Lithostratigraphy and deposition of the type Portlandian
WILLIAM GEOFFREY TOWNSON
SUMMARY The "Portland Beds" of Dorset (Portlandian environment. The dolomite formed by in situ of English usage) are described in terms of a replacement of limelime mud. The middle cycle Group comprising two Formations and seven consists of cherty fine-grained limestones Members. Facies and thickness variations deposited on the outer part of a carbonate indicate the presence of a swell separating an shelf. The abundance of replaced sponge East from a West Basin. The swell may be due spicules adequately accounts for the amount to the movement of Triassic salt. The environenviron- of chert. The upper cycle consists of cherty mental history of the Portland Group is limestones passing up into shallow-water described in terms of three cycles consisting of grainsgrainstones. tones. Ooid shoals developed over the major regressive and minor transgressive swell. These marine limestones are overlain by phases superimposed on an overall regression. stromatolites and evaporites which formed on The lower cycle consists of siliciclastics and the basin margin. dolomite deposited in a relatively deep marine
OVERLYINGOVERLYINO THE UPPER JURASSIC KIMMERIDGE CLAY in southern Eng-Eng land and northern France is a sequence of strata known as the "Portland Beds" (for references from 1816 to 1936I936 see Arkell 1935,I935, 1947). The Dorset outcrop provides the geographical name of Portland and in this area the Portlandian of English geologists comprises three ammonite zones (Cope in Torrens 1969) correlated with the middle Volgian (Casey 1967, 1973)I973) and the upper Tithonian (Zeiss 1968):I968) : PurbeckPur beck Beds non-marine Titanites giganteus Portland Beds Glaucolithites gorei Portlandian Progalbanites albani Topmost Kimmeridge "Epipallasiceras"EpipaUasiceras sp." late Clay Pavlovia rotunda Kimmeridgian In the past the "Portland Beds" have been studied mainly for their faunal content (Cox 1925,I925, 1929i929 and references in Arkell 1935)i935) and are still poorly defined lithostratigraphically. The detailed bed measurements of 15 sections by Arkell (i935)(1935) are very useful but his rock descriptions are less satisfactory. A lithostratigraphy is presented here based on study of 233 vertical sections, io721072 hand specimens and 68o680 stained thin sections (Townson I97I1971). ). A detailed lithofaciesli thofacies correlation of all the exposures has been made (Table I, Figs. i-3).1-3). Nomenclature The "Portland Sand" (Table x)I) is redefined as the Portland Sand Formation, divided into four Members. The base is lowered to a mappable change from
Jl geol. Soc. vol. x3x131,, I975,1975, pp. 6x9-63861g-638, , 5 figs. PrintedPrinted. in Northern Ireland.Ireland.
Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021
620 W.W. G. Townson
TTA ABLE ]3 L E I:I" Lithostratigraphy of thethe Portland Group of Dorset. Arrows indicate regressive sequences.$e( vuences .
i 19351955 ARKELL 1947 ARKELL 1956 THIS PAPER ~Ouj~ PURBECK Upper 22m Durlston Formation PURPURBECK BECK :E~EZ Z BEDS Middle 47m =EO BEDS Lulworth Formation GROUP
Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 The type Portlandian
argillaceous sands and silts to shale, instead of the present j unction within a sandy or silty sequence (Arkell 1947). The top of the Formation is not changed but is in fact the top of a widespread thick dolomite and not a sandstone (Townson 197 I). The "Portland Stone" (Table I) is redefined as the Portland Limestone Formation, divided into three Members. The base is taken in east Dorset as the junction between limestone and underlying dolomite and in the west between clay and underlying dolomite. The only change is to include the "Portland Clay" (Damon 1884, Strahan 1898, Arkell 1947). The top of the Formation is not moved. The term Portland Group replaces "Portland Beds" in their enlarged and lithostratigraphically defined and correlatable form. The repetition of "Portland" is justified on historical precedence and the absence of suitable geographical names which are not already used for the Dorset J urassic. The "Purbeck Beds", comprising the Lulworth and Durlston Beds (Casey 1963), are renamed Purheck Group, Lulworth Formation and Durlston Formation (Table I). Depositional setting Thicknesses vary across Dorset as follows (see Fig. 2 for localities): Portesham Ringstead Gad Cliff & Hounstout Isle of Area Bay Kimmeridge Area Portland Lulworth Formation 35 m 30 m 45 m 55 m 35 m Portland Group 65 m 40 m 90 m I 10 m 65 m Kimmeridge Clay 350 m 260 m 500 m ? 400 m A swell divided the area into East and West Basins. This swell plunged south wards as seen by the increased thickness in the Isle of Portland. Depositional thinning of the upper J urassic and lower Cretaceous across this swell is described by Arkell (1938, and earlier workers referred to therein) but no trend direction or causal explanation is given. Drummond (1970) describes the 'Mid-Dorset Swell' and its control on sedimentation during Albian to Cenomanian time (after a major tectonic phase). This feature, however, trends NW.-SE., at right angles to the direction of the late Jurassic swell (Fig. 5). The SW.-NE. orientated palaeo graphic maps of West (1975) for the basal beds of the Lulworth Formation follow on very satisfactorily from those for the Portland Group in Townson (197 I) . Lees and Cox (1937) state that the tectonic style of south Dorset is halo kinetic and it is here suggested that subsidence in the area was influenced by the presence of a Triassic salt pillow below the swell. Portland Sand Formation
(A) BLACK NORE MEMBER The name is that used by Arkell (Table I) from the section on the Isle of Portland, now spelt Blacknor (grid ref. SY 677 I).
Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021
622 w.IV. G. Townson
In the East Basin only two localities expose the complete succession: Hounstout-Hounstout St. Albans Head and Gad Cliff (Fig. 2). Four informal divisions exist (Table 1):I) : (i) Rhynchonella "Marls". Black, dolomitic, laminated argillaceous siltstones. (ii) Hounstout "Clay". Dark grey dolomitic silts and very fine sand with thin shale interbeds. (iii) Hounstout "Marls" and (iv) Emmit Hill Beds (modified term, see Table 1).I). The latter two consist of bioturbated calcareous dolomitic clay, silt and fine sandstone with thin beds or nodule layers of lime mudstone (details in Townson 1971).197 I). In the West Basin the junction with the Kimmeridge Clay is known only in a stream section near Coryates (SY 629856) 629856 ) and in a temporary exposure (SY 690833)69o833 ) over Tout Downs (Samuel 1969,I969, Townson 1974).I974). In both sections clay passes up into a mixture of dolomite and sand, but the western exposure contains up to 100/0 lO% glauconite (Townson 1971).I97i ). On the Isle of Portland the Member is best exposed at Arkell's type locality. Dolomitic silty clay passes up into dark grey argillaceous dolomite with thin nodular horizons oflightof light grey lime mudstone.
(B) CORTON HILL MEMBER The name is taken from the exposure in the cutting to Corton Farm, over Corton Hill. Thickness and facies changes are shown in Fig. I. In the East Basin the sediment is dominantly a mixture of clay, dolomite and fine quartz sand, with thin horizons of Exogyra. Towards the swell the proportion of carbonates and beds of Exo~raExogyra increases until the Member is mostly dolomite with current bedded Exog~ra.Exogyra. In the West Basin the dolomite passes into shelly limestones and at the most western exposure (Portesham, SY 6185, 2 m thick) contains quartz sand. On the Isle of Portland the Member is shelly argillaceous dolomite with lime mudstone nodules.
(C)C) PONDFIELD MEMBER The name is taken from the section on the east side of Pondfield Cove (SY 8779) at the west end of Gad Cliff (Fig. 2). This Member is similar in lithology to the one below but thinner, more argillaceous and without Exog~raExogyra beds. Thickness and facies changes are given in Fig. 2.
