even ostracods occur only occasionally in facies B and palaeosalinities were probably persistently near the values necessary for gypsum precipitation, about 124 parts per thousand according to Friedman and others (1973).

As in facies A, evidence of intertidal conditions in this facies are provided not only by thrombolites but also by algal mats. Similar mats occur at the present day in protected hypersaline, high-intertidal, zones of many lagoons including those of the Persian Gulf (Illing and others, 1965; Shearman, 1966) and of Shark Bay (Logan, 1961). As observed by Pugh (1969) large-scale polygonal dessication cracks in the Purbeck algal mats can be matched in modern examples (Fisk, 1959; Evamy, 1973).

The replaced gypsum crystals of facies B (West, 1964, plate 36, fig. 1) are in size and shape like the gypsum crystals of the intertidal zone of the Persian gulf lagoons (Shearman, 1966; Butler, 1969)

[END OF P.214]

[START OF P. 215 BELOW]

A single occurrence has been found in this facies of large gypsum crystals with characteristics like those of the modern supratidal sabkhas (West and others, 1969). Even this thin bed, however, contains evidence of algal-mat lanimation.

The gypsum crystals that occur within ooids could only have received oolitic coatings without dissolving, in water saturated or supersaturated for gypsum. Thus they demonstrate that water existed which was sufficiently hypersaline for direct precipitation of gypsum without requiring supratidal sabkha conditions [i.e. the Purbeck lagoon was hypersaline at this time].

The area in which these oolites are present [around Lulworth Cove] is that of the penecontemporaneous downwarping [probably a local fault block of the Late Cimmerian movements, which were starting at this time]. This small amount of depression may have formed a shallow lagoon within the broad belt of tidal flats. Similar in some respects is the small lagoon of Khor Odaid in the Persian Gulf [or Arabian Gulf], described by (Loreau and Purser, 1973). There, fine-grained oolitic sands occur in the intertidal and shallow sub-tidal zones in hypersaline water. In a nearby subtidal part of this lagoon precipitation of gypsum occurs.

Thus, there is abundant evidence of hypersaline intertidal conditions. As in similar modern environments, the tidal variation was probably mainly a product of wind-action (Rusnak, 1960; ( Illing, Wells and Taylor, 1965). The tidal-flat conditions were interrupted for an interval of unknown duration when hypersaline water withdrew from the area and soils were developed, forming the Great Dirt Bed [rendzina palaeosol].

(111) Facies C [main evaporite facies]

The replaced calcium sulphate of this facies is sometimes nodular. The Purbeck nodular sulphate (West, 1965) closely resembles nodular anhydrite in the supratidal sabkhas around the Persian Gulf lagoons, and (Shearman, 1966) and (Butler, 1969) have suggested that it was formed in a similar environment. Typical features of the Persian Gulf supratidal sabkhas are absent, however, and the present writer previously thought that a supratidal origin was unlikely. Blown sand, dolomitic sediments and gypsum sand-crystals (c.f. (Shearman, 1966); Evans, 1970) are missing. This is not evidence against supratidal conditions, however, since differences in climate and supply of clastic sediments may account for the lack of these features in the evaporites. A largely supratidal origin is thus probable, but some phases of intertidal conditions are implied by the presence of lenticular crystals of gypsum (now replaced by calcite or silica) like those of the Persian Gulf, intertidal-zone sediments (Shearman, 1966)

(iv) Facies D [ostracod facies - Cypris Freestones etc]

Faunas consisting mainly of abundant ostracods show that conditions were usually either intertidal or subtidal. The larger particle-size of the allochems and the abundance of ripple-marks and small-scale current-bedding suggests that wave or current action was more vigorous than during the formation of the other facies. High energy conditions were probably unfavourable for the development of algal-mats. The occasional sediment casts of halite crystals demonstrate that at intervals subaerial exposure took place and shallow water was evaporated. Halite crystals, so-formed, were probably dissolved by later submergences and the moulds filled with sediment.

The occasional evidence of evaporites, the presence of ostracods and the absence of normal marine faunas suggests moderate hypersalinities. Similarities in general faunal content between this facies and facies A perhaps indicates that palaeosalinities were, once again, frequently between about 50 and 70 parts per thousand. Where ostracods are the only abundant faunal remains, comparison can be made with those modern Persian Gulf lagoons where salinities of over 70 parts per thousand permit the survival only of ostracods (Hughes-Clarke & Keij, 1973). Nevertheless, palaeosalinities sufficiently high for precipitation of gypsum were rarely attained.

[END OF P. 215]

[START OF P. 216 BELOW]

Fig. 5 - Portesham

The diagram above is a modified version of:

Fig. 5. Vertical variation in the basal Purbeck strata at Portesham. Lithology of the Portland limestones is not shown. The interpretation in terms of palaeosalinity is discussed in the Appendix.

.

[END OF P. 216 FIG]

[START OF P.217 BELOW]

(c) Depositional environments of the dirt beds and marls

The trees rooted in the dirt beds confirm the origin of these as soils (Arkell, 1947). The presence of limestone pebbles in the Great Dirt Bed and the eroded surfaces of the underlying limestone suggests that lithification of that limestone had occurred prior to the development of the soil.

