The Fossil Forest, west of Lulworth Cove, Dorset, southern England, is a classic geological locality. It consists of a series of small embayments about half way up the cliff where a weak bed of evaporites or replaced evaporites crops out. The evaporitic beds and and overlying tectonic-evaporite breccia, the Broken Beds are eroded away on rare occasions by the action of great storms washing seawater up the cliff. The water returns to the sea with some debris through small run off channels, one of which is present in each small embayment.

The washing away of the evaporites and breccia reveals the remains of Late Jurassic trees and thrombolite mounds surrounding them. The trees are maily of an an ancient Cypress type (Protocupressinoxylon). They are rooted in a calcareous palaeosol (ancient soil), the Great Dirt Bed. Above the trees is are microbial mounds of thrombolitic ("stromatolitic") limestone. Above this the unusual Broken Beds, a limestone breccia that was originally evaporitic (a cargneule). This is the thin margin of the Purbeck Anhydrite under much of southern and south-eastern England and well-known to the oil industry as a seismic reflector. This marginal facies of the Purbeck evaporites is well-seen at the Fossil Forest and the adjacent Potters Hole.

.

Please go to separate webpage at:

PURBECK FOSSIL TREES AT THE FOSSIL FOREST AND ON THE ISLE OF PORTLAND.

(This webpage continues with discussion of the strata present at the Fossil Forest Ledge.)

THE LOWER SOFT CAP (THROMBOLITE BED):
(Limestone above the Great Dirt Bed of the Upper Fossil Forest)

Modern Analogues - General

Present-day environments with much resemblance to those of the basal Purbeck occur mainly in Australia. There are major similarities with both the Mediterranean type environments of South Australia, and semi-arid environments of Western Australia near Perth.

THE LOWER SOFT CAP

Modern Analogue:

The Coorong Lagoon, South Australia

An environment with some similarities in terms of semi-arid climate, latitude (although in wrong hemisphere) and presence of evaporites to that of the Purbeck fossil forest environment is that of the coastal Coorong area of South Australia, shown here. Trees grow and die at the margins of hypersaline lakes in this region. These lakes may largely dry out after summer drought. The slightly pink colour results from halobacteria or algae.

This association of trees and hypersaline lakes is not a common one at the present day, mostly because human activities have destroyed the trees in such areas. The Purbeck climate (at about 37 degrees north and clearly very seasonal) was probably Mediterranean but the present Mediterranean area is particularly poor for trees because of deforestation and the grazing of goats. South Australia, with its Mediterranean climate, provides better analogues for various Purbeck features (charophyte lakes, chert, dolomite etc.). Differences should also be noted. The grass and angiosperm shrubs have replaced some low Purbeck vegetation (ferns, horsetails and cycadophytes?). Note that the Coorong area of lagoons and lakes is separated from the sea by a long barrier beach. It is not suggested that the Purbeck environment was associated with a long barrier beach type. Such a beach has not been found and the whole Purbeck lagoonal area was not narrow and was very large (about 200 km across). Nevertheless, the Coorong and other South Australia areas are useful guides to the Purbeck palaeoenvironments.

Reconstructions -
Left: The very early stage of submergence of the Purbeck trees in saline lagoon water is shown here. In the summer this brine became sufficiently hypersaline at times to precipitate a little gypsum and, with time, thrombolites developed around the tree stumps. Notice that there is a pinkish colour to the water because of the presence of halobacteria and algae. The real topography would have been more subdued than this and the surrounding environments were probably forested rather than bushy. Local swells or low ridges did exist in Dorset, however, and the area over the Wytch Farm oilfield was one of these (raised in the Late Kimmerian Movements).

Right: At times the water of the shallow lagoon will have almost certainly dried up and the trees would have been encrusted in salt. The form of the thrombolites shows that the water depth was often less than a metre so this was very likely from time to time in summer. Salt has not been preserved at this level but evidence of calcium sulphate is common. Halite is so soluble that it may be ephemeral and in this semi-arid environment it would have been readily lost in the winter rains. The gypsum and probably halite-saturated conditions of the trees accounts not only for their general preservation but also for the great rarity of holes made by boring organisms within them (one example has been seen in a cycadophyte but they are normally absent in the cypress-type wood). The low vegetation would have been of ferns and horsetails and not grass which had not evolved at this time. In reality many more trees would probably have been visible in the distance, including scattered cycadophytes.

Notice from this reconstruction just how thrombolitic carbonate could accumulate beneath as well as above a near horizontal trunk. The fallen section on the left is lying above the floor of the lagoon. If the saline water was to rise and microbial ('algal') growth take place it could grow beneath and above this particular trunk. This type of growth has clearly happened in the basal Purbeck environment, and it also explains the "tree-holes", nearly horizontal holes left by tree branches, and found in the caps on the Isle of Portland.