(D) GAD CLIFF MEMBER The Member is named after Gad Cliff in the Isle of Purbeck (SY 8879, Fig. 2). Throughout the outcrop the rock is grey/brown fine-grained dolomite with less than 10%IO% clay, silt and very fine sand, all previously described as sandstone (Table I). It is usually bioturbated by RhizocoraUiumRhizocorallium (R. irregulare ofofFiirsich Fiirsich 1974)I974) up to a metre long but remnant original lamination (10-30(IO-3O #m)p.m) is also present. Excepting ammonites, body fossils are rare. Cross-cutting veins of ferroan calcite occur and dedolomitisation has taken place at certain horizons, producing black nodules ofpoikiliticof poikilitic calcite with celestite in places (Fig. 2). In the East Basin the lowest beds contain up to 200/02o% clay and locally up to 50%5o% quartz silt and sand.&and. Towards the swell the Member thins and locally passes into shelly limestone (current deposited Exogyra, Fig. 2). Over the swell the Member consists of brown dolomite with white nodules of lime mudstone, both containing ammonites,
Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 SY636855 SY 690833 SY 716823 SY763815 SYSI6799 1" SYB2179'8 SY 7987 SY 9575 COR TON 6km TOUT 35km OSMINGTON 4-7km RINGSTEAD 6km DUNGY HEAD 5km GAD Skm HOUNSTOUT 8 FARM DOWNS HILL BAY 8 STAIR HOLE CLIFF ST ALBAN'S HEAD 7 21 EAST
10
9
8 CORTON HILL MEMBER
7
I:: 6
5
EAST BA SIN 4
1»:<1 ~ I;!:::,!~:~~(;~~~~~e 11, + sand G ARGILLACEOUS 3 t~;;;;'~~lSPONGE SPICULE WACKESTONE I DOLOMITIC Emmit Hill Beds (RHAXELLA) 8DOLOMITE (tine grained) I CALCAREOUS CORTON HILL : ::z: 2 ii MEMBER SHELLY (Bivalve,) r:;:;:::::JQUARTZ SANDSTONE 4 17 E Mostly ~ (BLACK NORE [J[JMOOERATELY SANDY (25-30%Quartz) § Abundant } SERPULlDS MEMBER) 16 Common (GLOMERULA) S (Thicknesses ore for [J[JSlIGHTLi SANDY (IO-25%Quortz) 15 AMMONITE HORIZON total Block Nore <0 Member) 14 o DCLAY M FOR LOCALITIES SEE FIG 2 E BLACK NORE T t ....~~':JBIVALVE SHELLS SY 636855 = Ordnance Survey R MEMBER OJ"'..., .... WACKESTONE - PACKSTONE Grid Reference E BED NUMBERS ARE TERMINOLOGY OF ARKELL 193~ a 1941 S
FIG. I. Upper part of lower major regressive cycle in the Portland Group of Dorset. Variation in facies and thickness of the Corton Hill Member and the Upper Beds of the Black Nore Member, Portland Sand Formation.
Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021
624 W. G. Townson
serpulids and a moderately diverse bivalve assemblage. In the West Basin and on the Isle of Portland the Member is dolomite throughout with less than 55% % clay and sand.
Portland Limestone Formation
(A) L1THOFAC1ESLITHOFACIES Six facies are described using the terminology of Dunham 1962.I962. Facies I:i: Lime mumudstone. dsto ne. Micrite with less than 10%Io% grains, other than faecal micropellets.micropeUets. FaciesFacies2: 2 : Spiculewackestone.Spicule wackestone. Micrite with more than 10%I o % grains (matrix supported) of which more than 500/05o% are RhaxdlaRhaxella "spicules", usually preserved as calcite casts. In places the latter comprise up to 70%7o% of the sediment. Accessory com-com ponents include Pachastrella spicules, fine bioclast sand, shells and 0-5o-5% % very fine quartz sand. The amount of chert in rocks of this facies is directly related to spicule concentration (Figs. 2, 3). Facies 3: Fine hioclastbioclast packstone (to wackestone). Well sorted fine bioclast sand, usually grain-supported with a micrite matrix. Accessory components include Rhaxella spicules, serpulids (Glomerula) and quartz sand. Bivalves, ammonites, serpulids and ThalassinoidesThaIassinoides are moderately common. Cherts are rich in Pacha-Pacha strella spicules. Facies 4: BiodastBioclast grainstone. Sparite-cemented medium to coarse bivalve sand plus echinoderm fragments and unidentifiableunidenfifiable micritizedmicrifized grains. Bivalves, ammonites, serpulids and trace fossils are common. Quartz sand content is less than 1%. Pachastrella spicules are present and, when concentrated, acted as focal points for chert formation (Raisin II9o3). g03). Locally the rock is a packstone. Facies 5: OoidOvid grainstone. Poor to well sorted ooidsvoids (300-600(3oo-6oo/am /lm diam.) with a varying amount of sparite cement. Accessory grains include intraclasts and fragments of shell and algae. Shells are less common than in Facies 4. Locally the rock is a packstone. Facies 6: Bivalve shell grainstone-packstone. Grain-supported whole or fragmented bivalves with lime sand or mud matrix. Sorting is poor. Accessory components include echinoids, gastropods, bryozoans, serpulids, sponge spicules, red algae and intraclasts. Bioclasts range from medium sand to coarse pebble grade. Aragonitic bivalves and their debris are preserved as micrite-enveloped sparite casts or as moulds. Calcitic bivalves, echinoids and their fragments often contain well preserved algal bores.
(B)(3) DUNGYDuNoY HEAD MEMBERM M ER The name is taken from the section below Dungy Head, near Lulworth Cove (Fig. 2). Throughout the outcrop the base is the contact with the dolomite of the Gad Cliff Member. The top is a well-marked bedding plane on a widespread highly burrowed limestone bed rich in bivalves, Thalassinoides and ammonites
Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 SY SY SY SY SY SY Sy 611859 690033 763815 1l161'99 7987 918800 9575 '3 Km (5 Km Si Km
36 DUr~GY HEAD (}5 I Portland Clay MEMBER G-J 34 I DUNG\' I I __t------~~-,t_--HEAD 32 GAD CLIFF MEMBER 30 rg MEMBER 13- F I I - 28 ~ 26 ~E PORTESHAM FARM -- SWEl.L AREA 24
Isle of FRIAR WAODON Hill 22 Portland TOUT DO'lmS SV 6878 (DANCING lEDGE MEMBER) - / 20 '-./ " cb 13-17 v"" PORTLAND GAD CLIFF v WEST BASIN (incLA) " v >--< SHELL BED IViEMBER 16 v " r\ cb 3m DUNGY 24-28 I (ind A) I1 / PORTLAND HEAD 1 CLAY MEMBER !6
BIVALVE SHELLS 14 4m J (E"EXOGYR4 DOMINANTl / GAD CLIFF 10 a; :'5 I.>:.:.::.:. I PACKSTONE (IIIACKESTOWE) SERPULIDS / I MEMBER .. FINE 810CLAST SAND ABUNDANT <> (GWMERUf...Al I 8 <> JURASS/C rz==:::zJ DOLOMITE / / ~ I~ (FINE GRAI~lErl) RHIZ(JCOR4LI...IUM I WEYMOUTH - PU RBEC K >--(" THALIlSSINOIDES 6 II / c=JCLAY GAD CLIFF IO·2m ANTICl.INE / 30' 4 N SANDY (QUARTZ) VERTICAL LETTERS AND NUMBERS I1 I 50+ VAPlIATION IN ARE TERMINOLOGY OF Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf, CHIERT CONTENT ARKEll 1935 & 1947 by guest 1-/- PONDFIELD MEMBER ISLE OF ARGILtACEOUS (I!L/J,CK TO LIGHT 2 on 30 September1/-·, 2021 1·3m EAST BASIN PORTLAND GREV. TABULAR, vJ" i NODULE (LST. OR .DOL) NODULAR OR E (CORTON HILL MEMBER) o " 'J ~ASSI\IEl DEDOlOMITE (CALCITE) <> METRES (Sf. CELESTITE j FIG.2 MIDDLE MAJOR REGRESSIVE CYCLE IN THE PORTLAND GROUP OF DORSEt VARIATIONS IN FACIES .AND THiC~ I:'zi:;=tt'-al-:;t~---- WINS PIT MEMBER EAST BASIN Totol thicknesses LEDGE ,""'EMBER In~ k.rlown from I G S DURLSTON HEAD mapping and Irom wall. S" 034772 iOm 2 s:z;;, FACIES 6·5", Sm EHim 6'5m UPPER v WEST BASIN BEDS o=g LIME MUD STONE SWELL AREA WINSPIT SPONGE SPICULE (RHAXELLA) 2?IO'W MIDDLE .WACI(ESTONE R-U BEDS ISLE OF MEMBER :3 []JJ .PACKSTONE. FINE BIOCLAST· SAND.. PORTLAND o 0 R SET LOWER (-WACKESTONE) (WEST COAST) BEDS 14-19m O~Q PACKSTONE } MEDIUM TO COARSE SY 6756G5 (CONT I NUOU S 4 EXPOSURE) GRAINS TONE BIOCLAST SAND SOUTli EBB DANCING LEDGE Si. ALBANS MEMBER 5'5-7'5m 0000 PACKSTONE 1 HEAD K:N 5 000 0010 SAND 0000 GflAINSTONE f SV 959154 .. 0" 1-1--.....,...... PACKSTONE } 6 GRAINSTONE BIVALVE SHELLS r'-'''''..... ,.,-..."-" / Kilometres .....,."",,,-,,, 81 VALVE SHELLS VERTICAl. 14 o 4 5 SCALE ~ ·AMMONITE HORIZON IN 12 METRES TETRA)(ON SPONGE SPICULES HORIZONTAL ANI) VERTICAl. SCALES 10 /- COMMON ("PACHASTR£LLA"J o 5 10 ENGLISH CHANNEL ~IDm __mm __ 'b======1 Km. ARI: CON51S'fAfIT TKROU~HOUT 8 -;;-s-.s- SERPUUD RICH HORIZON (GLOM£RULA) 6 4 .