The black pebbles which occur within and at the base of this soil resemble modern pebbles of limestone which obtain a black pigment on the shores of hypersaline lakes Ward and others. As discussed above, the Purbeck pebbles probably originated in an uplifted area in which beaches may have developed. Similarly, black limestone pebbles in the evaporitic Purbeck strata of the Jura Mountains of Switzerland have been derived from shallow-water areas of penecontem-poraneous uplift, according to Carozzi (1948).

Providing important palaeoenvironmental evidence are the marls of fresh or brackish water origin with chert at the horizon of the Great Dirt Bed at Portesham (Fig. 5, bed 19). The chert contains not only pseudomorphs after gypsum but also charophytes and other apparently freshwater fossils, including mollusca and ostracods. Land plant remains are present (see Appendix). Comparison with modern environments with charophytes suggests palaeosalinities of less than 10 parts per thousand Groves and Bullock-Webster (1948).

An origin for this strange mixture is indicated by ephemeral lakes around the coastal lagoons of South Australia Alderman & Skinner, 1957). These occur in a mediterranean climate near the temperate border of the southern hemisphere belt of modern evaporite deposition. They therefore exist in an area climatically comparable to the Purbeck environments. These lakes are fresh or brackish during the winter, with abundant water plants, molluscs and crustaceans, but in summer they increase in salinity until a little hatite is deposited.

The lakes also suggest an origin for the chert. During intensive photosynthesis by the plants a pH sufficiently high for dissolution of silica is developed in the water. A much lower pH in the underlying sediments causes precipitation of gelatinous silica (Peterson and von der Borch, 1965). A source of silica for this and other Purbeck cherts (House, 1968) might have been diatoms, which would be expected in algae-rich sediments.

The charophyte chert is a local feature occurring only near the western margin of the basin. Thus it was probably formed in an ephemeral lake or temporary lagoon of small size (Fig. 6). Sulphate ions were presumably supplied by occasional floods of water from the gulf to the east.

.

(d) Palaeoenvironmental significance of sedimentary cycles

In Lower Purbeck strata of the Warlingham and Fairlight boreholes, cycles of sedimentation apparently like those of the Persian Gulf sabkhas have been recognised (Shearman, 1966; (Holliday and Shepard-Thorn, 1974); In Dorset, most of the vertical sections (Figs. 2 and 3) show the following sequences:

3. Gypsum (since calcitised) and limestone.
2. Thrombolitic and pelletoid (peloidal)limestone.
1. Dirt bed (palaeosol) with rooted coniferous trees, on an erosion surface. (at the base). One sequence of this type commences with the Lower Dirt Bed and another with the Great Dirt Bed.

In terms of palaeoenvironments the components of the eye-es may be interpreted in the following way. The dirt beds (1) are palaeosols, the remains of soils. The thrombolitic limestone (2) has been formed by a transgression resulting in intertidal conditions. The coating of upstanding tree trunks by algal limestone for up to 4.6 m. before they decomposed and collapsed (see Portesham section

[END OF P. 217]

[START OF P. 218 BELOW]

----- in Appendix) shows that submergence was rapid. The gypsum (3) originated on intertidal or supratidal flats and probably indicates a long regressive phase. It is sometimes followed by the dirt bed (palaeosol) of another cycle.

These cycles usually differ from typical sabkha cycles by lacking units of 'lagoonal' or subtidal origin. The dirt beds or soils represent the phases of maximum regression and such products of semi-arid conditions would, of course, be absent in desert sabkha cycles.

The Purbeck hypersaline gulf probably resembled the modern gulf of Kara Bogaz (Grabau, 1920) in its large size, shallowness and extensive evaporite flats. The Dorset strata originated in the western marginal area. Because of the distance of the open sea, drifted skeletal debris of normal marine origin is absent.

The basal Purbeck strata commence with the moderately hypersaline, subtidal and intertidal limestone facies (A) overlain by a limestone and gypsum facies (B) of more hypersaline, intertidal origin. The main gypsum facies (C), now calcitised, was probably formed in very hypersaline intertidal to supratidal conditions. Facies (D) above, originated in moderately hypersaline, subtidal to intertidal, conditions.

The cycles provide a guide to the positions of time-planes enabling some approximate palaeogeographical reconstructions to be made (Fig. 6). Shallow water and tidal-flats initially covered the whole east Dorset area. Regression caused soils with forests to spread over the former gulf floor (Fig. 6), forming the Lower Dirt Bed palaeosol. Rapid submergence resulted in extensive tidal-flats with thrombolites and some gypsum. Regression caused the Great Dirt Bed to be formed as the soil of an island or peninsular, to the west of which a small lagoon existed (Fig. 6). Another transgression resulted in tidal flats with hypersaline water covering the whole region and the precipitation of gypsum (Fig 6). Finally the formation of evaporites almost ceased, with further submergence of the area into the subtidal to intertidal environment.

The palaeoclimate was probably semi-arid so that as in the modern Laguna Madre of Texas (Rusnak, 1920) halite was not usually preserved. The boundary of the warm-temperate climatic zone was probably sufficiently close for the existence of the forest vegetation. At the present day, this boundary is nearest to southern England just south of the western Mediterranean Sea. There, mesophytic forests Linton exist in proximity to areas of formation of minor evaporites. If a large shallow gulf was a feature of that area at present, it would probably have generated similar sediments to those of the basal Purbeck Formation.