THE LOWER SOFT CAP (THROMBOLITE BED):

Thrombolites and Stromatolites of the Soft Cap - Introduction

Both thrombolites and stromatolites are carbonate development that form in shallow hypersaline water. Stromatolites are clearly laminated, microbial ("algal") mats. Throbolites are lumpy developments of carbonate of microbial origin in hypersaline saline water. They occur at the present day and in fossil condition in ancient salt lake deposits. [Incidently similar fossil thrombolites seem, from photographs, to be present on the planet Mars, although the circular features seen, might perhaps be of some inorganic origin, and thus the matter is not confirmed]. Thrombolites are well developed around late Jurassic, fossil tree remains on the Isle of Portland and at the Fossil Forest on the Isle of Purbeck.

.

.

The well-known microbial mounds at the Fossil Forest or on the Isle of Portland are often casually referred to as "stromatolites", and sometimes accidently here, but actually they are not well-laminated, and thus strictly are thrombolites. True stromatolites do occur at the Fossil Forest; they are well-laminated and mostly occur in about the middle of the Soft Cap.

Hypersaline lagoon effectively "pickled"; the remains of trees, which had by now lost their bark. The stumps and fallen logs were then coated in cyanobacterial limestone in conditions rather like (but not identical with) those of Shark Bay, Western Australia. The mounds shown here cover two stumps and a fallen trunk. The internal structure is not well-laminated as in a true stromatolite, but is thrombolitic (lumpy and irregular). Thus, strictly speaking, most of the Purbeck examples of microbial mounds are thrombolites. They are sometimes referred to casually as "stromatolites". Some true, laminated stromatolites are also present in the Soft Cap of the Fossil Forest. These are smaller and they do not coat trees.

I am very grateful to Alan Holiday for pointing out that the Purbeck cylindrical thrombolites are very similar in appearance to the thrombolites of Lake Thetis in Western Australia (Grey et al., 1990). It is interesting that these have been modelled mathematically using the Interface Evolution Equation (KPZ equation). The Purbeck thromobolites had the complicating factor of tree boles as nuclei.

For more information see: Proof of Life.
("An important piece of the origin-of-life-on-Earth puzzle has been discovered in a collaboration between researchers in mathematics and the earth sciences... Murray Batchelor (right), a Professor in the Mathematical Sciences Institute and the Research School of Physical Sciences and Engineering, and Dr Robert Burne, a Visiting Fellow in the Faculty of Science�s Department of Earth and Marine Sciences, have applied the mathematical tools of theoretical physics to show how slimy mats of microbes could have formed the unique conical shapes of certain stromatolites on the seabed over 3,450 million years ago.)

THE LOWER SOFT CAP continued:

Thrombolites - Microbial Domes and Vertical Microbial Cylinders (and Lake Thetis analogues)

Lake Thetis in Western Australia has an environment in which thrombolites like those of the Purbeck Formation develop at the present day. They do not seem, however, to be around tree stumps, as far as I am aware.

Notice the general similarity between the Purbeck thrombolites and the Durness Limestone stromatolites in cross section. Both microbial structures are closely associated with evaporites. Pseudomorphs after gypsum and the lutecite variety of chalcedony are abundant in the basal Purbeck Group with these structures. Similarly the Durness stromatolites occur in strata which contain, at least in parts, relics of evaporites in the form of anhydrite (also present in the basal Purbeck) and lutecite chalcedony (Young, 1977).

Very similar Purbeck dirt beds, fossil trees and thrombolites occur on the Isle of Portland.

LOWER SOFT CAP -

Further West - Durdle Door

For details regarding the equivalent strata to the west of Lulworth Cove go to the Durdle Door Webpage. Here is seen the western limit not of the Soft Cap, as such, but of the forest with trees on the Dorset mainland. Note that this is upper fossil forest, and trees of the lower fossil forest occur commonly north of Weymouth, but these are in the Hard Cap, not in the Soft cap.

At Durdle Door the tree moulds are of small size, in several cases only about 0.3m or less in diameter. This is a less well-developed part of the forest, probably because the Great Dirt Bed was not uplifted so high above lagoon level. Small diameter tree moulds are reminiscent of the lower fossil forest on the Isle of Portland, i.e. that of Hard Cap (above the less-well developed Lower Dirt Bed).

.

.

LOWER SOFT CAP continued:

Microbial Domes or Thrombolites around Trees - Cross Sections

THE LOWER SOFT CAP continued:

Thrombolites - Elongate on Fallen Trees (bedding parallel)

The fallen, almost flat-lying, trees have elongate thrombolites developed around the location of the trunk. The orientation of these is not parallel. Thus the tree trunks do not seem to have been blown down in the same direction in the same storm.