>---<: TRAC E FOSSIL THALliSSIIVO/D£S WM - 6-IOm Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdfCOMMON PUR BECK GROUP by guest DlM - 14-15m 2 on 30 September 2021 i ® CORAL LOCALITY (/SASTREAj PORTLAND Winspit Member o VERTICAL VARIATION LETTERS AND NUMBERS UPPER MAJOR REGRESSIVE CYCLE IN THE PORTLAND GROUP OF DORSET. PORTLAND LIMESTONE Dancing Ledge Member RED ALGAE HORIZON (SOl.£NOPORA) Ill! CHERT CONTENT ARE TERMINOLOGY OF ~ ARKELL 1935 & 1947 (SLACI( TO LIGHT VARIATIONS IN FACIES AND THICKNESS OF THE DANCiNG LEDGE AND FORMATiON Dungy Head Member GREY, TABULAR I GROUP -:::=:::::::::: LLH LATERALLY LINKED HEMISPHERES} BLUE- NODULAR OR MASSIVE). WINSPIT MEMBERS OF THE PORTLAND LIMESTONE FORMATION. FI G. 3 PORTLAND SAND FORMATION fri"v;'\\ S 1-1 STACKED HEMISPHERES GREEN I ALGAE DESICCATION CRACKED The type Portlandian 625 (Arkell's Bed J', Fig. 2, Table I).1). This bed in thethe East Basin was lithostrati-lithostrati graphically correlated with the "Basal Shell Bed" (Table i)1) in thethe West Basin and Isle of Portland by Townson (197 I). The same correlation has since been proposed on ammonite evidence by Cope & Wimbledon (I973).(1973). In the East Basin the lower beds consist of very cherty, brown, bioturbated Rhaxella-rich wackestones (Facies 2) in which rare ammonites are thethe only body fossils (Fig. 2). Towards the swell RhaxeUaRhaxella spicules comprise up toto 70% of thethe rock. The upper beds in the East Basin are less cherty and consist of grey-brown bioturbated fine bioclastic packstones (Facies 3) with common bivalves, serpulids (Glomerula gordialis) and ammonites. Changes of facies and thickness over the swell are shown in Fig. 2. In the West Basin and on the Isle of Portland the Member is divided into the Portland Clay overlain by the Portland Shell Bed ("Basal Shell Bed" of Blake 1880,I880, COXCox 1925,I925, Arkell 1947,I947, House 1969)'I969). The latter is a cherty bioturbated shell-packstone with less than 10%Io% RhaxeUaRhaxella casts. The most common fossils are pterioid and trigoniid bivalves, serpulids, ammonites, gastropods and echinoids but bryozoans, asteroid ossicles, crustacean fragments, fish teeth and reptile remains also occur (Cox 1925).i925). (c)C) DANCING LEDGE MEMBER The name comes from the exposure at Dancing Ledge (SY 9976) in the Isle of Purbeck (Fig. 3). The base is everywhere put at the sharp junction with the shelly limestone at the top of the Dungy Head Member. In the East Basin the top is put at the change from grey-brown Facies 3 to paler coloured Facies 4. Also, the amount of chert decreases and changes colour from black/dark grey to light grey/white. In the West Basin the top is put at the change from Facies 2 to Facies I. Facies and thickness changes are shown in Fig. 3. In the East Basin the lower beds are grey, very cherty, with shell beds and several ThalassinoidesThala~sinoides and Glomerula horizons accentuated by pressure solution. On the swell the beds contain less shell debris and weather white with large black chert nodules. In the West Basin the Member contains very few shells, scattered fragments of bivalves, echinoderms and serpulids, also rare foraminiferans and ostracods; quartz sand is absent. The higher beds contain up to 40% casts and moulds of Rhaxella spicules, the silica of which has been reprecipitated as discontinuous horizons of black chert nodules, up to o'So- 5 m thick. In these the spicules are preserved in their original siliceous form. On the Isle of Portland the thickest known sections of the Member are in the cliffs around most of the island (House 1969, details in Townson 1971).197 I). The lower beds are massive bedded with large black chert nodules, common Thalassinoides and only very few thin shellshcU beds. The upper third contrasts with the lower two thirds of the Member in that it contains thick current-bedded shell beds, common PachastrellaPacha~trella and a widespread current deposited GlomerulaSlomerula serpulite (up to 2 m thick). Also it is less silicified and the chert is dark to light grey. Overturned blocks up to 2 m across of the coral Isastrea oblonga occur in the massive shell beds and ammonites are more common than below. Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 626 W. G. Townson (D) WINSPIT MEMBER The name is taken from the exposures in the old clifftopclifftop quarries at Winspit (SY 977760) in the Isle of Purbeck (Fig. 3, Table I). In the East Basin the Member reaches a maximum known thicknessthickness of 19 m near Dancing Ledge. It is well displayed all along the coast from Durlston Head to St. Albans Head but is most accessible at Winspit, Seacombe (SY 983766) and Tilly Whim "Caves". An informal threefold division is practical, based on the vertical facies sequences (Fig. 3). The lower beds (4"5-9"5(4'5-9'5 m) are light to dark brown coloured Facies 4 with shell beds in the upper part. In the pure bioclast sand the only fauna is occasional Pachastrella spicules and Serpula sp. Cross bedding is not often visible but vectors NNW. and SSW. are present. The base of the middle beds (3.0-6.6(3'0-6·6 m) along most of the coast is a chert-rich Facies 3 horizon (Arkell's "Listy Bed" 1935). Above this the middle beds are Facies 4, with intraclastic shelly beds in the upper part. Small-scale planar cross-beds are sometimes visible with foresets dipping S., W., NE. and SE. High angle burrows occur (cf. Scolithos). The upper beds (2'5-6'5(2.5-6. 5 m, Fig. 3) have at their base 0'5-1o.5-1.5 '5 m ofbioturbated bivalve packstone/wackestone,packstonejwackestone, current bedded in places. The rock weathers to a grey flaggy honeycombed appearance. This is overlain by Facies I with common gastropods, scattered bivalves, burrows and occasional current-deposifionalcurrent-depositional structures. The topmost stratum is 0'50" 5 m thick over the whole East Basin and consists of a dark brown ostracodal stromatolitic limestone which marks the top of the Portland Group. The "Tilly Whim Oyster Bed" (Arkell 1947) is the only exception to the above description (Fig. 3). The basal shelly layer locally thickens to nearly 3 m and consists almost entirely of current-disturbed large bivalves and bioturbated coarse bioclast debris with occasional clasts of Solenopora (details in Townson 1971).197 I). On the swell the Member is well exposed on either side of Lulworth Cove and at Ringstead Bay. Facies and thickness changes are shown in Fig. 3. The oolite contains well preserved high-angle micrite lined burrows and trough cross beds 0'2-0'4o'2-o" 4 m thick. The uppermost 2'5-3'752"5-3"75 m changes across the swell from lime mudstone to ooid packstone and grainstone (Fig. 3). The thin overlying stromatostromato- litic limestone is present, enabling lithocorrelation with the East Basin. In the West Basin, west of Ringstead Bay, the member is exposed at only a few inland exposures (Fig. 3). Grainstones are absent. RhaxeUaRhaxella spicule casts comprise up to 400/04 ° % of the micritic sediment; also present are bivalves, gastropods, ammonites, fragments of echinoderms and crustaceans, Pachastrella spicules, foraminiferans and ostracods. The upper half of the Member is white biobio- turbated chalky mudmudstonestone composed of micrite less than II/~m pm across (no coccococco- liths were found by scanning electron microscopy). The junction with the Purbeck Beds at all exposures is at the ostracodal algal limestone. On the Isle of Portland 16016o vertical sections have been measured in an outcrop area of 88km km 2~ (see Townson 