ACKNOWLEDGMENTS

The author gratefully acknowledges help and advice from the late Professor Frank Hodson and the late Professor Peter C. Sylvester-Bradley. He is much obliged to Dr Roy Clements, Dr. D. Barker and Dr. John R. Merefield for reviewing drafts of this paper and to the late Professor Douglas J. Shearman and Dr. W. Geoff. Townson for valuable discussion.

.

[END OF P.218 AND MAIN TEXT.]

The following brief comments are concerned mainly with the occurrence of evaporitic beds. These are not always easily recognised in the field. Bed numbers given below are those of Figs. 2 and 3 to which references should be made. Details regarding the Portesham section are given in Fig. 5. Exposures are considered in sequence from west to east.

(i) Portesham (National Grid map reference SY 611857)

[Also go to webpage:

Portesham Quarry. ]

This is the most westerly exposure of the basal Purbeck Beds. At the base of the section there is a gradual transition over a fen centimetres between typical Portland and Purbeck facies. Betweem the junction and the Lower Dirt Bed (Beds 8-10) no evaporites are present and laminated limestones with ostracods and other fauna occur. Above this dirt bed are remains of trees which were probably once rooted in it. A large sheath of algal tufa [the "Fossil Elephant" - see above] which formerly coated one such tree was described by Woodward (1895) and can still be seen in the quarry.

The algal limestone is thickened in a series of encircling rings occurring at intervals of about 15 cm. The regularity and orientation of these suggests that the tree was standing vertically in water at the time of their formation. A depth of about 4 • 6 m. is implied by the length of the algal sheath.

At the base of some marls (bed 19), probably corresponding to the Great Dirt Bed, unusual chert nodules occur. Within these, perfect lutecite pseudomorphs after lenticular crystals of gypsum even contain traces of the original (010) cleavage. Also present within this chert are silicified remains of charophytes, freshwater ostracods, gastropods and land plants (Barker and others, 1975). Affinities have been suggested (West, 1961) between the fauna and flora of the charophyte chert and problematical Purbeck strata at Swindon. The marls associated with this chert contain similar freshwater fossils and other bands of evaporitic chert.

Gypsum replaced by calcite and chert is common in beds 21 to 23 and 26 to 31. Bed 28 and the lower part of bed 30 contain black intraclasts.

The major facies, described above, can be recognised in the Portesham section. Their limits are shown in Fig. 5. Within these broad facies, the effects of smaller variations in depositional conditions can be observed in terms of replaced evaporites and of skeletal debris. Six microfacies are listed below. The main characteristics of each appears to have resulted from particular palaeosalinity conditions. Their occurrence in the vertical section is shown in Fig. 5. Obviously, many minor fluctuations in palaeosalinity will remain undetected.

1. Carbonaceous soils and associated marls containing freshwater fossils. Subaerial or freshwater conditions are indicated.

2. Limestones with marine faunas including cephalopods, sponge spicules, thick-shelled bivalves and occasional echinoderm remains. Comparison with modern environments (Hughes-Clarke and Keij, 1973; Logan and Cebulski, 1970) suggests origins in water of less than 50 parts per thousand.

3. Limestones with foraminifera, ostracods and sometimes calcispheres. Usual marine macrofossils and bioturbation are absent. Facies A is mainly composed of such beds and discussion above suggests that they probably originated in palaeosalinities between about 50 and 70 parts per thousand.

4. Limestones either without fossils or with only ostracods. Discussion above indicates that such beds, of which facies D here consists, originated in palaeosalinities between about 70 and 124 parts per thousand.

[END OF P. 220]

[PAGE 221]

5. Unfossiliferous limestones with small quantities of replaced evaporites. Palaeosalinities probably occasionally attained about 124 parts per thousand necessary for sulphate precipitation.

6. Replaced evaporites, of which facies C usually consists, probably indicate persistent palaeosalinities above the value required for sulphate precipitation.

(ii) Upwey (SY 671851)

Although poorly exposed, this quarry section shows many similarities to that at Portesham. Supplementary information comes from Fisher's (1856) account of exposures, since grassed over, in the adjacent Ridgeway railway cutting.
[Some Further illustrations of the basal Purbeck Group in about 2010-2012 can be seen in the webpage: Ridgeway Railway Cutting and the Weymouth Relief Road Cutting.]

As at Portesham the lowest bed of Purbeck facies is fused to the Portland Stone. Here too, tree remains were found on the Lower Dirt Bed. Minor quantities of replaced gypsum occur in limestones above the Lower Dirt Bed and Fisher (1856) recorded freshwater gastropods in a marl (bed 9) probably corresponding to the Great Dirt Bed in the railway cutting. Correlation with the Portesham charophyte chert and associated marl is suggested.

Bed 10 is in part a replacement of gypsum or anhydrite and contains mud-clasts. Most chert replacement of gypsum and some brecciated limestone in beds 11 to 14 probably represents the evaporites of the Broken Beds. In bed 11 some quartz sand grains and black intraclasts are present. Finally, at the top of the section, harder limestones of 'Cypris' Freestone lithology follow.