THE LOWER SOFT CAP continued:

Thrombolites - elongate (bedding parallel) in cross section

Notice that horizontal or subhorizontal trunks or branches are often compacted by about a quarter to a half (the two photographs above are of the same specimen at different years). The thrombolite around shows about the same amount of compaction. This extent of compaction is compatible with an overburden of probably not more than 10 or 20m. Thus firm cementation took place quite early.

More common on the Isle of Portland is partial and later filling of tree moulds. This is sometimes seen in Hard Cap examples. The filling is at least in part of thrombolite type. This suggests very early decomposition of the wood and early, partial fill.

THROMBOLITES continued:

Analogous Thrombolites in the Lower Fossil Forest at Portesham Rocket Quarry

Go to:

Portesham Rocket Quarry webpage, to see analogous thrombolites that have formed in the Hard Cap (lower than the Soft Cap thrombolites of the Fossil Forest). A notable example is an elongate thrombolite with rings on the periphery. This was once known as the "Fossil Elephant" and is still a notable feature of Portesham Quarry, north of Weymouth.

Thrombolitic [i.e. not laminated as in the case of stromatolitic] limestone occurs around a mould of a coniferous tree. There is no doubt that this was a tree because Strahan (1898) has recorded silicified wood within it. In addition a silicified trunk has been found nearby at the time of quarrying. Silicified trunks and moulds occur commonly on the Isle of Portland and at the Fossil Forest Lulworth, although they are best developed there above the Great Dirt Bed, and not so well-developed at the Lower Dirt Bed level.

The thrombolitic coating 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.

The basal part of the now-missing fossil tree has left an impression in the thrombolitic ("stromatolitic") limestone. This could be just an impression or rotted timber, but it appears more likely that it is the mould of part of the bark of the tree. If this was a bark mould, then the bark seems to resemble the bark of a swamp Cypress tree, like the modern Taxodium distichum.

Further in the interior of the mould can be examined. This is shown in the photograph above. It can be seen that the tree trunk had well-defined straight grooves (the twisting of the trunk as shown on some reconstructions of Purbeck trees is just artistic licence by the artist). Straight grooves are common on various modern trees, including Cypress trees. Purbeck fossil trees have been described as Cupressinoxlyon ("proto - Cypress wood") by Professor Jane Francis. It is not known whether this particular tree was of Cupressinoxylon type. Araucaria-type debris is known from the Portesham Charophyte chert a little higher in the succession, and it seems likely that both Cypress-type and Araucaria-type trees were present in Purbeck environments. Whether anything resembling the Swamp Cypress existed there is not known. There was a swamp with horsetails (remains in the Portesham Charophyte Chert), and there was a freshwater lake at times. There are still so much unknown about Purbeck environments.

THE LOWER SOFT CAP continued:

Origin of the Thrombolites

The association of the Purbeck thrombolites with traces of evaporites and the general association with evaporitic strata above (the main Purbeck anhydrite bed) suggests that they originated in a hypersaline lake or lagoon. There are no freshwater gastropods associated and the salinity was generally too high for much fauna at all (although there are some worm tubes). See the discussion about facies at the Fossil Forest. No thrombolites or stromatolites are known in the Purbeck Formation in association with the freshwater facies which are common higher in the Formation (Cherty Freshwater Member etc). They do occur in association with another higher sequence of evaporites, the Soft Cockle Gypsum (probably anhydrite underground). These thrombolites are more isolated in an argillaceous sequence. Although the Victorian geologists did not understand the thrombolites, and thought that they were some form of "tufa", since the 1960s they have been known to be of hypersaline origin (see papers of West (1963,1964,1975 etc).

As shown above, the major comparison is with the remarkable thrombolites of Lake Thetis in Western Australia. This is a hypersaline lake environment that provides a superb modern analogue. There is little doubt that is the model for the environment, although the modern example at Lake Thetis is on a small scale compared to the late Jurassic development of the Purbeck Formation, discussed here.

Minor Comparison

Very small stromatolites growing at the present day on an old steel drum that has sunk into a soft sabkha with hypersaline brines at Umm Said, Qatar. A hard or raised surface for attachment seems favourable for stromatolite growth. Similarly the tree stumps submerged in the Purbeck hypersaline lake have provided attachment areas on which stromatolites or thrombolites developed.

(For a charophyte mystery concerning the Soft Cap see the charophyte section of the Purbeck Facies and Palaeoenvironments Webpage.)