1971197I for locality details and isopach maps),maps). The Member (6-10(6-1o m) consists predominantly of grainstones throughout the island with a lime mudstone intercalation in the north (Fig. 3). Oolite prepre- dominates but bioclast sand and lime mud are also present. Locally shells are the Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 The type Portlandian sole component. Small algal-oyster-bryozoan "patch reefs" occur and debris derived from these is common. The sixfold division used by Arkell (1947) is not applicable over much of the island due to rapid lateral variation. An informal threefold division is used here: the lower beds include the "Base or Best Bed" and "Base Bed Roach"; the middle beds include the "Curf" and "Flinty Bed"; the upper beds include the "Whit Bed" and "Roach". The lower beds (1.5-4.2 m) comprise a lower oolite unit (1·5-3·5 m) spread over the whole island and an uppershelly unit (0-2·5 m) more patchily distributed (Fig. 3). Planar cross strata are sometimes visible up to 2 m thick with dip directions SW. to SSW., except at Portland Bill where the direction is NW. The upper unit is a shell bed, usually less than I m thick, dominantly composed of disarticulated trigoniids in a matrix of ooids, bioclasts and lime mud. At Mutton Cove, however (SY 6797 I 2), the lower beds consist of a shell-supported mass (4·75 m thick) composed of large epifaunal and infaunal bivalves with gastropods, serpulids and partly silicified colonies of Isastrea. The middle beds in the northern half of the island (0-2·5 m) consist of Facies I and 2 with occasional trigoniids, ammonites, gastropods and crustacean fragments, but no thick shell beds. Grey and black tabular chert is common, up to 0·25 m thick. Southwards these beds contain more bioclast sand and pass into cherty cross-bedded grainstones (Fig. 3). The upper beds in the northern half of the island (3·6-6·0 m) comprise a lower oolite unit overlain by a shelly unit. The lower unit (2-5 m) consists of Facies 5 with accessory shells, fragments of shells, red algae (Solenopora) , Paehastrella spicules, oomicrite intraclasts and lime mud. Erosional planar beds are some times visible dipping mainly to the SW. The upper, shelly, unit is widespread over the northern part of the island ("The Portland Roach"). The characteristic fossils are Laevitrigonia, Isognomon, Aptyxiella and Solenopora, in a current deposited grainstone matrix. Nodular and tabular dark brown chert is common and in this the aragonite shell structure oftrigoniids and intra-ooid laminae is well preserved. The middle and upper beds further south are grains tones throughout (3·75- 6·7 m). Across the centre of the island Facies 3 passes southwards into cherty current bedded Facies 4 and 5 (Fig. 3) which soon pass into massive pure oolite up to 6 m thick with bedding plane joints 2-3 m apart (e.g. Broadcroft Quarry at SY 700719 and in the railway cutting near Rufus Castle SY 697712). South of this the thickness decreases to 4-5 m and the sequence comprises thinly bedded oolite units (0·5 m average). Each unit consists of an upward fining basal shelly layer (with or without Solenopora debris) overlain by oolite and often oomicrite. Cross bedding is usually obscured by bioturbation. There are some local variations to the above general description of the upper beds. In the shell beds there are blocks of oyster boundstone 0.05-1 m + across, consisting of mutually cemented Ostrea with subordinate Plicatula and sometimes Solenopora, bound by multilamella encrusing bryozoa. Local oyster "patch reefs" are present in situ up to 2 m high and 20 m long (e.g. SY 69137235). Also common are blocks of "patch reefs" composed of Solenopora bound by bryozoa, with subordinate Ostrea and Plicatula. Examples up to 3 m wide by 1 m high occur in situ (e.g. tipped blocks at SY 703716 and on the coast at SY 6850568950 and Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 628 W. G. Townson 6822568700).6822568700 ) . Both the oyster and the algal-dominated "patch reefs" are bored by Lithophaga, sponges, algae and annelids. Solenopora debris from boulder to sand size is recognizable by its fine banded structure. On the west coast at SY 6787036787o 3 the uppermost 2-1'52-t. 5 m of the Member is entirely composed of bioclast sand with a brown microstalactitic cement similar to that described by Taylor & Illing (1969) and Purser (1969)(I969) as forming subsub- aerially in high intertidal to supratidal environments. Along the northeast face of Broadcroft Quarry the oolite gives way northwards to cherty lime mudstone with an erosional sloping contact (see Townson 1971).I97 I). Interpretation (A) PALAEOECOLOGYPALAEOECO LOOY The palaeoecology of the benthonic macrofauna in the Portland Group is summarized in Table 2. The great majority of the genera listed are suspension feeders. Present day forms are known to be largely confined to sandy or firm mud bottoms, whilst deposit feeders are common on soft muddy substrata (Rhoads & Young 1970).197o). (B) DEPOSITIONALD .POSITIO, AL MODELS The transition from Kimmeridge Clay through Portland Group into the Pur-Pur beck Group is considered to be the result of a worldwide fall in sea level (Hallam 1969).x969). In Dorset it appears that this was interrupted by at least two minor transgressions resulting in three phases of regression (Table i,I, Fig. 1-3).I-3). Applying Walther's "Law of Facies" (1893-94)(t893--94) the basic model proposed for deposition during these three phases is that of a marine basin with a marginal shallow carbonate shelf. This sloped into deeper offshore water where siliciclastics accumulated (Fig. 4). The sediment distribution was mainly depth controlled in the sense that areas of high, medium and low turbulence or "energy" existed at different depths. (I) The deeper water facies (Fig. 4) In the deepest parts of the basins bituminous claystones (Kimmeridge Clay) were deposited in quiet water (50-100(5o-Ioo m deep?)deep ?) which was sometimes anaerobic. Nearer the source of terrigenous material quartz silt and sand accumulated. Between the carbonate shelf and the siliciclasticsiliciclastie basin clay and lime mud were deposited. The Kimmeridge Clay contains thin beds of dolomite and this mineral is present in varying proportions in the Portland Sand Formation. For reasons given below the dolomite is regarded as synsedimentary with a relatively deep marine origin. Friedman & Sanders (1967)(x 967) conclude that synsedimentary dolomite is formed from hypersaline brines in two main environments: (a) in supratidal and intertidal sabhkas by evaporation, and (b) in deeper water by magnesium-magnesium enriched brine concentrated by basin margin evaporation. This Mg-rich brine flows by density contrast over the sea bed to a bathymetric low and dolomitises aragonite and high-Mg calcite mud (Adams & Rhodes 1960,t 96o, Schmalz 1969).t 969). Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 TABLE 2: Palaeoecology Of the benthonic macrofauna of the Portland Group of Dorset. Based mainly on Stanley (1970), Barthel (1969), KaujJman (1969), Nichols (1969), Rudwick (1970), and Fursich (1974); complete list in Townson (1971). -- GENERAL FEEDING ACTIVITY METHOD ('0 fo< 11) BENTHONIC MACROFAUNA IN THE I Arca A S ARCOIDA Barbatia A S Parallelodon A S MYTILOIDA Falcimytilus A S E Musculus A S E Camptonecte s (A) US' S E Plagiostoma (A) US S E ~ -( Isognomon (A) US 5 E Z PTERIOIDA Oxytoma A S :J Exogyra A S E -f, Ostrea A 5 E la. il: Plicatula A 5 lzl .