(iii) Chalbury Camp (SY 693838)

Nodules of silicified lenticular gypsum occur in the Lower Dirt Bed at this locality. Some recemented limestone breccia (bed 10) is probably of evaporitic origin. The main evaporite bed (bed 13), lying about 0-5 m. above, consists of calcitised calcium sulphate and contains pseudomorphs after gypsum both in the limestone and in nodules of chert.

(iv) Poxwell Quarry (SY 744835)

See also the Poxwell Webpage.

Few indications of evaporites were found here. Calcite pseudomorphs after gypsum occur near the middle of the section (bed 10) and crystal moulds are present in the overlying marly limestone. Chert nodules, probably replacements of gypsum, occur in the Lower Dirt Bed as at Chalbury Camp. Sponge spicules and gastropods have been found by Clements (1967) in beds 1 and 2.

(v) Holworth House Cliff Section, Ringstead (SY 762816)

Chert with pseudomorphs after gypsum occurs in a limestone breccia (bed 14) at about 4 m. above the base of the Purbeck Beds. The thin breccia appears to be a representative of the Broken Beds.

[See also the Ringstead - White Nothe Webpage.

(vi) Durdle Door (SY 807802), near Lulworth Cove.

A section is accessible at the eastern end of the promontory. Here, calcitised, brecciated, secondary anhydrite occurs at the base of the Broken Beds.

[See also the webpage: Durdle Door]

(vii) Dungy Head (SY 815800), near Lulworth Cove.

In the Hard Cap minute lenticular crystals of gypsum, since replaced by calcite, form the nuclei of small ooids. Calcitised evaporites are again present in the lower part of the Broken Beds.

[See also the webpage: Dungy Head]

(viii) Stair Hole and Lulworth Cove (SY 821799-827797)

At Stair Hole, just west of Lulworth cove, chert in the Soft Cap contains both small pseudomorphs after gypsum crystals and silicified remains of the gastropod Hydrobia. A small nodule of celestite was found in the same bed [on the central ridge]. The former vertical distribution of evaporites in the Caps and Broken Beds at Lulworth Cove and Stair Hole is otherwise similar to that of the nearby and better exposed Fossil Forest section, described below.

[See also:
Stair Hole Webpage.

Lulworth Cove, Purbeck Formation - West Side of Cove - Webpage.

Lulworth Cove, Purbeck Formation, East Side of Cove - Webpage .]

[END OF P. 221]

[START OF PAGE 222]

(ix) Fossil Forest, Lulworth Cove (SY 832796)

At this locality, east of Lulworth Cove, calcite pseudomorphs after gypsum occur in a fine-grained [lagoonal] oolite of the Hart Cap, as at Dungy Head. Chert replacements of gypsum form an almost continuous bed in the Soft Cap. Calcitised secondary anhydrite is conspicuous in the same unit and in the basal part of the Broken Beds. Much of it is poorly lithified with the superficial appearance of weakly-cemented sand [there is no quartz sand here]. At the western end of this exposure there is a thin bed within the Soft Cap consisting of algal-laminated limestone with calcite pseudomorphs after large lenticular crystals of gypsum. These crystals were so orientated that their c-axes were nearly horizontal and their planes of flattening were in near-vertical positions.

(x) Bacon Hole, near Mupe Bay, East of Lulworth Cove (SY 839797)

Calcite pseudomorphs after gypsum were found in parts of the Hard Cap. In the Soft Cap a porous, crumbly, saccharoidal limestone consists of calcitised anhydrite (bed 16) and is the lateral equivalent of the quartz-calcite rock of Worbarrow Tout discussed below. More calcitised evaporites occur in the lower part of the Broken Beds.

(xi) Worbarrow Tout and Pondfield Cove (SY 868795)

[At the southern end of Worbarrow Bay]

Microbial limestones occur in the Hard Cap below the main evaporitic strata on the west side of Worbarrow Tout. The lowest of these, bed 3, consists of thrombolitic limestone whereas beds above it consist largely of the even-laminated type of microbial mat stromatolitic limestone.

On the west side of the Tout the Hard Cap contains some celestite which has replaced gypsum or anhydrite. The Great Dirt Bed palaeosol (see Arkell(1940, plate 4) contains both celestite and lutecite in addition to carbonaceous debris [charcoal from forest fires is abundant in the Great Dirt Bed at other places, and probably here]. Evaporites from overlying strata have impregnated this bed, prior to the date at which silification ended in the basal Purbecks.

A conspicuous relic of evaporites is a porous rock (bed 15) consisting of sparry calcite with scattered euhedral quartz crystals. This is calcitised anhydrite. Net-texture, nodular structure and spherulites of lutecite are obvious, and minute inclusions of anhydrite occur within the quartz crystals. Calcium sulphate was also probably once present at the base ot the overlying Broken Beds. At the adjacent section of Pondfield Cove a similar succession was recognised.

(xii) London Door Quarry, Encombe (SY 945793)

In this section [a very old and small quarry on the Encombe Estate on the road from Encombe House; it may be completely overgrown now], which was in the 1970s poorly exposed, calcitised secondary anhydrite occurs above the Lower Dirt Bed. Limestone with fine lamination, probably of microbial-mat origin, is present near the horizon of the microbial-mat-laminated, Hard Cap of Worbarrow Tout. Brecciation here, however, descends to a very low level and involves the upper part of the Hard Cap. These brecciated beds were once evaporitic and contain calcitised anhydrite. [London Door quarry is close to the edge of the basin-shelf boundary. It is probably just within the basin facies. The low level of the base of the brecciation probably relates to the progressively lower base of the evaporites in an eastward direction , until at Durlston Head they are less than half a metre from the top of the Portland Freestone. Much further east, at Portsdown, nodular anhydrite occurs at the very base of the Purbecks.]