STROMATOLITES (true stromatolites - laminated)

Microbial Mat Type of Stromatolites

Although the Fossil Forest Ledge which exposes the basal Purbeck strata has long been known for its microbial mounds, the relationship of thrombolite and true stromatolites has not been clearly expressed. The mounds over trees, the most well-known features have been referred in past years in general terms as "stromatolites" (e.g. West, 1975) because of general, overall morphological similarity to the stromatolites of Shark Bay, Western Australia. In fact, they are generally thrombolites with a lumpy and irregular fabric, with only crude lamination. It is actually above this level in Bed FF 13 of West (1975) that true, well-laminated stromatolites occur. They are smaller than the metre-large microbial mounds over the trees and often only about 30cm or so in diameter. They are best seen at the eastern end of the Fossil Forest Ledge. They often contain lenticular gypsum crystals that are "on edge", i.e. with the plane of flattening vertical and the c-axis near-horizontal. There is other close association with evaporites. The main Basal Purbeck evaporite horizon, the "Broken Beds" with brecciated and calcitised evaporites is just above. This is the fringe, coastal equivalent of the much thick Purbeck Anhydrite of the English Channel Inversion, the Isle of Wight area, and the extensive Weald Basin. It is worked economically for gypsum at Brightling and Mountfield in Sussex. There is no question that the true stromatolites of the basal Purbeck originated in very hypersaline conditions (in West's, 1975, Facies C), even more so that the thrombolites (in West's, 1975, Facies B) with salinities near 124 parts per thousand (at the boundary of gypsum precipitation).

Evaporites - Gypsum, Anhydrite, Flowage and Calcitisation.

Evaporites are very obvious in the upper part of the Soft Cap, particularly in thin-section. Some large gypsum moulds, on-edge in orientation (i.e. with c-axes horizontal) can be seen in a laminated limestone near the barbed wire at the cliff edge at the western limit of the Fossil Forest. The bed can be followed out of the Fossil Forest ledge towards Potters Hole, and further search can be made in the cliffs there. However, note that this is a dangerous area above vertical cliffs of Portland Stone.

Gypsum is readily replaced by calcite and by silica. Halite is rather easily dissolved. Its former abundance here is liable to be underestimated. The best pseudomorphs actually occur in the Cypris Freestone Member, above the Broken Beds. It is very likely though, that the gypsum crystal relics were associated with some halite, as in modern environments, such as northern Egypt and elsewhere. No evidence is known, at present though of any thick beds of halite in the Purbeck Formation, such as occur in the Trias of Devon. Beds of anhydrite have been reported in boreholes, but beds of halite have not. Halite crystals occur at the level of the dinosaur footprints in Lower Purbeck strata of the Isle of Portland. The horizon, the Hard Cockle Member is higher than the Caps and Broken Beds at the Fossil Forest.

The lagoon floor and particularly its western margins here dried out from time to time. The last relics of brine evaporated to produce a white crust of halite crystals. These crystals displaced the underlying carbonate mud. Late in the summer there were phases when the dried-up lagoon flooded again, perhaps as a result of the first major rains or because winds from the east backed up the lagoon waters and blew them over this marginal area. The flooding quickly dissolved the salt and filled the moulds of the crystals with a mixture of carbonate sand and some quartz sand. Thus were formed the pseudomorphs shown in the photograph. Insects and fine plant debris has been washed over flats and is preserved because it settled on the top, the other side, of the sandy lamina. Some pseudomorphs at the Fossil Forest have a fill of faecal pellets rather than quartz and carbonate sand.

(See other sections of this webpage for the diagenetic sequence involving change from gypsum to anhydrite, and with brecciation and calcitisation taking place. This is not fully explained in this part, but is shown in diagenesis diagrams of Ian West.)

With a little imagination, we can to some extent reconstruct this environment. Here is a small branch of the great lagoon beginning to dry out as seen from a hill on the Wytch Farm High (an east-west axis which was uplifted in late Jurassic/early Cretaceous times because of the late Kimmerian movements). The view is southwestward. The main lagoon was actually very large and extended well over the horizon and for more than 200 km. The salt crust dissolves in the winter rains. Some Iguanodons have left trails of footprints across the salt flats.

Comparison, is made again with a modern salt lake at Akrotiri, Cyprus, drying only in summer.

UPPER SOFT CAP

More Details - Thrombolite - Evaporite Relationships

The relationship between the evaporites of the Upper Soft Cap and the thrombolites beneath is of interest. This is a topic which requires further study, and the initial interpretation is not proven, and might not necessarily be correct.

At the Fossil Forest exposure an interesting relationship can be seen between the calcitised evaporites (calcium sulphate, gypsum to anhydrite) and a thrombolite. The evaporites seem to have been distorted around the thrombolite.