-- GA C GASTROPODS PLEUROTOMARIDAE CEHITHIACEA GA H E 1':CllfNOIDS IIEMICIDAHOIDEA GA C nJtACTtIPOl)S lUIYNC HONE LLIDA A S TEH.lmH.A TU LIDA A S SPONGI-';S A 5 DRYOZOA A S St<:H.PULIDS A S E CORALS A S MYTIT.OIT>A Modiolus A S E L~thophag" A S- E nrVALVES Astilrte US S E ..:l V"~NF.ROroF.A ,-f, Corbicellopsis US S. E ..... Z ::E:J lzl NUCULOIDA Nucula GA ri Nuculana GA D Laevitrigonia US S E TRIGONIOlDA Myophorella US oS E ..:l "Trigonia" US S E TRACE Chondrite 5 animal GA D FOSSILS Rhizocorallium animal GA D Pholadomya US S E ..:l PHOLADOMYOIDA Pleuromya US S E !lo< BIVALVES Thracia US S E lzl§ US S E lzl (A) =Attached when young US' =Probably capable of swimming Euryhaline = Capable of withstanding fluctuations in salinity Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 630 W.W. G. Townson The occurrences in thethe Portland Sand Formation are considered to have the latter origin because:-because :- (I)(z) No characteristic features of sabhka deposits are present, such as mud cracks, birdseyes, boudinage, scour & fill, channels, ripples, cross bedding, algal structures, gypsum/anhydrite or signs of their former presence. Also, marine fossils are present. (2) In the Black Nore Member, dolomite rhombohedra occur in clay laminae which alternate with laminae of quartz silt and sand concon- taining low-Mg calcite shell debris and glauconite. These sediments contain ammonites and are intercalated with thin lime mudstones with marine bivalves. The whole sequence was originally laminated but subsequently bioturbated. (3) In the CortonCarton Hill Member the ExogyraExogvra beds sometimes have finefine- grained dolomite as matrix to the low-Mg shells, indicating selective replacement of aragonite and/or high-Mg calcite mud. A good example of this is at Stair Hole where the Gad Cliff Member contains a low-Mg calcite limestone (Exogyra-rich)(Exog~ra-rich) which passes laterally into shellshell- free fine grained dolomite. (4) The Gad Cliff Member is generally a dolostone, devoid of body fossils except ammonites. On the swell at Ringstead Bay, however, it passes laterally into a shallow marine shelly limestone which is only partly dolomitised (Fig. 2). L L ~------F PORTLAND SAND FORMATION ------..- 14---E PORTLAN0PORTLAND LIMESTONE FORMATIONfORMATION J I SHALLOW WATER CARBONATE I DEEPER SHELF AND I RELATIVELY DEEP I SHELF 'J SLOPE IJ BASIN (Generolly(Generally well oxygenated) I (Transitional zone) J(GenerallyI( Generally anaerobic sea bed) A i Sea ~. ~.-,o-.~,~ ,3- '' TURBUL.ENTTURBULENT ZONE Level - • .... "- " "9 "-" *:%"~.~k-..',~. - ~...~i~• "~" "6:-" " " o" - -'0~- ~~:..~ ....v. .-...... ~:::~:~ ,,#. ,r,~-:, , _ ~. , :,,-,...~ , ...... o \_ % s,o,., . • ~ .. ,-- . -,-.....!.~;. General faunalfauna I LOWLow EnergyEneroy J Diverse fauna ~ diversity low I I Low Energy' I~CARBONATE SAND---~ NONo ~entho~cbenthonie foanofauna I Low Energy' JI CLAY, SILT &a SAND "LIME MUD DEPOSITION L I ~---- Dolomitiaation --'------;~ P Dolomiti=otion Sea Level ·~~=rb~ul~en~f-z~on-e------Turbulent Zone SWELL ExaggecatedExaooerated figurefioure toto show CRITICAL __~ ~ ~ ~'.~..,__._. thethe influenceinfluence of thethe swell on 4 , ~ :'~'.;'" .. -- ' LEVEL depths of water and hence --~ / / J..-l-"~'~Z,"b ,'T~ Lime on dolomitisotlondolomitisation inin thethe / / .,~ Mud Portland Sand Formation. BASIN J fringe of partial I BASII~ doiomitisQtiondolomitisation I dolomitisotion J dolomitlsationdololnitisation For keys toto symbols see Figures 1 a 3'3 Fzo.FIG. 4. General model for environments of deposition of the Portland Group of Dorset. Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 The typetype Portlandian 631 These observations also preclude a hydrothermal metasomatic ongIn,origin, a subaerial origin related to soil formation or an allochthonous origin (wind or water transported). The dolomite is interpreted therefore as of submarine origin. Although no present day examples are known to the author it would appear that aragonite and/or high Mg-calcite lime muds entering an alkaline decay environenviron- ment would be dolomitised at or within the seabed, especially if the brine in which it was suspended was warm and magnesium enriched (Friedman & Sanders 1967, Schmalz 1969, Bathurst 1971).197 I). A general genetic model of deep water evaporite deposition has been proposed by Schmalz (1969). During the first stage evaporation commences, calcium carbonate is precipitated, the sea is well oxygenated throughout, with a normal marine nectonic and benthonic fauna. As evaporation continues heavier brine displaces normal sea water from the basin floor and a euxinic phase starts. Nectonic fauna still inhabit the surface water but the benthos is anaerobic and reducing conditions occur, possibly with free HH~S.2S. According to Schmalz, if evaporation continues the surface water along the basin margin becomes hypersaline and gypsum is precipitated. This is redissolved, however, before being transported to the basin floor; fauna is sparse or absent at all levels. The basin will be occupied for long periods by brines depleted in Ca-ions relative to Mg and this results in dolomitisation of calcium carbonate. This model for dolomitisation is attractive when the Portland Group and the basal beds of the Purbeck Group are considered together (Figs. 4,4,5). 5). The latter are rich in calcium sulphate minerals which are inter to supratidal in origin (West 1964, 1975, Shearman 19196666 , Cope, Clements & West in Torrens 1969I969 Pt 13). Such an evaporite facies may have existed along the basin margin during deposition of the Portland Group. Precipitation of calcium sulphate in a marginal sabhka would increase the Mg/Ca ratio of the seawater without significantly altering the proportion of carbonate ions. Warm Mg-rich seawater could have flowed off the shallow marine carbonate shelf into basinal waters where the dolomite of the Portland Sand Formation was formed. The carbonate shelf was sufficiently shallow for aeration of the water, thus a moderately diverse molluscan fauna flourished. However, scarcity or absence of stenohaline organisms (corals, brachiopods, crinoids, echinoderms and belemnites) on the shelf might be a reflection of slightly abnormal chemical composition of the sea. It is postulated, therefore, that during deposition of the Portland Group there was a "C1itical"c~itical level" in the sea, below which oxygen content was low, pH was high, conditions were stagnant and dolomitisation of aragonite and high-Mg calcite mud took place at or within the sea bed. (2) The carbonate shelfshelffaciesfacies (Fig. 5) The model depicted in Fig. 5 shows, in an exaggerated fashion, how thethe carbonate facies are related to depth, general level of turbulence ("energy levels") and the presence or absence of a swell. It is not intended toto imply synchronous deposition of the three Formations. The arrangement of thethe facies around thethe swell is closely comparable with that of Recent sediments around salt dome islands and swells in the Persian Gulf (Purser 1973). Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 Z N . a at OR COMMON VERY OSTRAC SAND, DEPOSITION a ENERGY ~,~ MUD, ~ ~_ ~ ~ ~ S04 versification RESTRICTED LOCALLY FAUNAS PELLET ABSENT. ALGAE STROMATOLlTIC LIME Ca di FORMATION LOW LULWORTH LAGOON HYPOSALlNE ,,+< EVAPORITIC ...J ~: BY facies -I13 ,,=, ,=, @ ,_ o) < -- l,aJ ..7'+ = ~ .,' Z ,.m, o ~ o. .-~ DIVERSITY ~,~.~r-.~ ~,~ WACKESTONE 0 ~.~ ENERGY DEPOSITED °~°GASTROPODS, ~ -- LAGOON SHELTERED maximum fades. BEDS. SHOAL. DENSITY. ALGAE. ~,.,<~.>FAUNAL < =+- o~ o.~ ~ ,~.~ o= Numbers PATCH Formation ANGLE ENERGY W SHOAL DIVERSITY DENSITY. SWELL • o.'~ ENERGY swell; 010 REEFS. GRAINSTONES LOCAL r" :2. © BURROWS HIGH AND ON ++~ LOW .<--..,-.,.+ ~ +"+ LOW -.~ _1 of I I I Limestone HIGH ~oo .-<.. . ,- ~ plunge AND BIVALVE FAUNAL HIGH SHOAL FRONT SAND OF IN ates >- >.,., ~ ,->.- ~ Portland ° ,,.0 = ,.<, 0010 ,,.~ o..= .~ the SHELL n,- "~ Z ~ "1": ~ 0 ~ indic AREA OF BIOTURBATED DEMOSPONGES SERPUUDS TETRAXON DENSITY DIVERSITY. MODERATE of Z LIMESTONE ENERGY tion Arrow SAN 0.. IN si . E ~,.-',,.'fi,:o " ~ _o ~. f,~ o-':+-.'-+ l. depo o ___ swel GEOIID 100km?------~;.~ SHELL SHAPED -- ILl for the OF MEDIUM ," ~ ~+_. =~)~DIVERSITY Y - SPICULES FINE MATRIX ~ + ,-+oCOMMON. _I++ • - ~ >~ o over model DENSITY FAUNAL MUD o o .,+ | ]~ I"~lk water SPONGE ~)0 N ~,. 0 .~ PORTLAND BURROWS HORIZONTAL DEMOSPONGES. LOW HIGH AND LIME . Generalised shallow "+X+ of /,'5 MUD V.LOW LAM. DIV/DEN LIME time FIG.5. ~------~ - ~.~, ENERGY LOW \/<:: O0 s Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 The typetype Portlandian 633 Above thethe critical levellevel limelime mud was not dolomitized and Rhaxella sponges were able to survive (Facies 2). The sea bed was soft, eliminating a diverse bivalve population with the exception of those which encrusted ammonite shells. The characteristics of Facies 3 and 4 are shown in Fig. 5. The water over the swell eventually shallowed to the point where it became saturated with calcium carbonate and was warm and agitated sufficiently frequently for ooids to form (Facies 5). The distribution of the fauna in the grainstone facies (4 & 5) is comcom- parable with that in Recent examples where ooid sand dunes are biological deserts which are colonized only during periods of quiet. The maximum density and diversity of fauna is on the seaward margin where the water is turbulent enough to be well oxygenated, there is a plentiful food supply and the substrate is more stable (Newell & Rigby 1957,i957, Ball 1967). Landward of the swell and protected by it the water was shallow, well oxygenated, food was plentiful and a rich fauna existed (Fig. 5). The area was subject to occasional current action indicated by the presence of cross bedded shells in places. Bordering this zone on the landward margin was a belt of stromatolitic algal mounds (Logan et al. 1964) which formed a barrier sufficient to separate water of near-marine salinity from hypersaline water in which evaporites were precipitated (West et al. 1969). In areas away from the swell the facies changed from barren deep water lime mud (dolomitised) to fossiliferous shallow water mud with only an intermediate transitional facies zone inhabited by Rhaxella, as seen in the West Basin (Fig. 5).5)- (c)(C) ENVIRONMENTAL HISTORY Within the area of study (33 X× 15 km) deposition of each Member of the Port-Port land Group was essentially synchronous. Correlation by "events" is possible in the sense that short periods of deepening of the water during the overall shallow-shallow ing are reflected in the sequence whether on the swell or in the basins. The three major regressive/minor transgressive cycles present (Figs. 1-3)I-3) have been recognized by the author (197(I97i) I) at the outcrops of Portlandian strata in southern England (Dorset, Wiltshire, Oxfordshire, Buckinghamshire), in the Bas Boulonnais of NE. France and in subsurface SE. England. Publication of detailed ammonite stratigraphy is awaited before synchroneity can be assumed but it is possible that the cyclicity was caused by eustatic sea level changes. The same mechanism is postulated by Talbot (I(1973) 973) for the Oxfordian of southern England. Lower cycle (Table I, Fig. II). ) This began with continued deposition of the Kimmeridge Clay after the break marked by the rotunda-nodule horizon (Arkell 1947, Cope in Torrens 1969, Casey 1971971). I). Relatively deep water clay deposition was widespread and thethe swell influenced only the thickness of the accumulations. The Black Nore Member was deposited as the regression continued and fine quartz sand and silt entered the area. Aragonite and high-Mg calcite were transported from thethe carbonate shelf and deposited as laminae of lime mud which were dolomitised when thethe seabed was below the critical level. The water was shallower towards thethe western edge of the West Basin where quartz sand and glauconite were deposited. A westerly Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 W. G. Townson source is suggested for the siliciclastics by the increase of sand and clay in that direction and by the heavy mineral studies of Neaverson (1925) and Latter (1926). The Corton Hill Member was formed in shallower water. On the eastern flank of the swell only dolomitic clay accumulated but on the top pure lime mud was deposited and only patchily dolomitised, the critical level being in the vicinity of the seabed. In the East Basin the small epifaunal bivalve Exogyra nana was abundant. Other benthos was rare except in thin limestones which contain deeply infaunal bivalves indicative of a soft seabed. The alternation of dolomitic clay with thin limestones probably results from cyclic shallowing with the restricted Exogyra horizons forming at the critical level and limestones above it. On the swell limestone deposition predominated, current action is indicated and the cycles are less distinct. West of the swell the matrix of the shell beds was dolomitised but on the margin of the West Basin a relatively thick shelly limestone succession was formed (Fig. I). The fauna was less restricted and includes tri goniids, small serpulids, sponge spicules and pterioids. Above the critical level the sea was still moderately deep and a soft mud bottom limited bivalve diversity . .Nfiddle cycle (Table I, Fig. 2) A slight sea level rise caused deposition of the Pondfield Member mainly below the critical level. In the East Basin dolomitic clay accumulated but on the swell the Member is a very thin argillaceous dolomite or is absent. In the West Basin a mixture of clay and shelly lime mud was deposited. The critical level was in the vicinity of the sea bed indicated by only partial dolomitisation. The Gad Cliff Member was deposited as the regression continued. Sand and clay failed to reach most of the area and lime mud was dolomitised everywhere except on the swell (Fig. 2). The extremely tolerant epifaunal oyster Exogyra nana was able to exist on the flanks of the swell at or just below the critical level. Elsewhere benthos was very rare or absent, apart from the spasmodic activity of the Rhizo corallium animal. The Dungy Head Member was deposited as the sea bed rose above the critical level. Clay was deposited in the West Basin but on the swell and in the East Basin lime mud accumulated (Fig. 2). During deposition of the lower beds lime mud laid down on the outer part of the carbonate shelf was colonized only by Rhaxella and the Thalassinoides crustacean. The sponges were adapted to living on soft mud but suspension feeding bivalves were unable to survive. Ammonites are rare which may be related to food scarcity on the seabed, or it may be that these beds accumulated faster than those with common ammonites. The overlying beds were deposited in shallower water and in the East Basin fine bioclast sand was added. Here the characteristic animal was the small serpulid Glomerula gordialis which adapted to living freely on soft substrates by self-encrusting in a knotted shape. The only other fauna present were rare bivalves, occasional ammonites and burrowing crustaceans. Slow deposition at the end of the cycle is indicated by the concentration of encrusted ammonites, bivalves, gastropods, serpulids, bryozoans, sponge spicules, burrows and crustacean fragments. Once ammonite shells and large epifaunal bivalve shells accumulated, hard substrates were locally Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 The type Portlandian 635 available for colonization. Spasmodic current activity produced bioclastic debris which made the sea bed firmer, enabling further colonization by suspension feeders. Upper cycle (Table I,1, Fig. 3) The sea level rose before the overall regression recommenced. Thus, in the East Basin, the Dancing Ledge Member has the same vertical fadesfacies change from 2 to 3 as has the Dungy Head Member (Figs. 2,3).2, 3). During deposition of the lower beds of the Dancing Ledge Member lime mud was deposited throughout the area and was colonized by Rhaxella, the Thalassinoides animal and occasional bivalves. The serpulid Glomerula was common in the East Basin. Over the swell the water was shallower and fine shell sand was deposited with the lime mud. In the West Basin sponges were only sporadically present. As shallowing proceeded fine shell sand made the substrate firmer and Pacha-Pacha strella, Glomerula and large epifaunal bivalves could flourish. The massive cross-cross bedded shelly units on the Isle of Portland (Fig. 3) indicate spasmodic turbulent conditions (hurricanes?)(hurricanes ?) on the swell before the water was sufficiently agitated for grainsgrainstones tones to form. During deposition of the lower beds of the Winspit Member the fadesfacies pattern was similar to that idealized in Fig. 5. Ooids were transported southwards on Portland and eastwards on the mainland. The shell sand in the East Basin was generated in front of the ooid shoal area. In the Isle of Portland area a submarine ooid dune or dune complex developed with its crest in the north, trending ENE.-WSW. During periods of minimal turbulence (i.e. most of the time) the ooid and bioclast sand was colonized by bivalves, Pachastrella and burrowers. The shells were redistributed during storms or hurricanes and debris was swept into depressions on the dune topography (Fig. 4 and isopachs in Townson 1971).1971 ). Muddy intercalations (Fades(Facies 2 & 3) within the grainstones in the East Basin and on the swell at Portland may indicate a further slight sealevel rise but the overall shallowing soon proceeded. The sediment pattern for the middle beds of the Winspit Member approximately repeated that for the lower beds. In the West Basin mud continued to be deposited but bivalves became more common as the water shallowed. The final shallowing during deposition of the Portland Limestone Formation resulted in the oolite shoal on the swell being surrounded by lime mud colonized by a diverse bivalve fauna. Occasional storms agitated the seabed producing local patches of cross bedded shell sand in a lime mud matrix. The ooid shoal was reduced in width and ooids were mixed with the onlapping lime mud along the margin (Fig. 3). In the north of the Isle of Portland small patch reefs of calcareous red algae and oysters colonized the slightly deeper water (2-3(2- 3 m?) on the flanks of the dunes (details in Townson 1971).197 I). Storm erosion cut into the stabilized dunes and channels were filled with lime mud (Broadcroft Quarry). The water over the dunes was very shallow and local beaches formed composed of bivalve and gastropod debris. The southward limit of Aptyxiella across the Isle of Portland probably indicates a temporary strandline along the swell. At the south end of the Isle the crests of the highest dunes were exposed to the air and a beachrock Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 636 W. G. Townson formed locally. The swell was probably the site of a chain of low islands by this time. Stromatolitic algae then spread across the whole Dorset area but the swell continued to influence sediment distribution and rate of subsidence. Facies of the basal beds of the Lulworth Formation are summarized in Fig. 3. For more details see Brown (1963,(I963, 1964),x964) , Pugh (1968),(I968), and West (1964,(x964, 1965,I965, in Torrens 1969,I969, 1975).I975). Relatively thick algal limestones accumulated on the swell and westwards whereas in the east thinner algal limestones formed with a stromatolite morphology suggesting growth in water 1-2 I-2 m deep. At times, trees grew on the algal mounds with their roots below water and soil profiles formed. With re-re striction and evaporation in very shallow water hypersalinity resulted and gypsum, halite and lime mud were deposited, mainly east of the swell. Occasionally, when salinity was lowered, ostracods abounded and burrowing organisms produced fine faecal sand. (0)(D) CONCLUSIONScoNc .usio s ((I) I) Three major regressive/minor transgressive cycles are recognized in the Portland Group of Dorset (Table I,x, Fig. 1-3).x-3). (2) Deposition commenced in a relatively deep marine siliciclastic basin. As a result of overall regression the area became a shallow open-marine carbonate shelf and finally a restricted-marine evaporitic margin (Fig. 4, 5). (3) A swell separated an East Basin from a West Basin and influenced sediment thickness and facies for at least 25 million years (Kimmeridgian-Portlandian span). The swellswell may be due to a Triassic salt pillow at depth. (4) The siliciclastics of the Portland Sand Formation came from the west but the dolomite present was formed in situ in relatively deep water. Brine, concentrated by evaporation on the basin margins, flowed into a euxinic environenviron- ment and replaced high-magnesium calcite and aragonite mud on or within the sea bed. A critical level is postulated below which dolomitisation took place; the presence of this level also largely controlled faunal distribution (Fig. 4). (5) The Portland Limestone Formation was formed on a shallow marine shelf (Fig. 4). Ooid sand formed when the water depth over the swell was particularly low (Fig. 5). The substrate conditions largely controlled faunal distributions but the rarity or absence of certain groups may have been due to slight hypersalinity. (6) The Lulworth Formation (Purbeck Group) was formed on the margin of the carbonate shelf. Stromatolitic algae formed a barrier behind which circulation was restricted and evaporites were formed (Fig. 5). The swell continued to influence facies distribution. ACKNOWLEDGEMENTS. I thank Dr N. L. Banks, Dr A. Hallam,HaUam, Dr T. J. A. Reijers, and Mr I. A. West for discussion. Research for this paper was financed by N.E.R.e.N.E.R.C. and supervised by Dr A. Hallam. [The Society is indebted to Dr Townson for a contribution towards the cost of Figs. 2, 3.] References AnAMS,ADAMS, J. E. & RHODES, M. L. 1960.x96o. DolomitisationDolomitlsation by seepage refluxion.reflurdon. Bull. Am. Ass. petrol. Geol.44,Geol. 44, 19x9 12x2-20. - 20• Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 TheThe typetype PortlandianPortlandian 637 ARKELL,AnK~LL, W.W. J.J. 1933.i933. JurassicJurassic SystemSystem inin GreatGreat Britain.Britain. ClarendonClarendon Press,Press, Oxford.Oxford. 344344 pp.PP. -- 1935.1935. TheThe PortlandPortland BedsBeds ofof thethe DorsetDorset Mainland.Mainland. Proc. Geol. Geol. Ass. 446,6, 301-47.3ox-47. -- 1938.x938. ThreeThree tectonictectonic problemsproblems ofof thethe LulworthLulworth district:district: studiesstudies on thethe middle limblimb of thethe PurbeckPurbeck fold.fold. Q.Q. JI.all. geol.geol. Soc. Soc. £ond.Lond. 94,94, I-53.x-53. -- 1947.x947. GeologyGeology ofof thethe countrycountry aroundaround Weymouth,Weymouth, Swanage,Swanage, CorfeCorfe andand Lulworth. Mem. geol. Surv.Surv. U.U. K.If. 386 pp. BALL, M. M. 1967.x967. Carbonate sandsand bodies of Florida andand Bahamas. J.8. sedim. Petrol. 37,556-91.37, 556-9 I. BARTHEL,BAaTHEL, K. W. 1969.1969. Die obertithonische, regressiveregressive Flachwasser Phase der Neuberger Folge inin Bayern. Abh. Bayer. Akad. Wissensch.Wissensch. Math.- naturw. Kl.,Ifl., N.F. 142,142, 1-174.x-x74. BATHuRsT,BATHURST, R. C. G. 1971.x97 x. Carbonate sediments and theirtheir diagenesis. Developments inin Sedimentology 12.x2. Elsevier, Amsterdam. 620 pp. BLAKE,BLAKe, J. F. 1880.i88o. On thethe Portland rocksrocks of England. Q. J.8. geol. Soc. Lond. 36, 189-236.I89-236. BRowN,BROWN, P. R. 1963.1963. Algal limestoneslimestones and associated sedimentssediments inin the basal Purbeck of Dorset. Geol. Mag. 100,xoo, 565-73. -- 1964.I964. Petrology and origin of somesome upper JurassicJvrassic beds fromfrom Dorset, England. J. sedim. Petrol. 34, 254-254-69.69. CASEY,CAStY, R. 1963.I963. The dawn of the Cretaceous Period in Britain. Bull. S.-East Un. scient. SocsSots 117,x x7, 1-15·x-x 5. -- 1967. The position of the Middle Volgian in the English Jurassic. Proc. geol. Soc. £ond.Lond. 1640,164o , 128-33·128--33. --- 1971.197 I. Facies, faunas and tectonics in late Jurassic-earlyJurassic--early Cretaceous Britain. In F. A. MiddleMiddle- miss & P. F. Rawson (ed.), Faunal provinces in space and time, Geol. J. Spec. Issue 4,193-266.4, 193-266. -- 1973.I973 . The ammonite succession at the JJurassic-Cretaceous urassic-Cretaceous boundary in eastern England. In R. Casey & P. F. Rawson (ed.), The Boreal Lower Cretaceous, Geol. J. Spec. Issue 6, 193-266. COPE, J. C. W. & WIMBLEDON,Wx~mL~VON, W. A. 1973. Ammonite faunas of the uppermost Kimmeridge Clay, the Portland Sand and the Portland Stone of Dorset. Proc. Ussher Soc. 2, 593-8. Cox, L. R. 1925. The fauna of the Basal Shell Bed of the Portland Stone, Isle of Portland. Proc. Dorset nat. Hist. antiq. Fld Club 46, 113-72.x t 3-72. -- 1929.I929. Synopsis of the Lamellibranchia and Gastropoda of the Portland Beds of England. Proc. Dorset nat. Hist. antiq. Fld.FId. Club 50,5o, 131-202.13 I-2o2. DAMON, R. 1884.I884. Geology of Weymouth, Portland and the coast of Dorsetshire (2nd ed.). Weymouth. DaUM~ONV,DRUMMOND, P. V. O. 1970.I97O. The Mid-Dorset Swell. Evidence of Albian-Cenomanian Movements in Wessex. Proc. Geol. Ass., IIx,81, 679-7i679-714. 4. DUNHAM, R. J. 1962.i962. Classification of carbonate rocks according to deposifionaldepositional texture. In W. E. Ham (ed.) Classification of carbonate rocks. Am. Ass. Petrol. Geol. Tulsa, Oklahoma, 108-io8--i3i.131 • FaX~DMAN,FRIEDMAN, J. M. & SANDERS, J. E. 1967. Origin and occurrence ofdolostones.of dolostones. InIn G. V. Chilingar, H. J. Rissell & R. W. FalrbridgeFairbridge (ed.), Carbonate rocks, origin, occurrence and classification. Elsevier, Amsterdam. 267-348267-348. . FORSICH,FURSICH, F. 1974. IehnogenusIchnogenus Rhizocorallium. Paldont.Paliiont. 7..Z·4 48,8, 1-23. I-~LLAM,HALLAM, A. 1969. TectortismTectonism and eustasy in the Jurassic. Earth Sd.Sci. Rev. 5,5,45-68. 45-68. HousE,HOUSE, M. R. I969.1969. Portland Stone on Portland. Proc. Dorset nat. Hist. archaeol. Soc. 9 x,1, 38--938-g.. KAUFFMAN, E. G. I969.1969. Form, function and evolution. InIn R. C. Moore (ed.),(ed.) , Treatise on In-In- vertebrate Paleontology, Part N, Mollusca 6. Bivalvia, 129-2o129-205. 5. I.~TT~R,LA'ITER, M. P. I926.1926. Petrology of thethe Portland Sand of Dorset. Proc. Geol. Ass. 37, 73-973-91. t- L~Es,LEES, G. M. & Cox, P. T. I937.1937. The geological basis for thethe search for oiloil inin Great Britain by thethe D'Arcy Exploration Co., Ltd. Q. J. geol.geol. Soc.Soc. Lond.£ond. 93, I56-9o.156-90. LOOAN,LOGAN, B. W., R~zAx,REZAK, R. & GINSBURG, R. N. 1964. Classification and environmentalenvironmental significancesignificance of algal stromatolites.stromatolites. J.J. Geol. 72, 68-83. N~'AVERSON,NEAVERSON, E. I925.1925. The petrology of thethe Upper Kimmeridge Clay andand Portland Sand in Dorset, Wiltshire, Oxfordshire andand Buckinghamshire. Proc.Proc. Geol. Ass.Ass. 36,36, 24o-56.240-56. NEWALL,NEWALL, N.N. D. & RIGBY, J. J. K. 1957. Geological studiesstudies on thethe Great BahamaBahama Bank.Bank. InIn R.R. J.J. Le Blanc & J.J. G. Breeding (ed.),(ed.), RegionalRegional aspectsaspects of carbonatecarbonate deposition.deposition. Soc.Soc. Econ.Econ. PaleontPaleont Min.,Min., Spec.Spec. Publ.Publ. 5, I5-22.15-22. Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021 638 w. G. Townson NICHOLS, D. 1969. Echinoderms. Hutchinson University Library, London. 192 pp. PuGH, M. E. 1968. Algae from the Lower Purbeck limestones of Dorset. Proc. Geol. Ass. 79, 513-33. PuRSER, B. H. 1969. Syn-sedimentary marine lithification of Middle Jurassic limestones in the Paris Basin. Sedimentology 12. -- 1973. Sedimentation around bathymetric highs in the southern Persian Gulf. In B. H. Purser (ed.), The Persian Gulf. Springer-Verlag, 157-77. RAISIN, C. A. 1903. Formation of chert and its microstructure in some Jurassic strata. Proc. Geol. Ass. 18, 71-82. RHOADS, D. & YOUNG, J. 1970. Influence of deposit-feeding organisms on sediment stability and community trophic structure. J. mar. Res. 28, 150-78. RUDWICK, M. J. S. 1970. Living and fossil brachiopods. Hutchinson University Library, London. Igg pp. SAMUEL, E. M. Ig6g. Formations exposed in the trench for the natural gas main in the area south-east of Bincombe. Proc. Dorset nat. Hist. archaeol. Soc. 91, 39-4 I. SCHMALZ, R. F. 1969. Deep water evaporite deposition: a genetic model. Bull. Am. Ass. petrol. Geol. 53, 798-82 3· SHEARMAN, D. J. 1966. Origin of marine evaporites by diagenesis. Trans. Inst. Min. Metall. B 208-15. STANLEY, S. Ig70. Relation of shell form to life habits of the Bivalvia (Mollusca). Mem. geol. Soc. Am. 125, 2g6 pp. STRAHAN, A. 18g8. Geology of the Isle of Purbeck and Weymouth. Mem. geol. Surv. U. K. 278 pp. TALBOT, M. R. Ig73. Major sedimentary cycles in the Corallian Beds (Oxfordian) of Southern England. Palaeogeogr. Palaeoclimatol. Palaeoecol. 14, 293-317. TAYLOR, J. C. M. & hLING, L. V. Ig6g. Holocene intertidal calcium carbonate cementation, Qatar, Persian Gulf. Sedimentology 12, 6g-107. TORRENS, H. S. (ed.) 1969. International Field Symposium of the British Jurassic. Excursion No. 1 Guidefor Dorset and south Somerset. Geology Department, Keele University. 71 pp. TOWNSON, W. G. 1971. Facies analYsis of the Portland Beds. Univ. Oxford D.Phil. thesis (unpubl.) 284 pp. -- 1974. Geology of Coombe Valley. Proc. Dorset nat. Hist. archaeol. Soc. 95, 7-8. WALTHER, J. 1893-4. Einleitung in die Geologie als historische Wissenschafl-Beobachtungen uber die Bildung der Gesteine und ihrer organischen Einschlusse. I, G. Fischer, Jena. WEST, I. M. 1964. Evaporite diagenesis in the Lower Purbeck Beds of Dorset. Proc. Yorks. geol. Soc. 34, 315-30 • -- 1965. Macrocell structure and enterolithic veins in British Purbeck gypsum and anhydrite. Proc. Yorks. geol. Soc. 35,47-58. --, SHEARMAN, D. J. & PUGH, M. E. 1969. Whitsun field meeting in the Weymouth Area, 1966, Proc. Geol. Ass. 80, 331-40. -- 1975. Evaporites and associated sediments of the basal Purbeck Formation (Upper Jurassic) of Dorset. Proc. Geol. Ass. 86 (in press). ZEISS, A. 1968. Untersuchungen zur Palaeontologie der Cephalopoden des Unter-Tithon der si..idlichen Frankenalb. Abh. Bayer. Akad. Wissensch. Math.-naturw. Kl. N.F. 132, 1-190. Received 9 January 1975; revised typescript received 7 March 1975. WILLIAM GEOFFREY TOWNSON, clo Shell U.K. Exploration & Production Ltd., Shell Centre, London SEI 7NA. Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/131/6/619/4885055/gsjgs.131.6.0619.pdf by guest on 30 September 2021