(xiii) Worth Quarry (SY 971787)

The Purbeck overburden of this Portland Stone quarry consists of the Caps and the Broken Beds. Small folds are present like those in the breccia of the Lulworth area and these contorted beds contain evidence of evaporites. A section measured on the east side of the quarry, and not shown in Figs. 2 or 3 is described below. Beds 1 to 5 probably belong to facies A, beds 6 and 7 probably belong to facies B. beds 8 to 10 facies C, whilst the uppermost beds belong to Facies D.

[END OF P. 222]

[START OF P. 223]

Bed

..... Cypris Freestone

13. Laminated pelletoid limestone and micrite with ostracods. 0.3m.
12. Brown shale. 0.46m.
11. Ostracodal pelletoid limestone with scattered ooids. 0.58m.

Broken Beds.
[calcitised evaporites - quite thick -4.4m. Unlike the Lulworth Cove area the breccia is entirely in evaporites and the underlying Great Dirt Bed and Soft Cap with thrombolites of the Fossil Forest is absent, while instead much anhydrite was present. This is Basal Purbeck, evaporite basin facies.]

10. Contorted, very friable, calcitised anhydrite with abundant pseudomorphs after anhydrite. 3.8m.
9. Shale and calcitised anhydrite with net-texture, partly brecciated. 0.24m.

Caps [mostly Hard Cap]

8. Laminated, calcitised secondary anhydrite with ghosts of anhydrite crystals and also of earlier lenticular crystals of gypsum. 0.82m.
7. Laminated marl. 0.26m.
6. Thrombolites of limestone with chert passing laterally into laminated limestone. 0.16m.
5. Carbonaceous calcareous shale (Lower Dirt Bed). 0.09m.
4. Thrombolitic limestone with chert. 0.39m.
3. Calcareous shale. 0.02m.
2. Laminated, sparry calcitised anhydrite with net-texture and 'ghosts’ of anhydrite crystals. 0.08.m.
1. Laminated pelletoid [peloidal] limestone with ostracods, gastropods and foraminifera (overlying and fused to the Portland Shrimp Bed). [i.e. The widespread Transition Bed or Skull Cap.] 0.08m.

(xiv) St Aldhelms Head [or St.Albans Head](SY 963 753)

Here only about 1 m. of basal Purbeck strata is present. Thrombolitic limestone, with moulds probably left by dissolution of a sulphate mineral, is overlain by calcitised anhydrite with authigenie quartz and poorly-developed pseudomorphs after anhydrite.

Between St. Albans Head and the ledge on the cliff top, referred to below, there are many, often inaccessible, exposures of the Caps and Broken Beds in the upper parts of the cliffs. A thin bed of laminated limestone [the Transition Bed] usually overlies the Portland limestones. Some irregular beds of thrombolitic limestone follow and these are succeeded by laminated, probably evaporitic, limestones [calcitised evaporites]. The overlying Broken Beds of this area mostly consist of soft, brown, friable carbonate replacements of contorted anhydrite. At the old Edboro Quarry [or Headborough Quarry] (SY 992768), the Broken Beds, about 5 m. thick, are separated from the Portland limestones by about 1.6 m. of thrombolitic and laminated limestone. There is a similar section at Dancing Ledge (SY 998769).

(xv) West of Anvil Point (SZ 011768)

A ledge on-the cliff top nearly 2 km. west of Anvil Point provides a particularly good exposure. Here, thrombolitic limestones and the Lower Dirt Bed lie above the Portland Formation. Silicified remains of a tree found at this locality have probably come from above this dirt bed [I now suspect that this silicified piece of tree trunk has come from higher in the Purbeck Group, since silicified wood has now been found above the Cinder Bed in Durlston Bay. No other silicified wood has been found in the basal Purbeck strata west of Bacon Hole and the Fossil Forest] . Above the thrombolitic limestones the Caps and Broken Beds consist of soft, poorly lithified carbonate replacements of secondary anhydrite with broken fragments of evaporitic chert. The brec-ciated nature of the Broken Beds is less obvious than at Lulworth.

(xvi) Durlston Head [celestite exposure](SZ 035773)

This section contains beds of celestite (West, 1960) which have replaced nodular anhydrite. They occur in a sequence of coarsely crystalline, calcitised anhydrite, occupying most of the succession between the Portland Formation and ‘Cypris’ Freestones. These limestones contain pseudomorphs after anhydrite, and net-texture and lutecite. Intense contortions and local variations in thickness of the Broken Beds (West, 1960) have probably resulted from movement of the anhydrite northwards due to tectonic forces [associated with faulting] before calcitisation occurred.

[END OF P. 223]

[START OF P. 224]

The celestite, which occurs with calciostrontianite (Salter and West, 1965) , was probably formed by reaction of anhydrite with Sr2+ ions in groundwater (West, 1973). The localisation of the strontium deposits may be due to fault planes facilitating circulation of brines.