BROKEN BEDS - THE CARBONATE-EVAPORITE CARGNEULE (BRECCIA)

General

See also:

Broken Beds at Lulworth Cove.

The "Broken Beds" is an old traditional name for a peculiar limestone breccia which puzzled many geologists before it was studied microscopically and geochemically. It extends eastwards from here to Durlston Bay, Swanage, where it is thickest and westward to Durldle Door where it is very thin.

This breccia has a simple explanation, although it is complicated in detail. It is actually a rauhwacke or cargneule of carbonate-evaporite type. It does not look at first sight like the cargneules of the Pyrenees or Alps but this is because it is not in Permo-Triassic red bed facies. Like them, though, it contains euhedral quartz crystals but they are mostly microscopic and not an obvious field indicator. In places it is strontium-rich and it contains celestite at Durlston Head (with calciostrontianite). Small quantities of celestite occurs beneath it in the Soft Cap at Stair Hole. Celestite is also present in the Caps at Worbarrow Tout.

Much of the lower part of the breccia and the upper part of the underlying Soft Cap consists of clasts and matrix of very porous limestone, often soft and sandy in appearance (but certainly not with quartz sand!). This is calcitised anhydrite, an unusual type of limestone produced when calcium sulphate is changed to calcium carbonate and any remaining sulphate dissolved .

This lower part of the breccia (and the bed immediately beneath) originally consisted of beds of gypsum, deposited in a hypersaline lagoon. These evaporites of calcium sulphate occur in boreholes in the Isle of Wight and southeast England at this horizon as anhydrite (calcium sulphate without water of crystallisation). They come close to the surface near Mountfield and Brightling in the Weald of Sussex where they are changed by groundwater to secondary gypsum (calcium sulphate with water of crystallisation) and are mined to make plaster of Paris and plaster-board. There are special reasons (perhaps you can think of some) why here, at the Fossil Forest the calcium sulphate has been replaced by limestone.

The upper and thicker part of the breccia consists of angular blocks of pelletoidal limestone with numerous small shells of ostracods (the Cypris Freestone facies). The strata have been laterally displaced northward, folded with small overfolds and fragmented into a breccia by tectonic action. The ductile evaporites provided a plane of weakness and the overlying brittle limestone has become involved in the movement. The date of the tectonism could have been Alpine (Tertiary - compressive) or Late Kimmerian (Cretaceous - extensional). Association with extensional faulting in the Isle of Purbeck points towards a Late Kimmerian date.

STRATIGRAPHY:

Broken Beds - Cargneule, Tectonic Evaporite Breccia - Sedimentology

The images here show the evaporitic part of the Broken Beds has calcitised anhydrite both as blocks and as matrix.

Under the microscope here there is abundant lutecite chalcedony, a common feature of evaporites, and there are also pseudomorphs after anhydrite. Some microscopic euhedral quartz occurs.

This anhydrite was a replacement, under burial, of primary gypsum and excellent pseudomorphs of the original lenticular gypsum crystals are preserved in chert nodules which are present in this lower evaporite part of the breccia. In places there are some macroscopically visible moulds left by lenticular crystals of gypsum with the planes of flattening vertical. (These are in the Soft Cap near the outer cliff edge at the extreme eastern end of the Fossil Forest section). A little celestite (in nodules of about a cm across) , which is an insoluble diagenetic product and residue of evaporites, occurs in places in the Soft Cap of the Lulworth region (at Stair Hole - again in a place not recommended for safety reasons!). It is much better to see this where it is well-developed in the Broken Beds at Durlston Head, Swanage.

An interesting aspect of the Broken Beds is that the chert here is, as mentioned above, a replacement of primary gypsum. Further east, near Swanage, it is a replacement of anhydrite. A possible explanation for this is that the chert formed at about the same time during early burial when the calcium sulphate here was still gypsum (West, 1964). Further east anhydritisation probably commenced at this time because the area was away from the shelf (Lulworth area) and into the basin (Swanage area). Such a process has been described in the Miocene of southern Poland (Kasprzyk and Orti, 1997) and has been attributed to the probable occurrence of higher salinity pore fluids in the basin centre.

The evaporites of the Broken Beds (and the Soft Cap beneath) originated in the hypersaline lagoon that "pickled" the dead Purbeck trees before they could rot completely. Silicification of the initially carbonaceous submerged forest was probably largely the result of high pH conditions in lagoon water at times (silica is very soluble at high pH and the tree stumps would have provided locally lower pH conditions suitable for precipitation. The silicified tree remains weather to a pale colour but are actually more carbonaceous and, to some extent, more sulphurous than they look. They can be black when sawn through)

The upper part is a breccia of angular joint-bounded limestone blocks without any substantial relics of evaporites. Notice the absence of matrix in this part. Much of this consists of ripple-laminated ostracodal pelsparites and is similar to the "Cypris" Freestones Member above. The fauna suggests an origin in moderately hypersaline lagoon water, perhaps about twice the salinity of seawater (no accurate estimate is possible because of the summer to winter variation salinity variations in the seasonal Purbeck climate).