[The problem is that the faults have not been dated. Thus the question remains as to whether the faults were Late Cimmerian (sub-Albian) or Tertiary, and whether the calcitisation and celestite development were of Cretaceous or Tertiary age. The fact that the faults are extensional is support for a Late Cimmerian age but it does not prove it.]

(xvii) Isle of Portland

The details of numerous basal Purbeck sections of the Isle or Portland are not considered since it is intended to discuss them in a separate account [Unfortunately, this has not been done, but needs doing. There is much description of Portland quarry sections available but the emphasis is not on evaporites.]. Onh a representative section at Perryfield Quarry (SY 695711) is described here and shown in Fig. 2. Some features of this section have been shown diagrammatically by Pugh (fig. 1 in West, and others, 1969).

As in the Weymouth area (e.g. at Poxwell and Portesham) three main dirt beds [or palaeosols] can be recognised. One is present at the base [Basal Dirt Bed] whilst about 1 - 5 m. above a dirt bed occurs corresponding to the Lower Dirt Bed of the mainland sections. The Great Dirt Bed lies at about 1 -75 m. above this [it contains pebbles, as at the Fossil Forest and it is very easily reconised].

Many moulds left by the decomposition of tree trunks and branches occur in thrombolitic limestone above the Lower Dirt Bed. The Great Dirt Bed contains black and buff-coloured limestone pebbles like those in the same bed at Luiworth. Above the Great Dirt Bed thrombolites surrounds silicified tree remains. Overlying this there is only about 0-3 m. of calcium sulphate replaced by carbonate. As at Holworth House thin evaporites of facies C were originally present at a level corresponding approximately to the base of the Broken Beds. No major brecciation is present, however, on Portland.

[FOLLOWING - a figure explanation for reworking] .

.

Fig. 6. (facing page). Hypothetical palaeogeography of the south-east Dorset area at particular times during formation of the basal Purbeck strata. This is based on data in Figs. 2 and 3 and on the general form of the Purbeck basin (Howitt, 1964).
1 — At the time of deposition of the Lower Dirt Bed.
2 — During deposition of the limestones between the Lower Dirt Bed and the Great Dirt Bed.
3 — During formation of the Great Dirt Bed. 4 — During deposition of the main evaporites above the horizon of the Great Dirt Bed. Figures refer to thicknesses in metres of the main evaporite facies (C).

.

.
Alderman, A.R. and Skinner, H.C.W. 1957. Dolomite sedimentation in the south-east of South Australia. American Journal of Science, vol. 255, pp. 561-567.
.
Allen, P. and Keith, M.L. 1965. Carbon isotope ratios and palaeosalinities of Purbeck-Wealden carbonates. Nature, London. 208, 1278-1280.

Allen, P., Keith, M.L., Tan, F.C. and Deines, P. 1973. Isotopic ratios and Wealden environments. Palaeontology, vol. 16, pp. 607-621.
.
Anderson, F.W. 1973. The Jurassic-Cretaceous transition: The non-marine ostracod faunas. In: R.Casey and P.F. Rawson (Eds.) The Boreal Lower Cretaceous. Seel House Press, Liverpool, Special Issue of the Geological Journal, 5, 101-110.

Anderson, F.W. and Bazley, R.A.B. 1971. The Purbeck Beds of the Weald (England). Bulletin of the Geological Survey, U.K. 34, 1-173.
.
Arkell, W.J. 1938. Three tectonic problems of the Lulworth district: studies on the middle limb of the Purbeck Fold. Quarterly Journal of the Geological Society, vol. 94, pp 1-54.

Arkell, W.J. 1947. The Geology of the Country around Weymouth, Swanage, Corfe and Lulworth. Memoir of the Geological Survey of Great Britain. 386pp. With Wright, C.W.and Melville, R.V. 2nd edition - 1952 with Addenda and Corrigenda.
.
Barker, D., Brown, C.E., Bugg, S.C. and Costin, J. 1975. Ostracods, land plants and charales of the basal Purbeck Beds of Portesham Quarry. Palaeontology, 18, 419-436.
A thin cherty layer near the base of the Purbeck Beds in Dorset has yielded a flora of land plants (including stems of Equisetum mobergii, isolated seeds referred to Carpolithes rubeola, C. glans, C. rhabdotus, C. cocos, C. gibbus, C. acinus and C. westi spp. nov., and cones of Araucarites sizerae sp. nov.) and freshwater Charales (Clavator westi sp. nov.) as well as freshwater ostracods. The discovery of Clavator westi in Lower Purbeck deposits opens up the possibility of separating the Lower Purbeck from the Middle Purbeck using the Charales. Among the ostracods Cypridea dunkeri Jones is shown to have stratigraphic value. Correlation of these basal Purbeck Beds in Dorset with the Swindon Series further north indicates that at Swindon either Purbeck conditions were established earlier than in Dorset, or that Portland conditions recurred later.
.
Brown, L.S. 1931. Cap-rock Petrography. Bulletin of the American Association of Petroleum Geologists, vol. 15, pp. 502-522.,

Brown , P.R. 1963. Algal limestones and associated sediments in the basal Purbecks of Dorset. Geological Magazine, vol. 100, pp. 565-573.