The brecciation was not simple collapse over evaporite, although some collapse may have occurred. It is mostly tectonic with small folds which have east-west axes and vergence towards the north. There are two possible dates for the tectonic brecciation at this plane of weakness produced by the evaporites. Tectonism occurred in the mid-Cretaceous (Late Cimmerian or Kimmerian) and in the Tertiary (Alpine). It is has not yet been firmly proven as to which phase was responsible and different authors have had different theories. The the reason for the northward movement of the overlying strata (as shown by the small folds) is also disputable.

Although the Broken Beds are a type of cargneules or rauhwackes, like those frequently associated with thrusting in the Alps and Pyrenees and elsewhere, they are different in certain respects. This particular breccia is not dolomitised and does not have associated red marl.
.
The Broken Beds are monomict cargneules(West, 1964; 1975; Warrak, 1974), at least in their lower part. Calcitisation at this particular locality is probably by meteoric water (Quest, 1985) although hydrocarbons were involved further east. Some collapse may have occurred in addition to tectonism but it seems to have been on a limited scale. Thus the breccia is not referred to as a collapse breccia.

Broken Beds and Soft Cap Evaporite
Petrography - Pseudomorphs after Lenticular Crystals of Gypsum

Early or penecontemporaneous crystals of gypsum are very commonly lenticular in form. Later, diagenetic gypsum is rarely so. The early lenticular gypsum is often replaced and present at lutecite pseudomorphs after gypsum, calcite pseudomorphs after gypsum or celestite pseudomorphs after gypsum. This a major part of the common evidence for "Vanished Evaporites". Such pseudomorphs are abundant in the Lower Purbeck Formation of Dorset, but are not normally visible in the field. Thin-section petrography is needed.

The most notable feature of the early or primary Purbeck calcium sulphate is that it was in the form of lenticular gypsum, as is abundant at the present day in the sabkhas of Northern Egypt (see for example Ali (1981). Pseudomorphs of calcite, chalcedony (particularly lutecite) celestite after such gypsum are very common. In most cases they are too small to see without a microscope so they are ofen misssed by field geologists. Some illustrations are shown above. The pseudomorphs after lenticular gypsum occur at most localities in Dorset where the Lower Purbeck Formation is exposed, and this includes the Fossil Forest Ledge. They are to be found in the lower part of the Broken Beds and in the Soft Cap. They occur again in some abundance around the levels of the Soft Cockle gypsum at Worbarrow Bay and Durlston Bay, higher in the Lower Purbeck. They are rarely found above the Mammal Bed (base of the Middle Purbecks) because of the well-known change to a rather wetter climate.

The Purbeck evaporite are dominantly calcium sulphate and the Stage 1 (of West's (1964) diagenetic scheme was broadly similar to that of the Eocene gypsum of Paris. This French gypsum is notable for lenticular gypsum crystals and they have been very well described in the the classic work of Alred Lacroix (1897) on La Gypse de Paris et Les Mineraux qui l'Accompagnent. It is interesting the French gypsum has survived largely unchanged for about 40 million years. The Purbeck gypsum, about 100 million years older, has undergone diagenesis from primary gypsum to secondary anhydrite, and, at the surface, back to gypsum again but of secondary type (largely anhedral and with negligible porosity.

The sequence of diagenetic stages West's (1964) is fairly obvious (although there differing views on the details of the anhydrite stage). However, it is still not known at what date the Purbeck lenticular gypsum was converted under burial to anhydrite. Late Wealden (about Barremian) would be a reasonable guess.

The crystallography of the primary lenticular gypsum was probably similar to that in the modern deposits. An interpretation by Masson (1955) of the details of the habit and crystallography of Laguna Madre (Texas) lenticular gypsum is given above.

Upper Broken Beds

Brecciation

The brecciation of the Upper Broken Beds (upper part of the Broken Beds without obvious evaporites) is clearly tectonically driven with movement towards the north. There are, as shown in photographs in this webpage, clearly defined overfolds in a northward direction. There is a general lack though of calcite veins on joints and slickensides are not normally seen, in spite of a search for them. However, as shown in the associated photograph, there are impact marks and some short but broad scratches.

STRATIGRAPHY:

Cypris Freestones Member

The 'Cypris' Freestone Member consists of very laminated buff, blocky limestone above the breccia of the Broken Beds. They are thus above the ledge of the Fossil Forest and in the back cliff, but not far up and quite accessible. Not all of these limestone are rich in ostracods, but some parts are. Look with a handlens for the small, bean-shaped ostracods (crustaceans).

In more technical terms this facies has been discussed by West (1975). The following notes are partly from this

Facies D - Limestones with Ostracods - 'Cypris' Freestones

The name has originated because these limestones are full of ostracods, referred to in early Victorian days as 'Cypris'. In fact they are very featuresless ostracods and difficult to identify. An important aspect is the low salinity ostracods, such as Cypridea (with beaks) and its relatives are generally absent in the 'Cypris' Freestones, because of the prevailing high palaeosalinity. The name "Freestone" applies because at some localities, mainly Upwey, some of this limestone has proved hard enough for building purposes, and can be cut in various directions (i.e. free-stone).

The facies D, discussed here largely corresponds to the the 'Cypris' Freestones, but not exactly. The upper part of the "Broken Beds" is of this facies, but the "Broken Beds" have been defined by the brecciation (tectonic and evaporite), not by the facies.

This facies consists of ripple-laminated limestones of pelletoids and ostracods. It contrasts strikingly with the underlying facies in its abundance of ostracods and relative lack of replaced gypsum. Normal marine faunas, such as those containing oysters and Corbula etc. do not occur, and neither (normally) do freshwater faunas with pondsnails such as Viviparus, Physa etc. The molluscs present are small euryhaline gastropds, Hydrobia and Valvata; these are not abundant.

The dominant rock types are ostracodal biosparites, intrasparites and pelletoidal limestones. Calcareous shales occur and in the western part of the Dorset outcrop there are some quiet-water oolite (i.e. lagoonal). The most common evidence of evaporites consists of occasional casts of small halite crystals, although these have been noted particularly near the base of the Hard Cockle Member, next above.

STRATIGRAPHY:

Hard Cockle Member

BROKEN BEDS continued:

Fish Remains in the Broken Beds

A small fish, Ichthyokentema purbeckensis, of a generally marine type but possibly adapted to hypersaline conditions has been found in the upper (moderately hypersaline) part of the Broken Beds (Griffith and Patterson, 1963) at the Fossil Forest locality and elsewhere. It seems to occur in the the hypersaline facies of the upper Broken Beds of West (1975).

This fish was only 6 to 7 cm in length and related to the marine fish genus Pholidophorus. The bones were found to be very abundant, several hundred specimens having been collected. Specific information on the horizon has been given by Griffith and Patterson: " At Lulworth there is a 0.15m (6 inches) band of yellow-brown, sandy limestone, darker in colour than the surrounding purer limestone, lying at about 1.91m (6 feet, 3 inches) below the top of the Broken Beds. 0.20m (8 inches) below this band there is a similar band, 3 to 5 inches thick. The fish were found in the 1.06m (3 feet, 6 inches) of limestone directly below this band. The specimens were collected from a single large block, about 3.05m (10 feet) long and 1.22m (4 feet) thick, lying immediately above the "fossil forest" where it is exposed in the cliff a quarter of a mile east of the mouth of Lulworth Cove. The bones were distributed through 1.06m (3 feet, 6 inches) of sediment and were common everywhere, though more abundant in and near the clay partings which occur irregularly in the rock. The fish remains are all disarticulated, only the neurocrania, mandibles and pectoral girdles being preserved reasonably intact. Thought the bones lie scattered throughout the rock, they are perfectly preserved, and show no sign of rolling or abrasion. There are some signs that the bones have been water sorted, for while no specimen shows scattered remains of a single skeleton, it is common to find two or three nerurocrania lying close together with no other bones in their vicinity. The vast majority of the bones are of Ichthyokentema purbeckensis: we have parts of at least 50 neurocrania of this species. A few fragments of other fish are present.. Associated with the fish are fragmentary plant remains and abundant ostracods of the two species "Cypris" purbeckensis Forbes and Fabanella bononiensis (Jones) (equals Candona bononiensis), with the former more abundant.

These particular "griesly fish" did not, in fact, swim through the tree branches (fish remains are rare in the Soft Cap) but lived a little later in the moderately hypersaline lagoon. And unlike those of the Bishop of Dunkeld flooded forest they did not of course, shriek and yell so as to "fordeafit the heiring with their schouts"!

Cherty Freshwater Limestone Blocks

Fallen blocks of white micrite are conspicuous features of the Fossil Forest Ledge, especially in the central part. These are quite distinctive because they are not laminated but they do contain chert. They contain freshwater gastropods and sometimes charophyte remains. These blocks are from the Flint Bed of the Cherty Freshwater Limestone Member of the Middle Purbeck Formation.