Brown, P.R. 1964. Petrography and origin of some upper Jurassic beds from Dorset, England. Journal of Sedimentary Petrology, 34, 254-269.
.
Butler, G.P. 1969. Modern evaporite deposition and geochemistry of coexisting brines, the Sabkha, Trucial Coast, Arabian Gulf. Journal of Sedimentary Petrology, vol. 39, pp. 70-89.
.
Carozzi, A. 1948. Etude Stratigraphique et Micrographique du Purbeckien du Jura Suisse. Faculte des Sciences, Universite de Geneve, 175 pp. Stratigraphic and Micrographic Study of the Purbeckian of the Swiss Jura. Thesis presented at the Faculty of Sciences at the University of Geneva for the degree of Doctor of Science in Geology and Mineralogy. Thesis No. 1122.
.

Clements , R.G. 1967. Some notes on the Purbeck Beds. Proceedings of the Dorset Natural History and Archaeological Society, 88, 43-44.

Clements, R.G. l969. Contribution to section on the Purbeck Beds : In : Guide for Dorset and South Somerset. In Torrens,H.S. (Editor) International Field Symposium on the British Jurassic. Keele University. 71 pp.

Clements, R.G. 1993. Type-section of the Purbeck Limestone Group, Durlston Bay, Swanage, Dorset. Proceedings Dorset Natural History and Archaeological Society, 114 for 1992, 181-206. (a reprint of the 1969 paper, and a classic Purbeck log)
.
Drummond, P.V.C. 1970. The Mid-Dorset Swell. Evidence of Albian-Cenomanian movements in Wessex. Proceedings of the Geologists' Association, vol. 81, pp. 679-714.
.
Evamy, B.D. 1973. The precipitation of aragonite and its alteration to calcite on the Trucial Coast of the Persian Gulf. In The Persian Gulf (ed. by B.H. Purser), New York, pp. 329-342.
.
Evans, G. 1970. Coastal and nearshore sedimentation: a comparison of clastic and carbonate deposition. Proceedings of the Geologists' Association, vol. 81, pp. 493-508.

Evans, G. and Bush, P. 1969. Some oceanographical and sedimentological observations on a Persian Gulf lagoon. Mem. Simp. Intern. Lagunas Costeras, UNAM-UNESCO, Mexico, 155-170.
.
Fisher, O. 1856. On the Purbeck Strata of Dorsetshire. Transactions of the Cambridge Philosophical Society, vol. 9, pp. 555-581.
.
Fisk, N.H. 1959. Padre Island and the Laguna Madre flats, coastal South Texas. Second Coastal Geography Conference, Louisiana State University, pp. 103-152.
.
Folk, R.L. and Pittman, J.S. 1971. Length-slow chalcedony: a new testament for vanished evaporites. Journal of Sedimentary Petrology, vol. 41, pp. 1045-1058.
.
Friedman, G.M., Amiel, A.J., Braun, M. and Miller, S.D. 1973. Generation of carbonate particles and laminites in algal-mats [microbial mats] - example from sea-marginal hypersaline pool, Gulf of Aqaba, Red Sea. Bulletin of the American Association of Petroleum Geologists, vol. 57, pp. 541-557.
.
Garrett, P. 1970. Phanerozoic stromatolites: noncompetitive ecological restriction by grazing and burrowing animals. Science, vol. 169, pp. 171-173.
.
Grabau, A.W. 1920. Principles of Salt Deposition. Geology of the Non-Metallic Mineral Deposits. New York.
.
Green, R. 1961. Palaeoclimatic Significance of Evaporites. In: Descriptive Palaeoclimatology , (ed. A.E.M. Nairn), London, 61-88.
.
Griffith, J. and Patterson, C. 1963. The structure and relationship of the Jurassic fish Ichthyokentema purbeckensis. Bulletin of British Museum Natural History, vol. 8, pp. 1-43.
.
Groves, J. and Bullock-Webster. 1924. British Charophyta. Book. Ray Society, London.
.
Holliday , D.W. 1970. The petrology of secondary gypsum rocks: a review. Journal of Sedimentary Petrology, 40, 734-744. [Purbeck gypsum]

Holliday, D.W. 1973. Early diagenesis in nodular anhydrite rocks. Transactions of the Institute of Mining and Metallurgy, vol. 82, pp. B81-84.

Holliday, D.W. and Shephard-Thorn, E.R. 1974. Basal Purbeck Evaporites of the Fairlight Borehole, Sussex. Institute of Geological Sciences [now British Geological Survey], Natural Environment Research Council. Her Majesty's Stationery Office. Report No. 74/4, 14pp. with fold-out graphic log of the basal Purbeck evaporites of the Fairlight Borehole.
.
Hollingworth , S.E. 1938. The Purbeck Broken Beds. Geological Magazine, 75, 330-332.
.
House, M.R. 1968. Purbeckian calcareous algae. In Dorset Natural History Reports - 1967 - Geology. Proceedings of Dorset Natural History and Archaeological Society for 1967. vol. 89, p.41-45.
.
Howitt, F. 1964. Stratigraphy and Structure of the Purbeck Inliers of Sussex (England). Quarterly Journal of the Geological Society, London, vol. 120, pp. 77-113.
.
Hughes-Clarke, M.W. and Keij, A.J. 1973. Organisms as producers of carbonate sediments and indicators of environments in the southern Persian Gulf. In the Persian Gulf (ed. by B.H. Purser), pp. 33-56.
.
Illing, L.V.A., Wells, A.J., and Taylor, J.C.M. 1965. Penecontemporaneous dolomite in the Persian Gulf. In: Dolomitisation and Limestone Diagenesis (eds. L.C. Pray and R.C. Murray). Society of Economic Paleontologists and Mineralogists, Special Publication, No. 13, Tulsa, pp. 89-111.
.
Lees, G.M. and Cox, P.T. 1937. The geological basis of the search for oil in Great Britain. Quarterly Journal of the Geological Society. vol 93, pp. 156-194.
.
Linton, D.L. 1951. Map to show existing natural vegetation. In The Oxford Atlas (eds. C. Lewis and J. D. Campbell) London, , viii, ix.
.
Logan, B.W. 1961. Cryptozoon and associate stromatolites from the Recent, Shark Bay, Western Australia. Journal of Geology, vol. 69, pp. 517-533.

Logan, B.W. and Cebulski, D.E. 1970. Sedimentary environment of Shark Bay, Western Australia. Pp. 1-37 in: Carbonate Sedimentation and Environment, Shark Bay, Western Australia (eds. B.W. Logan, G.R. Davies, J.F. Read and D.E. Cebulski). American Associatin of Petroleum Geology, Memoir 13, Tulsa.
.
Loreau, J.P. and Purser, B.H. 1973. Distribution and Ultrastructure of Holocene Ooids in the Persian Gulf. In the book - The Persian Gulf (ed. B.H. Purser), New York, pp. 279-328.
.
Lotze, N.F. 1964. The distribution of evaporites in time and space. In Problems of Palaeoclimatology (ed. A.E.M. Nairn) London.
.
Masson, P.H. 1955. An occurrence of gypsum in south-west Texas. Journal of Sedimentary Petrology, vol. 25, pp. 72-77.
.
Peterson, M.N.A. and Von der Borch, C.C. 1965. Chert: modern inorganic deposition in a carbonate-precipitating locality. Science, vol. 149, pp. 1501-3.
.
Pugh , M.E. 1969. Algae from the Lower Purbeck limestones of Dorset. Proceedings of the Geologists' Association, London for 1968, 79, 513-523.
.
Rusnak, G.A. 1960. Sediments of the Laguna Madre, Texas. In Recent Sediments Northwest Gulf of Mexico (eds Shepard, F.P, Phleger, F.B and van Andel, T. H.) American Association of Petroleum Geologists, Tulsa, pp. 153-196.
.
Salter, D.L. and West, I.M. 1965. Calciostrontianite in the basal Purbeck Beds of Durlston Head, Dorset. Mineralogical Magazine, vol. 35, pp. 146-150.
.
Shearman, D.J. 1966. Origin of marine evaporites by diagenesis: Institute of Mining and Metallurgy, Transactions, 75, B 717, 208-215. [Purbeck evaporites of the Warlingham Borehole]
.
Strahan , A. 1898. The Geology of the Isle of Purbeck and Weymouth. Memoirs of the Geological Survey, England and Wales. 278pp.

Strahan, A. 1901. An abnormal section of Chloritic Marl at Mupe Bay, Dorset. Geological Magazine, vol. 38, pp. 319-321.
.
Taitt, A.H. and Kent, P.E. 1958. Deep Boreholes at Portsdown (Hants) and Henfield (Sussex). British Petroleum Company Limited, London.
.
Ward, W.C., Folk, R.L. and Wilson, J.L. 1970. Blackening of eolianite and caliche adjacent to saline lakes, Isla Mujeres, Quintana Roo, Mexico. Journal of Sedimentary Petrology, vol. 40, 548-555.
.
West , I.M. 1960. On the occurrence of celestine on the Caps and Broken Beds at Durlston Head, Dorset. Proceedings of the Geologists' Association, 71, 391-401.

West, I.M. 1961. Lower Purbeck Beds of Swindon facies in Dorset. Nature, London, 190, p.526.

West, I.M. 1964. Evaporite diagenesis in the Lower Purbeck Beds of Dorset. Proceedings of Yorkshire Geological Society, 34, 315-330.

West, I.M. 1965. Macrocell structure and enterolithic veins in British Purbeck gypsum and anhydrite. Proceedings of the Yorkshire Geological Society, 35, 47-58.

West, I.M. 1973. Vanished evaporites - significance of strontium minerals. Journal of Sedimentary Petrology, 43, 278-279.

West, I.M., Brandon, A. and Smith, M. 1968. A tidal flat evaporitic facies in the Visean of Ireland. Journal of Sedimentary Petrography, vol. 38, pp. 1079-1093.

West, I.M., Shearman, D.J. and Pugh, M.E. 1969. Whitsun Field Meeting in the Weymouth Area, 1966. Proceedings of the Geologists' Association [London], vol. 80, pp. 331-340.
.
Woodward, H.B. 1895. The Middle and Upper Oolitic Rocks of England (Yorkshire excepted). Memoirs of the Geological Survey of Great Britain, UK.

.

Paper received by the Geologists' Association - 20th February, 1973.
Revised Version received 12th December 1974.

Online Version, with minor revisions, 28th August 2012. Dr. Ian West. ianmichaelwest@gmail.com.

.