They are interesting because the fauna and flora of the beds is of a very low salinity type. However, they usually contain moulds of halite crystals and, in some cases, silica pseudomorphs after halite. These are the relics of salt crystals that developed in the sediment soon after sedimentation. They show that the early interstitial brine contained NaCl in solution in spite of the "freshwater" conditions implied by the fauna.

LOCATION:

Mupe Bay

If time permits you can continue your field trip to Mupe Bay and Bacon Hole which is in the Ranges, not far to the east. Here you can see ammonites in the Portland Stone, a good sequence through the Purbeck Formation and oil sand in the Wealden.

ACKNOWLEDGEMENTS

I am particularly grateful to Professor Jane Francis for discussion on the Purbeck Formation. I much appreciate the comments of many visitors to the locality and the opportunity to consider with them various details of the Purbeck succession. It was very helpful to discuss evaporites with the late Professor Douglas Shearman and the late Professor Robin Bathurst. I much appreciate the help of Alan Holiday in drawing attention to the Lake Thetis analogue of the Purbeck thrombolites. Paul Ensom has kindly drawn attention to his discoveries of fossil trees at other Purbeck horizons. I very much appreciate the help of the Portland Stone expert, Dr. Geoff Townson, with regard to Shark Bay photographs and other matters. Regarding other sources, the submerged forest illustration is modified from Davey and Morcombe (1972) and the night reconstruction is partly after Dutton (1980). Dr. David Godfrey kindly provieded the Coorong photograph.

Bibliographies

See the Geological Bibliography of Lulworth Cove.

See also the Geological Bibliography of the Purbeck Formation.

PURBECK FOSSIL FOREST WEBSITES, PART 1 AND PART 2.

This is the end of:

THE FOSSIL FOREST: PART 1: GEOLOGY OF THE LEDGE.

To continue with Purbeck Fossil Trees - go to:

THE FOSSIL FOREST: PART 2: PURBECK FOSSIL TREES.

.

ACKNOWLEDGEMENTS

I am very grateful to the Dean of the Faculty of Natural and Environmental Science, to the Head of the School and the Staff of the School of Ocean and Earth Science for supporting the running of this website. I thank iSolutions for runnng the website from their server. The Channel Coastal Observatory, and in particular Dr. Travis Mason has been very helpful with regard to aerial photographs. I much appreciate the helpful discussion with various people on field trips in the past back to the late 1950s. I think Dr. Geoff Townson for kindly providing photographs of the Shark Bay stromatolites. I very much appreciate discussion in the field with the specialist on microbial mounds, Professor Dan Bosence. I particularly appreciate the boost to my career that was given by the late Professor Bathurst in the 1960s, as a result of a sedimentological field trip that I was leading here. My wife and family have provided support and assistance in various ways. I particularly thank Alan Holiday for having realised the similarity between the Purbeck thrombolites and the thrombolites of Lake Thetis. Alan has also noticed the comparison with Durness Limestone stromatolites, photographed them and kindly sent me a copy with is used in an illustration above. I thank Richard Cox for an excellent photograph of a thrombolite around a hollow mould of a coniferous tree at Portland Bill.

Copyright © 2016 Ian West, Catherine West, Tonya Loades and Joanna Bentley. All rights reserved. This is a purely academic website and images and text may not be copied for publication or for use on other webpages or for any commercial activity. A reasonable number of images and text may be used for non-commercial academic purposes, including field trip handouts, lectures, student projects, dissertations etc, providing source is acknowledged.

Disclaimer: Geological fieldwork involves some level of risk, which can be reduced by knowledge, experience and appropriate safety precautions. Persons undertaking field work should assess the risk, as far as possible, in accordance with weather, conditions on the day and the type of persons involved. In providing field guides on the Internet no person is advised here to undertake geological field work in any way that might involve them in unreasonable risk from cliffs, ledges, rocks, sea or other causes. Not all places need be visited and the descriptions and photographs here can be used as an alternative to visiting. Individuals and leaders should take appropriate safety precautions, and in bad conditions be prepared to cancel part or all of the field trip if necessary. Permission should be sought for entry into private land and no damage should take place. Attention should be paid to weather warnings, local warnings and danger signs. No liability for death, injury, damage to, or loss of property in connection with a field trip is accepted by providing these websites of geological information. Discussion of geological and geomorphological features, coast erosion, coastal retreat, storm surges etc are given here for academic and educational purposes only. They are not intended for assessment of risk to property or to life. No liability is accepted if this website is used beyond its academic purposes in attempting to determine measures of risk to life or property.

Webpage - written and produced by: