Content and licensingview original scan buy a printed copy

Geology of the country around Wells and Cheddar (Explanation of one-inch geological sheet 280, New Series)

By G. W. Green and F. B. A. Welch

Bibliographic reference: Green, G.W. and Welch, F.B.A. Geology of the Country around Wells and Cheddar (Explanation of One-inch Geological Sheet 280, New Series). London: HMSO, 1965

Department of Scientific and Industrial Research. Geological Survey of Great Britain. Memoirs of the Geological Survey of Great Britain England and Wales

Geology of the Country around Wells and Cheddar (Explanation of One-inch Geological Sheet 280, New Series) By G. W. Green, M.A., and F. B. A. Welch, B.Sc., Ph.D. with contributions byG. A. Kellaway, B.Sc., D. R. A. Ponsford, M.Sc., M. Brooks, B.Sc., Ph.D. (Geophysics) and M. Mitchell, M.A. (Palaeontology of the Carboniferous Limestone of Burrington Combe).

London: Her Majesty's Stationery Office, 1965 © Crown copyright 1965.

Preface

The Wells (280) Sheet covers much of the Mendip Hills and part of the Somerset Coalfield where the Armorican structures and the Mesozoic stratigraphy are of considerable scientific interest. In the limestone uplands of the Mendips, splendid sections of the Carboniferous Limestone Series are exposed in deep limestone gorges which include such picturesque areas as Burrington Combe and the Cheddar Gorge.

The Wells Sheet forms part of the region covered by the Old Series one-inch Sheet 19 which was geologically surveyed on the one-inch scale and published in 1845. Some of the geological features of the district were briefly mentioned by H. T. De la Beche in Vol. I of the Memoirs of the Geological Survey, published in 1846. Revision of Sheet 19 was undertaken between 1867–71 by H. W. Bristow, H. B. Woodward, W. A. E. Ussher, and J. H. Blake under the superintendence of H. W. Bristow and a 2nd Edition of the map was published in 1873. An explanatory memoir by H. B. Woodward of this and maps of adjoining districts—The Geology of the East Somerset and Bristol Coalfields—appeared in 1876. A 3rd Edition of Sheet 19, incorporating additional information collected during the preparation of J. Prestwich's report to the Royal Coal Commission of 1871, was published in 1899. All these maps were published hand coloured.

With the late Dr. R. W. Pocock as District Geologist, primary six-inch mapping of the New Series Wells (280) Sheet was started by Dr. F. B. A. Welch (1943–5) in the area of the Radstock Coal Basin and Mr. G. A. Kellaway (1943) in the Pensford Basin as part of a complete survey of the Somerset and Bristol Coalfields. In 1948, Dr. Welch and Mr. G. W. Green commenced work in the Wrington–Blagdon area under the superintendence of the late Mr. H. G. Dines and, with the assistance of Dr. R. Beveridge (1951) and Mr. D. R. A. Ponsford (1954–5), the work was extended to cover the whole sheet. The survey was completed in 1955 under the supervision of Dr. Welch, and the one-inch map was published colour printed in 1964.

Full details of the stratigraphy and structure of the Coal Measures are not given in this memoir as these will form part of a forthcoming memoir dealing with the Bristol and Somerset Coalfields as a whole.

We gladly acknowledge the courteous assistance of many individuals and firms, in particular that of the Bristol Waterworks Co. and of Messrs. F. G. Clements (Easton) Ltd. in supplying information on boreholes, trial pits, quarries and temporary sections, or in giving facilities for their examination by the surveyors themselves. Detailed acknowledgements are made in the text.

During the survey the main collection of fossils, additional to those collected by the surveyors, was obtained by Mr. S. W. Hester (Carboniferous) and Messrs R. V. Melville and F. G. Dimes (Mesozoic). Carboniferous fossils were studied and named by the undersigned, when Chief Palaeontologist, and Mr. M. A. Calver (Coal Measures), by Dr. W. H. C. Ramsbottom and Mr. M. Mitchell (Carboniferous Limestone and Millstone Grit); Mr. R. V. Melville and Dr. H. Ivimey-Cook named the Mesozoic fossils. For specialist assistance in naming the ammonites, thanks are given to the late Dr. L. F. Spath, F.R.S., to Professor D. T. Donovan (Lower Lias), and to Dr. M. K. Howarth (Upper Lias); also to Dr. D. V. Ager for naming some of the Lias brachiopods, and to Dr. W. G. Chaloner for similar service with the Coal Measures plants. Mr. Mitchell and Mr. Green, in an appendix to this memoir, discuss the faunal succession of the Carboniferous Limestone of Burrington Combe and some of its implications.

Some petrographical descriptions have been provided by Mr. R. W. Elliot; others have been written by Professor K. C. Dunham (when Chief Petrographer) and by Dr. R. Dearnley. Mineralogical determinations of baryte, celestine, and barytocelestine have been made by Mr. E. A. Jobbins. Mr. G. A. Sergeant has supplied a chemical analysis. A chapter on geophysical surveys in the area, contributed by Dr. M. Brooks, includes details of recent surveys by the Geophysical Department.

The compiling of the memoir was undertaken by Mr. Green and the editing by Mr. Kellaway. It has been seen through the press by Dr. J. R. Earp.

C. J. Stubblefield, Director,Geological Survey Office, Exhibition Road, South Kensington, London, S.W.7 March, 1964

(Front cover) View of Cheddar Gorge showing cliffs cut in Carboniferous Limestone (A9759).

(Rear cover)

List of six-inch maps

The following is a list of the 6-inch geological maps included, wholly or in part, in the 1-inch Wells (280) Sheet, with the initials of the surveyors and dates of survey. The names of the officers are as follows: R. Beveridge, G. W. Green, G. A. Kellaway, D. R. A. Ponsford, F. B. A. Welch.

The geological 6-inch ‘county’ sheets listed above, will be reconstituted on national grid lines, and the following sheets, including those marked with an asterisk which are to be published, will be available for reference in the Geological Survey and Museum Library

Wells and Cheddar—summary

The country around Wells and Cheddar includes much of the Mendip Hills, parts of the Somerset Coalfield and a portion of the Somerset 'levels'. Geologically it is best known for the splendid sections of the Carboniferous Limestone exposed in picturesque gorges such as Burrington Combe and Cheddar. The well-marked structures of the Armorican (Permo-Carboniferous) earth movements provide clear illustrations of the relationships between folds, various classes of faults and differing rock types. In Mesozoic times, extensive downwarping (Central Somerset Basin) south of the Mendips is combined with striking unconformities and non-sequences over the Mendip Hills themselves.

This memoir is an authoritative account of the geology of the area, written by the geologists who surveyed on the six-inches to one mile scale the Wells (280) sheet of the Geological Survey series of one-inch maps ; this sheet was published in 1964.

(Front cover) Jacket photograph : View of Cheddar Gorge showing cliffs cut in Carboniferous Limestone (A9759)

Other publications dealing with this and adjoining districts

Books

• British Regional Geology
• Bristol and Gloucester, 2nd Edition, 1948
• Obtainable from the Geological Museum Bookshop, Exhibition Road, South Kensington, London, SW7 2DE; Government Bookshops in London (post orders to PO Box 569, SE1 9NH), Edinburgh, Cardiff, Belfast, Manchester, Birmingham and Bristol, or through booksellers

Geological maps

• 'Ten mile' map of Great Britain (1 :625 000) (South Sheet)
• 3rd Solid Edition, 1977
• 1st Quaternary Edition, 1977
• One-inch to one mile (1 :63 360)
• Bristol District Sheet,Solid and Drift Edition, 1962
• Wells (280) Sheet, Solid and Drift Edition,1964
• Frome (281) Sheet, Solid and Drift Edition, 1965
• 1:50 000 Series
• Bristol (264) Sheet, Solid and Drift Edition, 1974
• Weston-super-Mare (279) Sheet, Solid and Drift Edition, in press
• 1:25 000 Series
• ST 45 Cheddar, Solid and Drift Edition, 1969
• ST 47 Clevedon and Portishead, Solid and Drift Edition, 1968
• 1 :10 560 Series
• Published six-inch National Grid geological maps of the Bristol and Somerset Coalfield areas are available from Ordnance Survey agents. In addition manuscript copies covering these and adjacent areas.are available for public inspection in the library at the Geological Museum, London SW7 2DE. Dyeline prints may also be obtained.

Geophysical maps

• 'Ten mile' Aeromagnetic map of Great Britain (1 :625 000) Sheet 2. Obtainable from the Geological Museum Bookshop and all Ordnance Survey Agents
• Diagram Edition (1 :250 000) Sheet 2 Aeromagnetic map Southern England and English Channel. Obtainable only from the Geological Museum Bookshop

Chapter 1 Introduction

Geographical features

The one-inch New Series Wells (280) Sheet covers a mainly agricultural region of North Somerset which possesses a great diversity of scenery ranging from extensive tracts of former swamp and marsh land at or near sea level to craggy limestone uplands with dry valleys and deep gorges. The largest centres of population are the small industrial town of Shepton Mallet and the old cathedral city of Wells. Hamlets and villages are built of various local building stones, and are often seen in attractive rural settings. This combination of scenery and architecture attracts many thousands of visitors each year.

The dominant topographic feature is the high steep-sided plateau known as the Mendip Hills, or Mendips, that stretches from north-west to south-east diagonally across much of the district. This tableland has an average height of about 800 ft above O.D.; it rises to 1000 ft at North Hill and Pen Hill and attains its greatest elevation at Black Down (1068 ft). At four other places the ground level reaches heights of between 940 and 980 ft O.D. West of Black Down the central part of the tableland has been eroded away and only the outer edges survive as a series of isolated ridges.

Situated to the north of the Mendip Hills there is a much-dissected plateau lying at an altitude of about 450 to 550 ft O.D. This forms the higher tracts of land extending from Chewton Mendip and Ston Easton to Hinton Blewett and High Littleton, and is separated by the wide valley of the River Chew from the flat-topped hills around Butcombe and Nempnett Thrubwell. West of a low watershed between Compton Martin and Nempnett Thrubwell the River Yeo has excavated a deep valley along the northern side of the Mendips, which widens westwards and finally merges into the alluvial flats of Banwell Moor.

To the south of the Mendip Hills and west of Wells lies the wide alluvial area of the Somerset 'moors' with a surface level of only 10 to 20 ft above O.D. This expanse of 'moor' is broken only by the low 'Isle' of Wedmore and the chain of low, steep-sided hills that extend eastwards through Theale and Farley to Wells and thence north-westwards towards Cheddar a short distance from the foot of the Mendip Hills. Between Wells and Shepton Mallet, to the south of the main Mendip hill-mass, the ground is broken into a series of small steep-sided ridges and hills.

Drainage

The main watershed is formed by the Mendips; on the south side the drainage is shared by the River Axe, with its tributaries the Yeo and Lox Yeo River, and by the Brue, with its tributary the Sheppey, all of which flow westwards into the Bristol Channel. North of the Mendips direct westerly drainage into the Bristol Channel is effected by the River Yeo, whilst farther to the east the rivers Chew and Cam drain north-westwards and flow into the Bristol Avon, thus reaching the Bristol Channel by a more circuitous route.

As would be expected in hills made largely of limestone, the drainage of the Mendip plateau itself is almost entirely underground. Rainfall on the limestone outcrop sinks directly into the ground and thence through a complex fissure system to the water table hundreds of feet below. The Old Red Sandstone and Lower Limestone Shale outcrops are drained by short streams which disappear underground through swallets when they reach the limestone outcrop. Part of this great underground reservoir of water is held back by impermeable strata along the sides of the Mendips, while the remainder emerges as streams and rivers from the foot of the hills to take a normal overground course out to sea.

Industries

Whilst some arable farming is carried on, the greater part of the land is devoted to dairy farming and grazing. On the well-drained south-facing slopes at the foot of the Mendips between Axbridge and Draycott intensive market gardening is practised. Here the soils are derived largely from Head consisting of stony loam or from weathered Dolomitic Conglomerate. The springs issuing in the low ground at the foot of the Mendips have not only been major factors in the siting of numerous villages and the towns of Wells and Shepton Mallet, but also for the establishment of a paper mill at Wookey and, in the past, for silk mills at Shepton Mallet. Much of this water is now impounded in the three large reservoirs of the Bristol Waterworks Co. which serve areas far beyond the confines of the Wells Sheet. Of heavy industries the most important one is the quarrying of limestone for roadstone and lime-burning. In the small part of the Somerset Coalfield which lies within the district two collieries are in work^‡1 . G.W.G., F.B.A.W.

Geological sequence

The formations shown on the Wells (280) Sheet and described in this memoir are as follows:—

Geological history

The oldest rocks exposed in the district depicted on Sheet 280 are lavas and tuffs of Upper Llandovery age in the Beacon Hill inlier, north-east of Shepton Mallet. Vulcanism was widespread for the nearest Silurian outcrop, some 30 miles to the north at Tortworth, includes similar lavas. This volcanic episode had apparently ended by Wenlock times as the succeeding formation in the Beacon Hill inlier consists of normal marine shales and sandstones (on Frome Sheet 281). The absence of deposits of Ludlow–Lower Old Red Sandstone age in the southern part of the district^‡4  may be due partly to non-deposition but mainly to erosion. In the adjacent areas to the north the deposition of the Lower Old Red Sandstone was terminated by the Caledonian earth-movements, which resulted in widespread uplift, accompanied by folding and faulting and prolonged erosion. The evidence for these events on the Wells Sheet is confined to the Beacon Hill area where the Upper Old Red Sandstone rests with angular discordance on different members of the Upper Llandovery volcanic sequence.

In the long period commencing with the deposition of the Upper Old Red Sandstone and lasting until the Armorican earth-movements at the close of the Carboniferous a thick series of essentially conformable strata was laid down in the Wells area. Sedimentation was discontinuous in the north-western part of the area where the Millstone Grit is thin or absent and the Pennant Series is thought to rest unconformably on the Carboniferous Limestone Series; elsewhere there is little evidence of major unconformity. The thick predominantly arenaceous sediments of the Portishead Beds represent rock waste derived from the erosion under semi-arid conditions of a land mass to the north and laid down in a partly enclosed sea. At the end of Old Red Sandstone times, continued submergence led to the spread of the open sea, which previously lay to the south, across the whole area. The muddy water in which the Lower Limestone Shale was laid down was later replaced by the shallow clear seas in which the main mass of the Carboniferous Limestone accumulated. Subsequent earth-movements resulted first in the Carboniferous seas becoming charged with sand and mud, which was laid down to form the Quartzitic Sandstone Group, and then to the emergence of the area to form part of the vast swamps and deltas of Coal Measures times.

Upper Coal Measures sedimentation was terminated by the Armorican (or Hercynian) earth movements which raised up chains of mountains of great structural complexity stretching across Europe and, although the areas of most intense folding lay to the south, the major structures in the Mendips themselves bear witness to the magnitude of the movements. At about this time a profound climatic change to semi-arid conditions took place and throughout the ensuing Permian and much of the Triassic period’s prolonged erosion resulted in the removal of immense thicknesses of strata.

The earliest Triassic deposits seen in the area are the Dolomitic Conglomerate and Keuper Marl which represent the detritus derived during the last phases of the wearing down of the Armorican mountains. The Keuper deposits to the north of the Mendips are relatively thin in contrast to those south of the Mendips, which are of considerable thickness. This difference is because in Keuper times the sediments did little more than infill the existing topography north of the Mendips whereas sedimentation in the Central Somerset Basin to the south was accompanied by subsidence. There followed a change to more humid conditions heralded by the striking colour change from the Keuper red marls to the Tea Green Marl and the Triassic period was brought to a close by the widespread and rapid marine transgression represented by the Rhaetic deposits. In Rhaetic times the highest parts of the Mendips remained as islands and shoals, and later, in Liassic times, these parts suffered intermittent emergence and submergence and it was not until Upper Inferior Oolite times that complete submergence took place. Meanwhile, to the north of the Mendips lay an unstable region of marine shallows, the Radstock Shelf, in which a thin Liassic sequence, showing much evidence of condensation and non-sequence, was deposited. To the south of the Mendips the steady development of the Central Somerset Basin gave rise to an unbroken succession of argillaceous rocks of great thickness.

The shallow water limestones of the Upper Inferior Oolite, the youngest rocks seen, bear witness to the establishment of virtually uniform sedimentary conditions throughout the area. By analogy with adjacent areas marine sedimentation may be presumed to have continued throughout the remainder of Jurassic and Cretaceous times, broken only by widespread warping, accompanied by faulting and erosion (possibly submarine), in pre-Albian times. Earth movement of Tertiary age uplifted and folded the district and during a prolonged period of erosion the present river system was initiated.

The lack of detailed evidence in the district precludes the exact delineation of events in Pleistocene times. In the glacial episodes which characterize those times this district lay to the south of the main ice-sheets although the higher ground may have supported small snowfields. Extensive head deposits were formed. During Pleistocene times there were considerable fluctuations of sea level as shown by the presence of marine sands which at the present time occur from about 30 ft above O.D. down to at least 100 ft below O.D.

By the end of the last Glacial episode the sea level had sunk to at least 100 ft below O.D. and the main rivers were entrenched in channels graded to this low level. The subsequent geological history is one of oscillatory, though mainly rising, sea levels with concomitant building up of the wide alluvial areas which now constitute the Somerset levels. G.W.G.

Chapter 2 Silurian

Volcanic rocks of Silurian age, occurring in the core of the Beacon Hill Pericline to the north-east of Shepton Mallet, form part of a much larger outcrop which extends eastwards into the Frome (281) Sheet.

A review of past research on these rocks has been published by Curtis (1955). They were first recognized by Charles Moore (1867), who regarded them as a basaltic dyke, and were first described as andesite by Teall (in Geikie and Strahan 1899). Later, Reynolds (1907) made a detailed field study of the rocks and showed them to be an extensive series of thick andesites with interbedded tuffs. Fossils found in the lowest exposed tuff beds were considered by Cowper Reed (in Reynolds 1907, p. 227) probably to be of Upper Llandovery age. Later, Reynolds's important discovery (1912) of normal marine mudstones and sandstones of Wenlock age overlying the volcanic rocks gave more precision to the upper age limit of the Silurian volcanic activity. Curtis (1955) has pointed out that the fauna from the tuffs could be of either Upper Llandovery or Wenlock age. The recent survey (of Frome Sheet 281) has confirmed Reynolds's suggestions (1912) that the upper series of Wenlock strata both overlie the volcanic sequence with angular discordance and also include no tuffaceous rocks (Green 1962, p. 29). These facts suggest that, not only had volcanic activity ceased by the time the (undoubted) Wenlock strata were being laid down, but also that a considerable time had elapsed between the outpouring of the lavas and the deposition of the Wenlock sediments. The volcanic activity of the Mendips is only matched in Silurian times elsewhere in Great Britain (see Wills 1951, pl. iii) at Tortworth, about 30 miles to the north of the Mendips, where andesitic lavas (The Upper Trap) are interbedded with rocks of Upper Llandovery age (Curtis 1955). It, therefore, seems reasonable to conclude that the volcanic rocks of the Beacon Hill Inlier are also of Upper Llandovery age.

The Silurian volcanic rocks are well exposed in large quarries south of Stoke Lane just beyond the eastern margin of Sheet 280 where the following succession has been seen (in ascending sequence)—fossiliferous tuffs about 120 ft, andesitic lavas, including an impersistent tuff near the base, about 150–200 ft, tuff about 35–40 ft, andesitic lavas (Main Andesite Group) about 500 ft, tuff and volcanic conglomerate about 100 ft to over 150 ft, andesitic lavas seen to 150 ft (Green 1962, p. 29). This succession can be traced westwards to the line of a north-easterly trending fault which cuts across the inlier east of the Three Ashes–Doulting road (Wells Sheet 280). West of this fault the geological structure of the inlier is not known in comparable detail owing to a dearth of exposures. A wide central area of andesite is flanked on the north and south by narrower outcrops of tuffaceous rocks. Most of the andesite probably belongs to the Main Andesite Group while the northern outcrop of tuffs apparently forms the westerly continuation of the variable series of tuffs overlying the Main Andesite Group in the Stoke Lane area (Frome Sheet 281). Reynolds (1907) has suggested that the volcanic conglomerate south of Beacon Hill is part of a volcanic neck but to the writer it appears it may be merely very coarse tuff such as is seen in the Moon's Hill area [ST 663 462] The stratigraphic relations of the tuff on the southern side of the inlier south-west of the Inn (Waggon and Horses) with the andesite to the north are not known. Reynolds (1907) correlated this tuff with that underlying the andesite at Sunnyhill. An alternative explanation is that the andesite at the 'Waggon and Horses' is anticlinally folded, the pitch of the fold being towards the west. In this case the tuff exposed on the northern and southern sides of the inlier are one and the same bed, and therefore overlies the andesite.

The continuation at depth of the Silurian volcanic rocks both to the north and south of the Beacon Hill Inlier is inferred from the results of geophysical surveys (see pp. 154–9).

For petrographical generalizations on the rock types it is necessary to refer to the well-exposed Stoke Lane area (Frome Sheet 281). Dr. R. Dearnley reports on the Main Andesite Group of Moon's Hill Quarry [ST 663 461] as follows:

Pyroxene–andesite. A compact medium to fine-grained volcanic rock. Phenocrysts of augite occur, partially or completely replaced by chlorite or serpentine. Plagioclase (oligoclase–andesine) phenocrysts are abundant and locally form glomeroporphyritic groups; the feldspar is partially altered to sericite and clay minerals. The groundmass of the andesite consists of flow banded microliths (up to 0.01 mm) of oligoclase, granules of pyroxene, usually altered to chlorite and abundant magnetite altering to haematite.

Details

At the western end of the Silurian outcrop a recent trench along the Fosse Way was examined by Dr. M. L. K. Curtis who reported (in litt.)that, although most of the trench was cut in Old Red Sandstone detritus, at a place 350 yd from the Wells–Frome road, andesite overlain by tuffaceous debris was noted. In this area Reynolds (1907, fig. 5, p. 229) reported coarse ashy conglomerate in a series of trial pits dug across the Old Red Sandstone contact and continued along the line of the rifle range to a place 250 ft to the south. A little farther to the east, the same rock was reported (Woodward and others 1909, p. 287) in two trial pits approximately 650 yd. N. 35°E. and 950 yd N.E. of Beacon Farm. Between this area and the Inn (Waggon and Horses) the greater part of the outcrop apparently consists of andesite, for though the rock is nowhere seen in situ, surface debris of andesite is abundant. A borehole [ST 649 547] for water 100 yd north-west of the Inn (Waggon and Horses) proved 'rock' to 95 ft, 'rock and clay' from 95 to 110 ft and 'rock' to 139.5 ft. The 'rock and clay' may represent tuff. On the south side of the inlier a pit [ST 647 455] 350 yd W. 27°S. of the Inn exposed 6 ft of rubble composed of reddish coloured, medium-grained, rather hard tuff of a type known from several parts of the succession farther east.

About 200 yd east of the Waggon and Horses Inn a well-marked north-easterly trending depression is apparently related to a fault which cuts both Silurian and Old Red Sandstone rocks. Discontinuous exposures in a ditch 650 yd east-north-east of the Inn show hard, brown, medium-grained, gritty tuff with traces of indeterminate fossils overlying soft yellowish tuff with poorly preserved casts of Orthids. The uppermost beds apparently dip at about 35° to 40° to the south-west and form a low feature sub-parallel to the strike of the Old Red Sandstone outcrop to the south but separated from it by a narrow depression (due to softer tuff?). The total thickness of the strata may be 120 to 150 ft. This is the westernmost exposure of the basal fossiliferous tuffs (Upper Llandovery) similar to those described by Reynolds (1907, pp. 222–225) at Sunnyhill Quarry (Sheet 281). About 150 to 200 yd to the north and north-east (partly on Frome Sheet 281) another ditch section showed discontinuous exposures in the overlying andesites and an associated tuff band. G.W.G.

References

CURTIS, M. L. K. 1955. A review of past research on the Lower Palaeozoic rocks of the Tortworth and Eastern Mendip inliers. Proc. Bristol Nat. Soc., 29, 71–8.

GREEN, G. W. 1962. in Sum. Prog. Geol. Surv. for 1961, 29.

MOORE, C. 1867. On Abnormal Conditions of Secondary Deposits when connected with the Somersetshire and South Wales Coal-Basin; and on the age of the Sutton and Southerndown Series. Quart. J. Geol. Soc., 23, 449–568.

REYNOLDS, S. H. 1907. A Silurian Inlier in the Eastern Mendips. Quart. J. Geol. Soc., 63, 217–38.

REYNOLDS, S. H. 1912. Further Work on the Silurian Rocks of the Eastern Mendips. Proc. Bristol Nat. Soc. (4), 3, 76–82.

TEALL, J. J. H. in Geikie, A. and Strahan, A. 1899. Volcanic Group in the Carboniferous Limestone of North Somerset. Sum. Prog. Geol. Surv. for 1898, 110–1. WILLS, L. J. 1951. A Palaeogeographical Atlas of the British Isles and adjacent parts of Europe. London.

WOODWARD, H. B., REYNOLDS, S. H., MORGAN, C. L. and WINWOOD, H. H. 1909. Investigation of the pre-Devonian Rocks of the Mendips and the Bristol Area. Rep. Brit. Assoc. for 1908, 286–91.

Chapter 3 Old Red Sandstone

Upper Old Red Sandstone, Portishead Beds^‡5 

The Old Red Sandstone crops out in the cores of the Blackdown, the North Hill, the Pen Hill, and the Beacon Hill periclines. These elongated and elliptical areas of sandstone form the highest ground on the Mendips with a general summit level of about 1000 ft above sea level and rise as smooth gently contoured hills about 200 ft to 300 ft above the level of the surrounding plateau occupied by the Carboniferous Limestone. When undisturbed by cultivation the Old Red Sandstone gives rise to dark and often peaty soil. Exposures, which are both poor and infrequent, are mainly confined to stream sections.

The dominant rock type is dull reddish feldspathic and usually quartzitic sandstone, interbedded with sandy green, red and purplish shales, fine-grained micaceous sandstones and, in the lower part of the succession, with pebbly sandstones and sandy quartz-conglomerates. Scattered grains of low-grade metamorphic and acid igneous rocks appear to occur throughout much of the succession (see below). The rocks are non-calcareous below about the top hundred feet. The sandstones are grey in colour in many places in the upper parts of the succession, elsewhere they are dingy red and purple due to the inclusion of haematite and, to a less extent, chlorite and clay minerals. The presence of fresh feldspars, the ubiquitous red, purplish and greenish colouration, the relatively poor sorting, the sharp lateral variations in the rocks and the lack of marine fossils have usually been taken as indications that the Old Red Sandstone rocks were derived from the denudation, under rather arid conditions, of surrounding high ground and deposited in shallow inland or partly enclosed seas.

Fossil remains in the Old Red Sandstone of the district under review are rare; a fish-bearing conglomerate has been observed in the top beds at Burrington while layers of plants at the same level may give rise to thin coal seams as, for instance, in the Black Down area.

In the Mendip area, the Old Red Sandstone appears to form one thick conformable series of sediments which are coarser towards the base. It has been correlated with the Portishead Beds (Kellaway and Welch 1948, p. 17 and 1955, p. 6) of the Bristol area which are of Upper Old Red Sandstone age. The fullest successions are exposed in the Blackdown Pericline where more than 1600 ft^‡6  of beds are present and the Beacon Hill Pericline where about 1350 ft of Portishead Beds rest unconformably upon Silurian rocks. It will be noted that there is a marked increase in thickness of the Upper Old Red Sandstone rocks southwards from the Forest of Dean, where they are 350 ft to 400 ft thick (cf. Kellaway and Welch 1955, pl. i).

In the Beacon Hill Pericline Mr. Ponsford has established the following succession: about 350 ft of red sandstone with pebbly and conglomeratic bands prominent in the basal part, overlain by the main sandstone block, comprising some 750 ft of red sandstone with thin mudstones. At the top are about 250 ft of grey sandstones with thick red and grey mudstones. A similar succession can be recognized in the Old Red Sandstone of the Blackdown Pericline.

Details

Black Down

An exposure [ST 469 577] 770 yd south of Read's Cavern showed quartzitic sandstone, with small white quartz pebbles, dipping at about 45° northwards. About half a mile to the east discontinuous exposures of quartzitic sandstone, conglomeratic in the basal portion, can be seen in the valleys of the two parallel streams leading into Burrington Combe. The dips are northerly at 40° to 50°. In an adit [ST 475 582] 1280 yd S. 15°W. of Burrington church, the following succession of upper beds of the Old Red Sandstone was recorded by Hepworth and Stride (1950):

Exposures in the stream banks about 0.75 mile south-south-east of Lower Ellick Farm show red and yellowish well-bedded rather fine-grained sandstone overlain by greenish-brown shale dipping at 40° to 45° to the north-north-east.

On the south side of the Blackdown Pericline, discontinuous exposures of Old Red Sandstone occur in the stream [ST 490 563] some 550 yd north-east of Lower Farm, Charterhouse, where 5 ft of massive quartzitic sandstone is overlain by about 50 ft of fine-grained, micaceous sandstone. An excavation [ST 462 575] 880 yd N. 30°E. of Longbottom Farm, in the lowest beds exposed in the pericline, revealed strongly disintegrated conglomerate and reddish sandstone. A thin section of the latter, (E24602)^‡7 , has been described by Mr. R. W. Elliot as follows: 'A purple medium-grained sandstone containing rounded subangular and angular grains of quartz, granulitized quartz, microcline, orthoclase, chloritized and haematitized intermediate volcanics, mudstone and chert. The quartz grains show silification. Haematite occurs as large irregular patches, grains and dusting some of the rock fragments and matrix.' The pebbles from the conglomerate consisted of dense reddish quartzitic sandstone and vein quartz, and rare pinkish quartz-porphyry and rhyolite (see (E24605)–(E24606)).

Another exposure of these lower beds is afforded by the crags [Holloway Rocks, [ST 457 575] 820 yd N. 15° W. of Longbottom Farm, where reddish quartzitic sandstone dipping 40° to 47° to the south-south-west underlies Triassic conglomerate which dips at 10° to the north-north-east.

At a point [ST 443 572] 200 yd south-south-west of Shipham church reddish-purple marl and micaceous sandstone dipping at 38° to 40° to the south are exposed in the roadside. G.W.G.

Shipham–Winscombe

In a water bore [ST 436 567] 240 yd south-west of Winterhead Hill Farm 150 ft of fine-grained brown sandstone were proved. At the western end of Hale Coombe [ST 427 567] hard fine-grained sandstone with occasional green marl inclusions and associated bands of green and purple shale dip at 36° to the south.

To the west and south of Winscombe hard fine-grained sandstone crops out to form the bracken-covered hills of Church Knoll and Broad Knoll. F.B.A.W.

Cheddar

About 1 mile south-east of Cheddar, near Carscliff Farm, there is a small inlier of Palaeozoic sandstone surrounded by Dolomitic Conglomerate (Triassic). The outcrop is marked by abundant surface debris of reddish quartzitic sandstone, while dry stonewalls in the area are built of blocks of the same rock obtained from shallow workings (no longer exposed). Originally mapped by the Survey as Old Red Sandstone, subsequent authors (Lloyd Morgan, 1890, pp. 178–9; Welch 1929, p. 63) have regarded the rocks as Millstone Grit (see also p. 138 below). Thin rock slices (E25058), (E27890) have been described by Mr. R. W. Elliot as 'medium-grained feldspathic sandstone'. He further reports that secondary enlargement of both the quartz and the feldspar has occurred, and that the rock fragments include fine-grained acid igneous rocks. The quartzitic sandstones of Millstone Grit age from the Mendips, which have been examined in thin section (see p. 52 below) lack the mineralogical variety of both the Old Red Sandstone and the rocks from the Carscliff Inlier. In conclusion, Mr. R. W. Elliot says that 'the lithological evidence though by no means conclusive, would favour a correlation of the Carscliff rocks with rocks of Old Red Sandstone age rather than Millstone Grit'. Dr. R. Dearnley adds that the presence of occasional grains of low-grade metamorphic rocks in the thin sections (cf. Old Red Sandstone, Beacon Hill and elsewhere—see below) supports this conclusion.

Borehole evidence indicates that there must be a considerable area of mudstone and sandstone underlying the Triassic rocks to the south and south-west of the inlier. A recent borehole [ST 4664 5202] about 0.5 mile to the south-west of the inlier proved basal Triassic breccias, apparently devoid of Carboniferous Limestone debris, resting on purplish silty micaceous mudstone^‡8  with two thin purplish-red sandstone bands (see p. 201). Dr. R. Dearnley reports that the sandstone in thin section (E31232)–(E31233) resembles that from the inlier. A further borehole [ST 473 516] in Triassic rocks about 0.5 mile to the south of the inlier (see p. 202) ended in breccia containing abundant very large fragments of sandstone and mudstone similar, respectively, to that seen in the inlier and the borehole previously mentioned.

The evidence given above favours the view that the Carscliffe Inlier is Old Red Sandstone rather than Millstone Grit^‡9 . This view also accords with the structural pattern of the northern end of the Cheddar–Wells Thrust Belt (see p. 138).

North Hill–Priddy

A borehole [ST 538 513] 1000 yd E. 8° S. of Priddy church, starting in Lower Limestone Shale, proved Old Red Sandstone between depths of 190 ft and the bottom at 300 ft. The strata consisted of grey, pink and brown calcareous sandstones with bands up to 2 in thick of conglomerate, and occasional beds up to 2 ft thick of brick-red and greyish-green shale and maroon micaceous shales. The dip of the overlying Lower Limestone Shale is recorded as 60° (Whittard 1949, p. 481). A roadside section [ST 547 515] 1850 yd south-south-east of the Castle of Comfort Inn, in the oldest rocks exposed in the North Hill Pericline, shows quartz conglomerates. A borehole [ST 544 524] 0.5 mile south of the Castle of Comfort Inn, which was sunk to a depth of 150 ft in Old Red Sandstone, is stated by Dr. F. S. Wallis (in litt.)to consist of 'the usual type of red sandstone, marls and conglomeratic sandstones'. On the eastern side of the pericline, hard red sandstone, near the top of the sequence, is seen dipping 28° to 30° to the north-east at a point [ST 576 517] 800 yd W. of Everard's Farm. Sandstone and vein material thrown out from numerous trial shafts is seen about mile west-north-west of Rookery Farm (Green 1958, p. 74). A specimen of this sandstone estimated to come from about 200 ft below the top of the Old Red Sandstone, was examined in thin section (E26575) by Mr. R. W. Elliot who reported as follows: 'A medium-grained sandstone containing angular and subangular grains of quartz, granulitized quartz, orthoclase, micropegmatite, microperthite, microcline, albite-oligoclase and fine-grained acid igneous rock and mudstone. The quartz grains may show secondary growth. Haematite quite often rims individual grains.

Pen Hill

At the western end of the Pen Hill Pericline, 1.33 miles north-north west of Wells Cathedral the Rookham Borehole (see pp. 205–7) showed that the upper part of the Old Red Sandstone included much veryhard light grey dolomitic sandstone. G.W.G.

In two old quarries [ST 568 480] on Prior's Hill, by the side of the Bath–Wells main road, hard, red-brown, quartzitic sandstone dips at 30° to 35° to the south-southeast. About 0.5 mile to the north, in the core of the pericline, an exposure [ST 568 487] showed nearly vertical and probably inverted sandstone. In the bottom of the steep valley (Biddle Combe) to the east of the main road a water bore [ST 469 485], starting within some 300 ft of the top of the Old Red Sandstone, proved sandstone with shale bands of considerable aggregate thickness to a depth of 402 ft (Richardson 1928, p. 122). A trial for water [ST 587 488] 1060 yd west-north-west of Whitnell Corner showed 15 ft of purple sandstone with red shaly partings. F.B.A.W.

Beacon Hill

The Portishead Beds outcrop on the northern limb of the Beacon Hill Pericline from a point west of Maesbury Castle eastwards to beyond the limits of Sheet 280. In the western part dips range from 16° to 40° but east of the Fosse Way dips of 60° to 65° are seen. The basal sandstones and conglomerates are exposed on the southern side of Beacon Hill where they form a prominent feature. A specimen of reddish-brown banded sandstone and conglomerate collected from an outcrop [ST 635 459], 850 yd north-north-west of Beacon Farm, was examined in thin section (E30220A) by Dr. R. Dearnley who reports as follows: 'A red poorly-sorted conglomeratic sandstone with grain size variable between 0.2 mm and 4 cm. The sub-rounded fragments are set in a matrix of finer-grained sandstone, with a small amount of silica cement. The fragments include vein quartz, quartz-schist, devitrified rhyolite showing rare spherulitic structure, silicified? tuff, rare mudstone fragments, abundant foliated, crushed and sutured quartzite grains, muscovite and chlorite schists, and some perthitic feldspar and micrographic intergrowths of quartz and feldspar. A few grains of micaceous schistoze grit and quartz grains with included vermicular chlorite also occur.' Dr. Dearnley comments that this assemblage of low-grade metamorphic rocks, quartz with vermicular chlorite and acid volcanics and tuffs (see also (E24605), (E24606), (E26575) described above) is similar to that described by Wallis (1928) from the Old Red Sandstone of the Bristol area.

The red sandstones (see p. 11) overlying the basal conglomeratic beds in the Beacon Hill area were penetrated by two boreholes to 134 ft and 150 ft near Warren Farm, and are seen in trial pits [ST 631 463], [ST 633 461] on the western side of Beacon Hill where they dip north-westwards at 40°.

Apart from an inlier north-west of Burnthouse Farm and some very small inliers in valley bottoms the Old Red Sandstone of the southern limb of the pericline is covered by Lias. A borehole [Stomacher Farm Borehole, [ST 625 458] 1250 yd south-south-west of Oakhill church, starting in Lias limestone (4 ft) proved 103 ft of thick marl and sandstone overlying 87 ft of sandstone with some marl (the main sandstone group, see p. 11). D.R.A.P.

References

GREEN, G. W. 1958. The Central Mendip Lead-Zinc Orefield. Bull Geol. Surv. Gt. Brit., No. 14, 70–90.

HEPWORTH, J. V. and STRIDE, A. H. 1950. A sequence from the Old Red Sandstone to Lower Carboniferous, near Burrington, Somerset. Proc. Bristol Nat. Soc., 28, 135–8.

KELLAWAY, G. A. and WELCH, F. B. A. 1948. Bristol and Gloucester District, 2nd edit. British Regional Geology, Geol. Surv.

KELLAWAY, G. A. and WELCH, 1955. The Upper Old Red Sandstone and Lower Carboniferous Rocks of Bristol and the Mendips compared with those of Chepstow and the Forest of Dean. Bull. Geol. Surv. Gt. Brit., No. 9, 1–21.

MORGAN, C. LLOYD 1890. Mendip Notes. Proc. Bristol Nat. Soc. (3), 6, 169–82.

RICHARDSON, L. 1928. Wells and Springs of Somerset. Mem. Geol. Surv.

WALLIS, F. S. 1928. The Old Red Sandstone of the Bristol District . Quart. J. Geol. Soc., 83 for 1927, 760–89.

WELCH, F. B. A. 1929. The Geological Structure of the Central Mendips. Quart. J. Geol. Soc., 85, 45–76.

WHITTARD, W. F. 1949. Temporary Exposures and Borehole Records in the Bristol Area: IV. Boreholes on Mendip. Proc. Bristol Nat. Soc., 27, 479–82.

Chapter 4 Lower Carboniferous

Carboniferous Limestone Series

The Carboniferous Limestone Series, totalling some 3000 to 3700 ft in thickness, has an outcrop of about 50 square miles and gives rise to some of the highest ground and the most striking scenery in the district. Formerly a significant repository of lead ore, the Carboniferous Limestone of the Mendips is now an important source of limestone of which about a million tons are quarried annually.

A twofold division into Lower Limestone Shale and Carboniferous or Mountain Limestone had remained the official classification (De la Beche 1846, Woodward 1879) until the present survey took place, though Lloyd Morgan (1890) further subdivided the Carboniferous Limestone at Burrington Combe. Since 1905, the classification of the Carboniferous Limestone Series has been based on the coral-brachiopod zones of the Avonian defined by Vaughan (1905, 1906) in the Avon Gorge at Bristol. These zones were applied by Sibly to Burrington Combe (1905) and later (1906) to the Mendip area as a whole. The Ethological and faunal succession of Burrington Combe was the subject of a joint paper by Reynolds and Vaughan (1911) in which some of the original Avonian zonal boundaries were redefined. Subsequently the whole area was mapped on a zonal basis by Bamber (1924) and Welch (1929, 1932, 1933) but the zonal boundaries of these authors, notably the Z, C and S zones do not always agree with those defined by Vaughan in 1911.

A stratigraphical account of the Carboniferous Limestone Series and a description of the various formations that have been mapped by the Geological Survey in the Mendip–Bristol area has been given by Kellaway and Welch (1955). In this account the formations are classified into four main groups which reflect the principal changes in fauna and sedimentary facies. The formations recognized in the district covered by Sheet 280 are as follows:

In terms of Belgian usage (Fourmarier and others 1954) these groups in ascending order equate with the Upper Famennian–Lower Tournaisian, the Middle–Upper Tournaisian, the Lower Viséan and the Upper Viséan respectively. The relationship between the mapped formations and the Avonian zones is discussed below (Appendix I p. 194).

Thickness variations in the Lower Limestone Shale (p. 17) follow the same pattern as those of the Old Red Sandstone. In the Black Rock Limestone the variations are relatively small (p. 17). A change in the pattern of sedimentation is indicated by the isopachytes of the Clifton Down Group (Figure 2) which show that the thickest succession is developed adjacent to the Radstock Coal Basin.

Lower Limestone Shale

This formation rests conformably on the Upper Old Red Sandstone. The junction has been recorded at Burrington Combe (p. 179) and in a borehole near Priddy (p. 34), and in both cases was seen to be sharp. The Lower Limestone Shale consists of grey-green to dark grey and black shale, sandy shale and siltstone with interbedded limestones. The last range from nearly black, impure, fine-grained types to coarsely crinoidal rocks which are in places oolitic. The coarser types, which are confined to the lowest 100 ft of the formation, show colours ranging from dark grey to red depending on the amount of haematite present, the deepest red being seen in the 'Bryozoa Bed'.

A generalized succession for the Mendip area is as follows. At the base, shales, some 50 to 100 feet in thickness, with lenses of coarse oolite and bioelastic limestones including the haematitized crinoidal limestone of the 'Bryozoa Bed' type, are succeeded by a thick series of dark, fine-grained limestones with shaly partings with a total thickness of 100 ft or more. These are overlain by about 100 to 120 ft of shale with occasional thin limestones, including grey polyzoan limestones in the Priddy area. The uppermost beds consist of banded dark fossiliferous limestones transitional in lithology to the limestones in the succeeding Black Rock limestone, but including an appreciable thickness of thin shale bands. This group is 80 ft thick at Maesbury, and at Burrington, where the top 60 ft of the formation is not seen, 110 ft are recorded. The junction with the Black Rock limestone forms a good mappable boundary.

Brachiopods form the main part of the Lower Limestone Shale fauna. Most abundant are small Chonetids including Chonetes failandensis S. Smith, Chonetes (Plicochonetes) stoddarti Vaughan and Chonetes (Rugosochonetes) spp.,while Productus (Avonia?)bassus Vaughan and Eumetria carbonaria (Davidson) are also present. Species that range up in to the Black Rock Limestone, such as Camarotoechia mitcheldeanensis Vaughan and Unispirifer tornacensis (de Koninck) are common. The index fossil of the Cleistopora (K) Zone, Vaughania vetus Smyth is rare and has not been found during the present survey.

The thickest development (500 ft) of the Lower Limestone Shale is in the Blackdown Pericline and the thinnest (400 ft) in the Beacon Hill periclinal area.

Black Rock Limestone

The Black Rock Limestone consists predominantly of dark grey to black rather fine-grained limestones with abundant crinoidal debris; massive and thin-bedded types occur, whilst thin shaly partings are common in the lower and middle parts of the succession. The lowest third of the Black Rock Limestone is characterized by somewhat coarse-grained crinoidal rocks, while the top 50 ft or so consists of grey granular crinoidal limestones transitional in character between the dark grey rocks beneath and the paler coarse-grained limestones of the succeeding Vallis Limestone and Burrington Oolite. At distances of 100 and 500 ft above the base of the formation are two horizons characterized by chert which occurs in the form of sheets and nodules; silicification of the fossils in the lower part of the formation is widespread. The lower cherty band is only present in the Blackdown Pericline while the upper band, some 50 to 100 ft in thickness, has been recognized over the greater part of the Sheet. The total thickness of the formation varies from about 850 ft in the western part of the Blackdown Pericline to about 1000 ft in the area of the Beacon Hill Pericline.

The Black Rock Limestone is the most fossiliferous formation of the Carboniferous Limestone Series of the Mendips. The faunal succession at Burrington Combe is treated at length below (Appendix I, pp. 180–7). In general this succession appears to represent that of the area as a whole though detailed comparison is not possible owing to lack of evidence.

At the top of the Black Rock Limestone about 75 to 100 ft of dolomite and dolomitized limestones are present in the western part of the Blackdown Pericline. Irregularly developed masses of dolomite up to 300 ft or more in thickness occur locally on the southern side of the Mendips between Cheddar and Croscombe. In the Wells area the dolomites are very hard and tend to stand out as strong features; elsewhere they are softer and may give rise to hollows. Their limits, both vertically and laterally, are often difficult to draw in the field due to their gradual passage into non-dolomitized limestone. South of Wells the dolomites extend downwards to the main chert horizon.

Clifton Down Group

The Group ranges from 1100 to 1500 ft in thickness and shows much facies variation. In general, however, coarse bioclastic and oolitic facies give way upwards to calcite-mudstone facies with the greatest variation occurring in an intermediate zone (Figure 3).

Burrington Oolite

The light grey oolite and oolitic crinoidal limestones which constitute the bulk of this formation are commonly very massive and show current-bedding. The thickness varies from 600 to 700 ft over most of the area with a maximum of some 750 ft in the Wells area. North-eastwards of a line joining Priddy and Shepton Mallet the thickness of the Burrington Oolite varies inversely with that of the Vallis Limestone below and declines to a minimum of 150 to 200 ft between Binegar and Chewton Mendip. The lowest quarter or third of the Burrington Oolite includes much coarse crinoidal debris and beds of coarse crinoidal limestone. The upper quarter or third may include occasional pale grey calcite-mudstone bands. Oolitic limestone pebbles and pale calcite-mudstone pellets are not uncommon at many levels particularly in the top half of the sequence. In the Wells area, a well-marked band of dolomite occurs towards the base of the formation.

The commonest fossils in the Burrington Oolite are Palaeosmilia murchisoni Edwards and Haime and Chonetes (Megachonetes) sp.of the papilionaceus Phillips group, in the lower part and Lithostrotion martini Edwards and Haime in the upper part while Productus (Gigantoproductus) θ Vaughan has a long range. Large Chonetids such as Chonetes (Megachonetes) magnus Rotai and Delepinea carinata (Garwood) are found in the lower part of the formation, and Davidsonina carbonaria (McCoy) near the top.

Vallis Limestone

South-west of a line joining Blagdon with Shepton Mallet the lenses of pale-grey coarse crinoidal limestone which occur within the Burrington Oolite cannot be separately mapped. To the north-east they increase in thickness and have been mapped separately. This facies, known as the Vallis Limestone, replaces the Burrington Oolite from the base upwards and reaches a maximum thickness of some 500 ft in the Binegar–Oakhill area. A lower Visean age for these beds is indicated by the presence of P. murchisoni, Caninia aff. subibicina McCoy, Caninia sp. cf. C. caninoides (Sibly), 'Zaphrentis' konincki aff. kentensis Garwood and the large Chonetid C. (M.) magnus.

Clifton Down Limestone

Over most of the district under review the following tripartite division of the rocks^‡10  can be recognized (Figure 3):

The lowest division typically shows a rapid alternation of rock types as seen in Burrington Combe (Appendix I p. 188). In the Cheddar area, however where this division is exceptionally thick, oolites and splintery limestones are separated into two distinct divisions; a basal dark limestone division 100 to 120 ft thick, the Cheddar Limestone (new name), being overlain by a white oolite 120 to 190 ft thick, the Cheddar Oolite (Green 1953, p. 20). These two formations are separately mapped from about 1 mile north-west of Cheddar up to the line of the Priddy Fault, some 4 miles to the south-east. In the area of the Cheddar Gorge the lower half of the Cheddar Limestone consists of dark grey granular limestone with a fauna including Carcinophyllum mendipense Sibly, L. martini, P. murchisoni; Composita ficoidea (Vaughan), D. carbonaria, and Productus (Striatifera)cf. striatus (Fischer de Waldheim). This fauna was described by Sibly (1906, p. 357) who classified it as S₁, but the association with D. carbonaria led Vaughan (in Reynolds and Vaughan 1911, p. 370) to regard it as S₂. Later workers (e.g. Welch 1929, p. 48) followed Sibly's grouping. Both in the gorge and farther west, the lower limit of the Cheddar Limestone is poorly defined, the granular limestones merging into the top of the underlying Burrington Oolite. A band of oolite, 160 ft thick, similar in composition to the Cheddar Oolite can be traced in the Binegar area but is not shown on the one-inch map. In the Wells area, the lower division of the Clifton Down Limestone consists of only 25 to 40 ft of calcite-mudstone with thin oolite bands, while the underlying Burrington Oolite shows a complementary increase in thickness, indicating that, here, a lateral passage of the lowest part of the Clifton Down Limestone into the Burrington Oolite facies has probably occurred.

In the middle division silicification of the fossils and the presence of chert nodules and bands are characteristic features. The upper boundary of this division is in places indefinite owing to the presence of calcite-mudstone beds near the top, as may be seen in the eastern parts of the North Hill and Pen Hill periclines.

'The upper division consists dominantly of calcite-mudstones although algal mudstones, oolites, pisolites and mudstone-pellet beds are common. The rocks are usually well-bedded and of a porcellanous texture ('chinastone–limestone'). In parts of the North Hill Pericline this division attains 250 to 300 ft in thickness.

The fauna of the Clifton Down Limestone is marked by an abundance of individuals but a paucity of species. It includes Lithostrotion martini (particularly in the lower and middle divisions), C. ficoidea, D. carbonaria (confined to the lower division), C. (M.) sp. papilionaceus group, and Productus (Linoproductus) corrugatohemisphericus Garwood. Basaltiform Lithostrotionids are also present.

Hotwells Limestone

This formation mainly consists of fossiliferous, massive, grey, crinoidal and oolitic, bioclastic limestones, thus contrasting strongly with the underlying Clifton Down Limestone; the junction between the two formations is usually very well-defined except on the rare occasions when one or more thin beds of calcite–mudstone are present at the base of the Hotwells Limestone. At the junction there is a well-marked faunal break; the bottom of the Hotwells Limestone contains a typical Lower Dibunophyllum (D₁) Zone fauna with Dibunophyllum bourtonense Garwood and Goodyear, Lithostrotion junceum (Fleming), L. pauciradiale (McCoy), P. murchisoni and Productus (Gigantoproductus) maximus McCoy. The upper part of the formation exposed near Compton Martin has a characteristic Upper Dibunophyllum (D₂) Zone fauna including the corals Lithostrotion portlocki (Bronn), Nemistium edmondsi S. Smith and Orionastraea ensifer (Edwards and Haime). The characteristic D₁ brachiopod Davidsonina septosa (Phillips) has been recorded only from Chewton Mendip at one locality at the base of the Hotwells Limestone.

In the Cheddar–Westbury area, limestones, with chert nodules and seams, about 10 ft thick overlying some 20 ft of flaggy, black, splintery limestone, form a useful marker band about 70 ft above the base of the Hotwells Limestone. The upper part of the formation is characterized by a diversity of rock types interbedded with the more typical grey limestones. These include 'rubbly beds' or pseudobreccias, limestones with chert, thin quartzitic sandstones, and, particularly on the north side of the Mendips, purplish shales with nodular, white, porcellanous dolomitic limestones. The thickness of the Hotwells Limestone apparently increases in a north-easterly direction from about 500 ft, or less, at Ebbor Rocks to about 700 to 750 ft along the south-western edge of the Radstock Coal Basin between Ubley and Ashwick. These thickness changes follow the same general pattern as those noted in the Clifton Down Group.

Lithology and conditions of deposition

The limestones in the Carboniferous Limestone Series may be grouped into three main types: bioclastic limestones, calcite-mudstones, and oolitic limestones.

The bioclastic limestones, which form the largest single group, consist of comminuted shells and crinoid stems and are a typical facies of rather shallow shelf offshore conditions. Widespread variants of these rocks include granular limestone, and the 'Petit granit' of Belgian geologists. The former is an even-textured medium to fine-grained rock; the latter consists of very coarse crinoidal debris set in a fine-grained matrix of the same material. Bioclastic limestones include the Black Rock Limestone, the Vallis Limestone, certain beds in the lower and middle parts of the Clifton Down Limestone, and the bulk of the Hotwells Limestone. The limestone bands in the Lower Limestone Shale are also bioclastic, but the 'Bryozoa Bed' type is unique in the presence of haematite, which probably represents an original ferruginous deposit accumulating under lagoonal conditions (Dixon in Dixon and Vaughan 1911, p. 515; Reynolds 1921, p. 218). Dr. K. C. Dunham, reporting on a thin section (E23827) of a specimen of coarse crinoidal limestone from this horizon collected in the Burrington Combe area stated that the matrix consisted of ankerite (ω=1.702).

The calcite-mudstone group comprises compact splintery textured rocks consisting predominantly of calcite mud and silt. The finest grained varieties of these rocks are of porcellanous texture with a characteristic conchoidal fracture and weather white, hence their older names of 'chinastone' or 'chinastone-limestone'. The conditions of deposition of these rocks was first discussed in detail by Dixon (1911, p. 511), while recent discussions are provided by Newell and Rigby (1957) and George (1958, pp. 254–9). The rocks were probably laid down in coastal and lagoonal' mud flats at, or a little below, sea level. The rock types though mainly calcite-mudstone include calcite-siltstones, dolomite-mudstones, oolites, calcite-mudstone-pellet rocks, algal limestones and pseudobreccias. The latter owe their texture to the patchy recrystallization of their calcitic or dolomitic mud content soon after they were laid down (Dixon 1911, pp. 507–11). Pseudobreccias are best developed in the Hotwells Limestone where the more incoherent varieties are termed 'rubbly beds'. The presence of calcareous algae in the calcite-mudstones is characteristic though it is still debatable how much of the calcite mud is the product of chemical precipitation and how much is finely divided algal tissue. Calcite-mudstones are mainly developed in the upper part of the Clifton Down Limestone though a subsidiary development may occur at the bottom.

The oolitic limestones form the least sharply defined group of the three types. All gradations of grain size occur between very fine oolite and pisolite. Increase of fossil debris results in the formation of intermediate types which pass into typical bioclastic limestones, while the increase of calcite-mud in the matrix or of calcite-mudstone pellets produces gradations to the calcite-mudstones. Macrofossils are usually sparse. The oolitic limestones were laid down in shallow shelf seas in an environment combining strong chemical precipitation with well-marked current action (Ming 1954; Newell and Rigby 1957). The main representatives of this group comprise the Burrington Oolite, the Cheddar Oolite, and parts of the Hotwells Limestone.

In addition to the rock types considered above, dolomitic limestones and dolomites form a distinctive group in the Carboniferous Limestone in the South-Western Province. Dixon (1907, pp. 11–20; 1911, pp. 484–5) and later workers agree that the dolomites, as distinct from the dolomite-mudstones, are secondary in origin i.e. the result of alteration of limestones. Dixon further subdivided these dolomites into 'contemporary' types in which the magnesian salts were probably derived from the Carboniferous sea, and 'subsequent' dolomites in which the dolomitization was much later, possibly Triassic in age (see also, George 1954, p. 283; 1956, p. 309). In the Mendips the evidence of secondary age of the dolomites is provided by the widespread occurrence of unaltered crinoidal debris in the dolomites themselves; their rapid lateral passage into into normal limestone, and the presence of a complete series of gradations between the altered and unaltered limestones.

The dolomites are mainly confined to the Black Rock Limestone and, to a much less extent, to the lower part of the Burrington Oolite. When fresh these rocks are dark grey in colour, sometimes tinged with purple which weathering alters to various shades of brown and grey. They are sugary in texture with dolomite grains ranging in size from 0.05 to 0.25 mm (E24624), (E24625), (E24626), (E24627) , (E25057), (E26779), (E26780), (E27566). They are probably mainly 'contemporary' in origin and, in many cases, the dolomitization can be proved to predate Armorican structures, whilst the recent discovery by Kellaway (1960, p. 26) of dolomite pebbles in an Upper Visean conglomerate at Kingsweston in the Bristol area lends additional support to this view. Examples of 'subsequent' dolomites are seen along the sides of exhumed Triassic valleys, where a wide range of stratigraphic subdivisions are affected; this alteration tends to be patchy in occurrence and of limited lateral extent. G.W.G.

Details

Blackdown pericline

Elborough–Loxton

In the core of the Blackdown Pericline the Black Rock Limestone has a broad outcrop caused by low dips and minor folding across the periclinal axis. South of Christon Plantation a sinuous ridge of much disturbed limestone appears to mark the line of the east to west Bleadon Thrust Fault, a dislocation which increases in size when traced farther west on to Bleadon Hill.

The lowest observed beds of Black Rock Limestone crop out immediately north of Christon Plantation, where, in a series of low crags, is exposed dark crinoidal limestone containing abundant Unispirifer tornacensis and dipping 18° to the north. South of this a southerly dip of 30° to 40° is maintained up to the line of the Bleadon Fault, but farther to the south the dips decrease to 8° to 12°. There are few exposures south of the fault, although debris of silicified and cherty crinoidal limestone is widespread.

Burrington Oolite occupies the steep northern slope of the limestone mass, the dips increasing progressively northwards away from the axis until they reach 54° south of Elborough. On the southern flank of the fold the oolite is exposed in a number of old quarries north-north-west of Loxton. Much of the rock has been affected by dolomitization and all stages from the unaltered rock to dolomite may be seen. In an old quarry [ST 374 562] 300 yd N. 30° W. of Loxton church the rock has weathered into a dolomite sand.

Clifton Down Limestone comprising splintery limestone, oolite, and calcite-mudstone with Lithostrotion has a small outcrop 0.25 mile west-north-west of Loxton.

Banwell

The Carboniferous Limestone is poorly exposed in this inlier which extends from the western end of Banwell Hill to Banwell Camp. Black Rock Limestone dipping at angles of 40° to 60° to the north, occupies the southern slopes of the mass. Burrington Oolite forms nearly all the northern face of the hill, and at some 120 ft above the base a thin band of rather dark grey crinoidal limestone containing gastropods occurs at intervals on Banwell Hill. In two old quarries [ST 399 589] 300 yd S. 27° W. of Banwell church some 200 ft of white slightly crinoidal oolite are seen dipping 65° to 70° to the north. Clifton Down Limestone consisting of splintery dolomitized limestone crops out in two narrow strips on the northern side of the inlier (see also, Appendix II, pp. 199–200).

Sandford Hill

Poorly exposed Black Rock Limestone, with a chert horizon some 400 ft below the top, crops out on the southern slope of Sandford and Lyncombe hills. Dolomite has been mapped along the southern part of Lyncombe Hill, where it is about 100 ft thick. The Burrington Oolite is well-exposed in Sandford Quarry [ST 421 591] where over 500 ft of white oolite and oolite-crinoidal limestone are seen dipping at 70° to the north, and the same rock is exposed in the deep mining trenches cut parallel to the strike of the rocks on the top of Sandford Hill. The Clifton Down Limestone comprising fine-grained splintery limestone and fine-grained oolite with Lithostrotion and Productids is exposed in the northern top of Sandford Quarry and in natural outcrops on the north slope of the hill. South of Churchill and east of Lyncombe Hill both Burrington Oolite and Clifton Down Limestone are inverted and dip at 40° to the south. F.B.A.W.

Windmill Hill, north of Churchill

The three inliers of Carboniferous Limestone which crop out through a cover of Triassic on this hill are probably all referable to the Hotwells Limestone. The two westernmost are poorly exposed and much dolomitized. Shallow quarries at the top of the hill [ST 444 602] expose light grey massive oolitic limestone folded into an anticline with dips to the north and south of 45° to 50°. Fossils collected from here included: Caninia cf. densa Lewis [of Hudson & Cotton 1945], C. cf. subibicina McCoy, isolated corallites of Lonsdaleia cf. duplicata (Martin); Trepostomatous polyzoa and Productus (Gigantoproductus)cf. maximus group. On the north side of the hill a small quarry [ST 443 603] exposes 15 ft of very massive, rather fine-grained limestone dipping 30° to the south-south-west.

Churchill–Burrington

The Lower Limestone Shale occupies a broad rather marshy depression extending in an east-south-easterly direction from Rowberrow. The Bryozoa Bed limestone forms a good feature and its position is sometimes marked by swallow holes. It is best exposed beside a footpath [ST 458 584] 1000 yd east-south-east of Rowberrow church, where 20 ft of red, thinly-bedded, crinoidal, more or less oolitic, limestones overlie 6 ft of flaggy, splintery, grey, crinoidal limestone.

The Black Rock Limestone is locally much dolomitized and its junction with the succeeding rocks may be difficult to determine. The lower two-thirds of the sequence is magnificently exposed in the cuttings and quarries on the west side of the main Bridgwater road (A. 39) and in Dolebury Warren. The lower chert beds are best seen in the cliffs above Read's Cavern [ST 468 584] where they attain 60 ft in thickness, but they die out westwards from Dolebury Warren. The main chert beds, some 100 ft in thickness, are conspicuous.

Dolomite can be mapped only in the vicinity of the Bridgwater road and Burring-ton Combe; elsewhere it passes laterally into patchily dolomitized limestone in the upper part of the Black Rock Limestone.

The Burrington Oolite is well seen in the cuttings and quarries on the west side of the Bridgwater road where the strata are inverted and dip southwards at angles of from 18° to 52°. Dolomitization, especially towards the base, may be extensive and individual dolomite beds may be followed along the crop for hundreds of yards as for example on Dolebury Warren.

The middle and upper divisions of the Clifton Down Limestone (p. 19) are well exposed along the northern edge of the Carboniferous Limestone outcrop. North of Dolebury Warren, cherts are developed in the uppermost calcite-mudstone group, as well as in the underlying Lithostrotion beds. The basal oolite-mudstone group, which is poorly exposed, apparently thins westwards from Burrington Combe, and is less than 70 ft thick in the Churchill area.

The Hotwells Limestone is exposed in a quarry [ST 453 594] 0.5 mile south-east of the main cross-roads at Churchill (Churchill Gate). Here 30 ft of massive grey crinoidal limestone overlie 12 ft of calcite-mudstones of the underlying Clifton Down Limestone, the junction between the two being sharp and undulose. The dip is 80° to the north.

Burrington Combe–Nordrach-on-Mendip–Castle of Comfort Inn

This area comprises the north-eastern part of the Blackdown Pericline and is bounded on the south and south-west by the Stock Hill Fault. The most complete and readily accessible section of the Carboniferous Limestone Series in the Mendips is in Burrington Combe (Figure 19) where some 2700 ft of strata can be seen. Owing to its importance this section is described separately (Appendix I, p. 177). A summary of the thickness of the formations is as follows:

The Lower Limestone Shale is not exposed in Burrington Combe itself but is seen in the West and East Twin Streams draining northwards from Black Down into the Combe. The lower half of the sequence including the coarse crinoidal and oolitic limestones of the Bryozoa Bed is best seen in the eastern stream (Reynolds and Vaughan 1911, fig. 2). Here the sequence is apparently more argillaceous than that described in the Axbridge R.D.C. adit in the West Twin Stream. The feature formed by the Bryozoa Bed can be traced for a mile eastwards from the East Twin Stream.

The lower beds of the Black Rock Limestone, consisting of dark grey crinoidal limestones with abundant brachiopods such as Unispirifer tornacensis, are exposed in many small quarries throughout the area, good examples being afforded by those at Paywell Farm [ST 508 569]. The lower chert band (see p. 17) extends eastwards from Burrington Combe to Paywell Farm where it dies out. A small quarry [ST 521 562] 300 yd north-east of Warren Farm, in dark crinoidal limestones dipping at 30° to the north-east yielded the following fossils Caninia sp.(cf. cornucopiae vesicularis Saloe), Fasciculophyllum omaliusi (Edwards and Haime), Zaphrentites delanouei (Edwards and Haime) and Leptaena analoga (Phillips). These beds lie about 150 ft below the main chert horizon, which is well developed in the northern limb of the pericline but dies out east of Nordrach. A small quarry [ST 492 579], 300 yd south-east of Lower Ellick Farm, in the upper chert horizon yielded a fauna including ?Caninia cornucopiae Michelin and Productus (Pugilis?)cf. bristolensis Muir-Wood.

A good exposure of Burrington Oolite is seen in an old quarry [ST 498 582] 1100 yd S. 37° W. of Blagdon church. This section is of particular interest because it largely bridges the 120 ft gap in the exposed succession, 620 ft above the base of the Burrington Oolite, in the Burrington Combe section, 1.25 miles to the west (Appendix I, p. 187). The section which starts some 35 ft below the top of that gap, is as follows:

Some 200 ft of jumbled cores from a borehole [ST 527 568] 120 yd north-west of Hazel Manor (Richardson 1928, p. 74) were examined in 1948 and were found to consist of very massive grey oolites and oolitic limestones typical of the upper part of the Burrington Oolite. In the area around Fernhill Farm where old mining trenches provide numerous exposures, the lower part of the succession includes much coarse crinoidal limestone indistinguishable from the Vallis Limestone. Trenches exposing beds at a high horizon in the Burrington Oolite, some 1000 yd east-south-east of Fernhill Farm show a local well-marked bed or lens of granular limestone packed with Lithostrotion as well as Carcinophyllum vaughani Salée, Syringopora cf. ramulosa Goldfuss; Composita aff. ambigua (J. Sowerby) and C. cf. ficoidea. This horizon may also be represented in the mining trenches to the north-west by grey oolitic-crinoidal limestones with scattered specimens of Lithostrotion and Carcinophyllum.

The succession in the Clifton Down Limestone is essentially similar to that of Burrington Combe. About 0.75 mile east of the Combe, an overgrown quarry [ST 488 587] 750 yd north-north-west of Lower Ellick Farm exposes 15 ft of typical Hotwells Limestone overlying about 175 ft of the upper calcite mudstone beds in which the dip is 60° to the north-north-east. A quarry [ST 498 582] about 0.5 mile to the east-south-east, which shows the junction of the Clifton Down Limestone with the Burrington Oolite, has been described above. Natural outcrops north-west of Ubley Hill Farm expose most of the Clifton Down Limestone sequence which dips at from 55° to 64° to the north-north-east. A similar succession can be seen in the upper part of the valley (Compton Combe) which leads south-westwards from Compton Martin church. In a quarry (Butt's Quarry) [ST 540 561], 1000 yd S. 30° W. of Compton Martin church, the junction of the Clifton Down Limestone with the Hotwells Limestone is well exposed, the beds dipping at 20° to 27° to the north-east. The section is cut by two small faults, marked by prominent calcite veining, and the top bed of the Clifton Down Limestone is a conspicuous, white weathering, nodular calcite-mudstone. Old mining trenches east of the Compton Martin–Wells road in the Gibbets Brow area crossing the easterly continuation of the Blackdown periclinal axis provide numerous exposures in the upper part of the Clifton Down Limestone. A large quarry [ST 544 552] adjacent to the road exposes 40 ft of Lithostrotion beds with much chert and thin beds crowded with Composita. Fossils collected included Palaeosmilia aff. murchisoni. A borehole [ST 541 544], 1350 yd N. 12° W. of the Castle of Comfort Inn, proved 55 ft of Clifton Down Limestone underlying Hotwells Limestone and Triassic (see p. 207). Evidence from scattered exposures and abundant mine waste in the area east of the road leading from Nordrach-on-Mendip to the Castle of Comfort Inn indicates that the basal oolite-mudstone group of the Clifton Down Limestone is about 200 ft in thickness, half of which is represented by pale grey coarse oolite. Small quarries near the road [ST 535 545] provided discontinuous exposures of the lowest 40 ft of the Clifton Down Limestone, the basal bed of which consists of black splintery limestone with colonies of Lithostrotion arachnoideum (M Toy).

Scattered exposures along the north side of the Mendips, such as in the quarry [ST 488 587] 0.75 mile east of Burrington Combe mentioned above, show the lower part of the Hotwells Limestone. The large disused quarry [ST 525 576] on the hillside southwest of Ubley shows the following section in beds near the top of the succession:

A rather overgrown quarry [ST 535 568] in Compton Wood, 700 yd east of Hazel Manor, exposes about 80 ft of massive limestone dipping at 30° to 35° to the northeast. The lowest beds, some 60 ft above the base of the Hotwells Limestone, include a bed with a rich coral assemblage consisting of Carcinophyllum vaughani, Dibunophyllum sp. juv., Lithostrotion junceum, L. cf. martini, L. pauciradiale, Palaeosmilia murchisoni, and Syringopora sp. Farther east in a quarry (Cliff Quarry) [ST 541 568], 0.25 mile south-west of Compton Martin church, beds at about the same stratigraphical position as those in the Ubley quarry are seen.

The rocks have an average dip of 40° to the north-east.

Natural exposures of lower beds in the Hotwells Limestone sequence are seen in the sides of the valley (Compton Combe) leading south-westwards from Cliff Quarry, while the basal beds are found in Butt's Quarry about 0.375 mile to the south (see p. 26). Quarries [ST 546 560] 1150 yd south of Compton Martin church, on either side of the road to Wells, expose massive Hotwells Limestone with cherty patches. The biggest quarry, in the north-east quadrant of the crossroads, shows some 55 ft of cherty limestone including a 3 ft 'rubbly bed' 15 ft below the top. Exposures of cherty limestone south-west of the crossroads include a quarry [ST 550 555], 700 yd south-south-west of Shortcombe Farm, which shows about 30 ft of limestone capped by a 2 ft coral bed packed with Lithostrotion pauciradiale. The mining trenches east of the Compton Martin–Wells road (see also p. 26) afford many exposures of the lower beds of the Hotwells Limestone, and in one place, 860 yd N. of Spring Farm abundant Diphyphyllum aff. lateseptatum McCoy was seen just above the base. A borehole, 0.75 mile north of the Castle of Comfort Inn (p. 207), proved Hotwells Limestone under a thick cover of Triassic. On the southern limb of the pericline, a small quarry adjoining the east side of the Nordrach–Castle of Comfort Inn road, 350 yd north-west of Hill Grange, shows massive limestone dipping at 40° to the south; see also Hill Grange Borehole (p. 203).

West of Stock Hill Fault–Charterhouse

The ill-exposed Lower Limestone Shale outcrop is marked by a marshy depression. The Black Rock Limestone is fairly well exposed, the lower beds with abundant brachiopods being seen in numerous small exposures in the neighbourhood of Charterhouse. A small quarry [ST512 533] at the south-western corner of Ubley Warren Farm yielded 'Zaphrentis' konincki forma a Carruthers in black fine-grained limestone estimated to lie about 25 ft below the main chert beds. The latter are well exposed in old mining trenches extending 0.5 mile westwards from the farm, where the dip is 15° to 25° to the south. Adjoining the Stock Hill Fault at a distance of 650 yd south-south-east of Warren Farm, a small quarry [ST 520 555] in black crinoidal limestone dipping at 15° to the west yielded Caninia cf. cylindrica (Scouler), and Fasciculophyllum omaliusi. This horizon lies some 150 ft below the top of the Black Rock Limestone. The top 100 ft of the Black Rock Limestone, best seen in the sides of Velvet Bottom, tend to be lighter in colour and coarser in texture than the underlying limestones. The junction with the overlying Burrington Oolite is usually obscured by dolomitization.

Many small quarries and crags in the area south-south-west of Charterhouse show discontinuous exposures of the Burrington Oolite and the lower part of the Clifton Down Limestone including the Cheddar Limestone and Cheddar Oolite, the succession being similar to that in the Cheddar Gorge.

Lower Farm, Charterhouse–Cheddar

The Lower Limestone Shale which gives rise to a belt of swampy ground north of Lower Farm, is best seen in the sides of the lane 150 yd north-west of the farm.

The Black Rock Limestone is exposed at intervals in Long Wood valley where the succession, in ascending order, is as follows: crinoidal limestones, including a horizon of silicified fossils 80 ft from base, 475 ft, main chert beds 75 ft, crinoidal limestones 400 ft.

The lowest beds of the Burrington Oolite, exposed in crags in the valley sides south of Long Wood, include much light grey coarse crinoidal limestone (cf. Vallis Limestone). Higher beds are well exposed in the quarry (Meredith's Quarry) [ST 496 547], 1400 yd S. of Lower Farm, Charterhouse, where massive vertical jointed oolites dip at 20° to 22° to the south-south-west. Throughout the remainder of the valley to its junction with the Cheddar Gorge, at Black Rock Gate, [ST 482 545] discontinuous exposures of Burrington Oolite, here 600 ft thick, and of the overlying Cheddar Limestone duplicate the section seen in the Gorge.

The upper beds of the Burrington Oolite and the Cheddar Limestone can be examined in the part of Cheddar Gorge (Figure 4) extending from Black Rock Gate westwards to the roadside reservoir 1000 yd S. 10° W. of Piney Sleight Farm. There is a gradual passage upwards from the Burrington Oolite into the Cheddar Limestone and the base of the latter is taken arbitrarily at a prominent bedding plane seen 2 ft above road level on the west side of the road, 240 yd W. 21° N. of Black Rock Gate; a vertical calcite vein, 2.5 ft across, associated with small-scale disturbances is seen on the west side of the road 51 yd farther north.

The section in this part of the gorge is as follows:

Beyond here, close to the reservoir enclosure, the gorge turns south-westwards and cuts through the Cheddar Oolite^‡12 , the top of the latter and a complete section through the overlying banded oolite-mudstone beds of the Clifton Down Limestone being seen at the foot of the precipitous cliffs bordering the zig-zag road known as the Horseshoe [ST 472 543]. The overlying Lithostrotion beds of the Clifton Down Limestone descend to road level immediately west of the Horseshoe.

The section in the area of the Horseshoe is:

Most of the remainder of the Gorge is cut through the middle and upper parts of the Clifton Down Limestone, the section being as follows:

Hotwells Limestone, overlying the uppermost beds of the Clifton Down Limestone, is seen in discontinuous exposures on the south side of the road in the area of the caverns (Long Hole, Gough's Cave, Cox's Cave). The most complete section, however, is that in the higher part of the crags above the Gough's Cave [ST 467 539], 1300 yd N. 41° E. of Cheddar church. The section here is as follows:

The cliffs at the cavern (Long Hole) [ST 4665 5388] 70 yd south-west of Gough's Cave, from the base of which the Cheddar River emerges, show a complete section in the lower beds. The base of the Hotwells Limestone, which roofs part of Long Hole Cavern itself, is sharply marked off from the underlying Clifton Down Limestone, of which 11 ft are seen, consisting of black well-bedded calcite-mudstone with shaly partings up to 2 inches thick. Small-scale thrusting is developed.

The area south of Cheddar Cliffs and the upper part of the Cheddar valley above Black Rock Gate is described below (see p. 36). G.W.G.

Cheddar–Tyning's Farm–Callow Hill–Shute Shelve Hill

The Lower Limestone Shale gives rise to a belt of low, often wet, ground to the north of the main limestone escarpment (Plate 2A). Old mining trenches [ST 461 568], 300 yd east of Longbottom Farm, expose about 25 ft of grey oolitic and crinoidal limestone at the base of the Lower Limestone Shale, in which the dip is 17° to 20° to the south. A swallow hole [ST 467 568] 900 yd east of the farm exposes similar limestone. A laneside section [ST 462 565], 300 yd south-east of Longbottom Farm shows about 90 ft of poorly exposed shales with thin beds of fine-grained limestone overlain by black limestone. Locally in the Hale Coombe area there is a development of reddish crinoidal limestone of the 'Bryozoa Bed' type, and in old quarries [ST 431 566] 550 yd south-west of Winterhead 12 ft of reddish crinoidal limestone with red-purple shale were seen. The uppermost beds of the Lower Limestone Shale, consisting of red dolomitized limestone and shale may be seen at Shute Shelve, in the cutting on the main road 1350 yd south-east of Winscombe church. G.W.G., F.B.A.W.

Traced westwards from the Long Wood valley, silicification at the level of the lower chert horizon in the Black Rock Limestone becomes more marked. These beds are seen in the crags at the entrance to the cave (G.B. cave) [ST 476 562], 0.5 mile east-south-east of Tynings Farm, and in long lines of crags striking from east to west on both sides of the Shipham–Cheddar road, about 0.5 mile south of Shipham. Crags on the north side of Callow Hill [ST 443 563] to the west of the Shipham road provide good exposures of the main chert beds. Farther south, the large quarries [ST 451 460] of Crow, Catchpole & Co., Ltd. on the east side of the road expose some 200 ft of black fine-grained crinoidal limestones in the upper part of the Black Rock Limestone sequence dipping at 30° to the south and containing a fairly abundant coral fauna of Caniniids and Zaphrentids. A remarkable assemblage of cephalopods and gastropods has been described by Wallis (1935, 1936) from material collected from this quarry. South of the main quarry, small roadside exposures show an upward gradation from granular crinoidal limestone into the lighter coloured and coarser crinoidal and oolitic limestones which characterize the basal beds of the Burrington Oolite.

Burrington Oolite is well seen in two large quarries on either side of the Cheddar Road about 1 mile south of Shipham. In the quarry [ST 445 557] of the Callow Rock Lime Co., Ltd. a lower horizon is exposed than in the Somerset C.C. quarry [ST 451 555] at The Perch. In both cases some 200 ft of oolites and crinoidal oolites are worked but in the former crinoidal debris becomes increasingly important towards the base. Farther west, dark grey rather crinoidal limestone [ST 426 551] 850 yd north-west of Axbridge church yielded Caninia caninoides (Sibly) and Carcinophyllum mendipense. This horizon lies near the top of the Burrington Oolite and probably corresponds with the lower part of the Cheddar Limestone farther east. In the area of Shute Shelve much of the Burrington Oolite is heavily dolomitized.

The Cheddar Limestone and the Cheddar Oolite thin westwards from the Cheddar Gorge area and cease to be separately recognizable west of Cheddar Woods. In the Cheddar Woods area the Cheddar Oolite has diminished from 190 ft (in the Gorge) to about 120 ft and the Cheddar Limestone is only represented by a band, 20 to 30 ft thick, of dark splintery limestone. West of Cheddar Woods, the base of the Clifton Down Limestone is marked by a band of dark splintery limestone with silicified Lithostrotion. The base of the Cheddar Oolite is seen in an old quarry [ST 470 550], 0.5 mile south-east of Piney Sleight Farm. Here 3 ft of grey oolite overlie 25 ft of dark grey and almost black fine-grained limestone with occasional Lithostrotion. The Cheddar Oolite is well exposed in the large quarry (Batts Combe Quarry) [ST 459 548] 1 mile north of Cheddar church, in which some 120 ft of light grey oolite underlie about 30 ft of banded oolite and calcite-mudstone. In the Cheddar area the beds of banded oolite and mudstone lying between the top of the Cheddar Oolite and the base of the Lithostrotion beds are about 30 to 40 ft thick. Apart from a distinctive bed, some 7 to 10 ft thick, of calcite-mudstone and porcellanous-textured oolite with abundant mudstone pellets directly overlying the Cheddar Oolite, these beds are not as clearly differentiated from the overlying Lithostrotion beds as in the northern and eastern parts of the Blackdown Pericline. This is due to the presence of interbedded dark grey splintery granular limestones with occasional Lithostrotion. The beds above the Cheddar Oolite are best seen in the top of Batts Combe Quarry (see above) and in the crags (Fore Cliffs) [ST 455 569] about 1.25 miles north-north-east of Cheddar church. The Lithostrotion beds are well exposed in the quarry (Chelm's Combe Quarry) [ST 463 545] 0.75 mile north-north-east of Cheddar church where 120 ft of dark grey limestones with Lithostrotion and chert are overlain by 25 ft of calcite-mudstone including a median 11-ft bed of splintery limestone with scattered silicified Lithostrotion. These beds and the succeeding calcite-mudstones are well seen in the crags of Fore Cliffs. In the weste'rn part of the area, the lower half of the Clifton Down Limestone crops out in a series of large dip slopes on the hill-side north of Axbridge where the rock consists of splintery fine-grained grey limestone and oolite with Composita. The beds are in places much dolomitized.

The main outcrop of the Hotwells Limestone lies to the south of Cheddar Cliffs and is described below (pp. 36–7). A small isolated area occurs at Warrens Hill 0.75 mile north-north-west of Cheddar church, in which a quarry [ST 457 542] exposes 7.5 ft of massive grey crinoidal limestone with corals overlying calcite-mudstones of the Clifton Down Limestone. G.W.G.

Shute Shelve–Crook Peak

Lower Limestone Shale gives rise to a belt of heavy clay ground at the foot of the limestone scarp on the north side of the hill mass. In the approach cuttings to a disused quarry (Winscombe Quarry) [ST 405 564] 800 yd south-west of Winscombe church were seen isoclinally folded olive-grey shales with nodules of crinoidal limestone containing Camarotoechia sp. and Spiriferellina octoplicata (J. de C. Sowerby).

Black Rock Limestone over 850 ft thick occupies most of the northern and central part of the hill mass where the beds dip in a south or south-south-westerly direction at 25° to 30°. In the eastern part of the area two chert horizons are developed at distances of 100 ft and 350 ft above the base of the formation; the lower band can be traced as far west as Winscombe Quarry, the upper band continues to the western end of the mass. Black Rock Limestone is exposed in a number of crags, notably Wringstone Rocks [ST 396 564] south of Barton. In North Quarry [ST 386 563], 1200 yd north-east of Loxton church, over 150 ft of black crinoidal limestone with abundant Spiriferids dips at 30° slightly west of south. At the top of the quarry the upper chert horizon is exposed. The succession in the lowest beds of the Black Rock Limestone in the disused Winscombe Quarry (mentioned above) is as follows:

The Burrington Oolite, which here reaches a thickness of some 600 to 650 ft forms most of the southern slope of Crook Peak, Compton Hill, and Wavering Down and is exposed in a number of crags the best examples of these being Crook Peak (summit) with its south-easterly prolongation towards Compton Bishop, White Cliff [ST 409 555] and Yew Tree Cliff [ST 415 554] respectively north-west and north of Cross. In this area, as for instance on the summit of Crook Peak, the Burrington Oolite yields Chonetids of the Delepinea carinata (Garwood) type, and numerous Palaeosmilia murchisoni.

Clifton Down Limestone crops out on the lower southern slopes of Crook Peak and in a small area north of Cross. The formation consists of bands of splintery fine-grained limestone containing Lithostrotion alternating with fine-grained oolite. The beds are locally much dolomitized. The best exposure is the abandoned Cross Quarry [ST 414 549] which shows some 56 ft of oolite, dolomite and thin impersistent calcite-mudstones (see Bamber 1924, pp. 85–6). F.B.A.W.

North Hill Pericline, Western half

This area is bounded by the Stock Hill Fault on the east and the South Western Overthrust on the south and west. The Cheddar Syncline is included in the western part of the area.

Hunter's Lodge Inn–Priddy (south of the Priddy Fault)

The Lower Limestone Shale, about 450 ft thick, gives rise to a belt of damp clay ground in which exposures are rare owing to a thick covering of Old Red Sandstone wash. A borehole [ST 538 513] 1000 yd E. 8° S. of Priddy church proved the lowest quarter of the succession (Whittard 1949, p. 481) in which grey unfossiliferous limestones with some interbedded dark grey shale dipping at 48° extended down to a depth of 73 ft and was underlain by dark blue-grey micaceous shale with occasional beds, up to 8 inches thick, of crystalline, crinoidal limestone down to a sharp junction with the Old Red Sandstone at 190 ft. A dip of 60° was measured in the lower beds. No representative of the Bryozoa Bed was present.

The junction of the Black Rock Limestone with the underlying shales is locally marked by a line of swallets, the best known of these being Eastwater Cavern [ST 539 506]. In the cliff above the cave are seen crinoidal limestones with lenticles of chert and silicified Spiriferids. A low horizon in the Black Rock Limestone can be seen in a quarry [ST 592 505] 900 yd north-west of Hunter's Lodge Inn where thinly bedded, black, crinoidal limestone with nodules and strings of chert and silicified Spiriferids and Zaphrentids dips at 30° to the south. Higher beds in the formation, consisting of dark grey, compact, crinoidal limestones, are exposed in a series of low cliffs immediately south-east of Lower Pitts Farm.

The Vallis Limestone, which is thick enough to be mapped separately a little to the south-east of Lower Pitts Farm, is best exposed in an old quarry [ST 547 499] 280 yd south-west of Hunter's Lodge Inn. Here very strongly jointed, coarse-grained, light grey, crinoidal limestone is exposed and contains a brachiopod fauna which includes Chonetes (Megachonetes)cf. magnus Rotai, and Productus (Linoproductus) sp. Another old quarry, 400 yd E. 30° S. of the inn, in the same formation, yielded a fauna including Caninia aff. subibicina McCoy and 'Zaphrentis'konincki aff. kentensis Garwood.

The Burrington Oolite is poorly exposed in this area. Its junction with the overlying Clifton Down Limestone can be seen in an old quarry [ST 536 499] 550 yd south-east of Lower Pitts Farm, in which white oolite dipping at 35° to the south-south-west underlies calcite-mudstone.

Exposures of the Clifton Down Limestone are limited to old quarries south-south-west of the Batch, Priddy in which typical Lithostrotion beds are displayed.

The Hotwells Limestone crops out in the angle formed by the intersection of the South Western Overthrust and the Priddy Fault. Adjacent to the Priddy Fault, on the east side of the road, are seen a series of crags in which calcite-mudstones of the Clifton Down Limestone, dipping at 35° to the south-south-west, are overlain by the lowest beds of the Hotwells Limestone which consist of coarse oolitic crinoidal limestones containing Palaeosmilia, Dibunophyllum and giganteid Productids. A strong chert nodule horizon is present 70 to 80 ft above the base. F.B.A.W.

North of the Priddy Fault

Boreholes have yielded most of the information on the Lower Limestone Shale sequence in this area. The only surface exposure is a laneside section [ST 534 519] 0.5 mile north-east of Priddy church, which shows brown sandy shales. A borehole, 400 yd north-east of Priddy church, in the bottom of a 21 ft well (Whittard 1949, p. 480) proved dark grey and some pinkish crinoidal limestone with subordinate shale to a depth of 45 ft, overlying dark grey calcareous shale with some limestone beds, 0.5 to 1.5 ft thick, to a total depth of 115 ft. Five bands of bryozoan limestones, 6 inches to 1.5 ft thick were encountered, two of these being in the top part of the boring. The dip of the beds was 20°. The spoil from an adjacent well 24 ft deep consisted of a mixture of thinly bedded limestones and hard blue-grey shale. Another borehole [ST 535 525] 1425 yd N. 33° W. of Priddy church, estimated to start about 50 ft below the top of the Lower Limestone Shale proved the following succession: shale to 20 ft, limestone to 70 ft and shale to 123 ft (Richardson 1928, p. 118). The dip here is estimated to be about 30° to 40°. These two boreholes prove different parts of a limestone group in the upper part of the Lower Limestone Shale which has been recognized elsewhere on Sheet 280 (see p. 17). An abortive water bore [ST 539 524] 1 mile north-east of Priddy church, proved Lower Limestone Shale, dipping at 33° to 44°, to a depth of 350 ft under a cover of 90 ft of Triassic rocks (see p. 202).

The lowest beds with abundant brachiopods of the Black Rock Limestone are seen in numerous small exposures along the strike of the rocks between Priddy church and Will's Farm to the north. The main chert horizon is seen north-west of Priddy and can be traced eastwards towards the Stock Hill Fault, the best exposures being in old mining trenches about 0.25 mile north and north-west of Chancellor's Farm, where the beds measure about 100 ft in thickness. The top 200 ft of the Black Rock Limestone consists of granular crinoidal limestones with scattered Caninia cylindrica, as may be seen in old quarries 110 yd W. of Priddy church.

The Burrington Oolite is seen in natural outcrops and small quarries [ST 531 515], 1 mile W. of Priddy church, which together expose about 150 ft of massive, grey, oolitic and oolitic-crinoidal limestones at the base of the formation. Discontinuous sections of the Burrington Oolite, here some 525 ft thick, are seen in crags on either side of the Priddy–Cheddar road in the area of Cheddar Head Farm. Other exposures in the Burrington Oolite lie about 1.5 miles east-north-east of Cheddar Head Farm, where old mining trenches expose a nearly complete section. The strata, which are vertical, have a total thickness of about 700 ft and form a remarkably uniform series of grey oolites and oolitic-crinoidal limestones. This great increase of thickness of oolite accompanied by a corresponding diminution in thickness of the overlying oolite-mudstone beds of the Clifton Down Limestone suggests that a local replacement of the latter by an oolite facies has occurred (cf. the Wells area).

The divisions of the Clifton Down Limestone referred to above (p. 19) are present in the area north of the Priddy Fault. The lowest division shows a similar sequence to that in Cheddar Gorge, being subdivided into the Cheddar Limestone (at the base), the Cheddar Oolite, and banded oolite-mudstone beds. The Cheddar Limestone, some 120 ft in thickness, is best seen in crags on either side of the Priddy–Cheddar road in the area of Cheddar Head Farm, where it consists of black splintery limestones and dark grey granular crinoidal limestone occasionally oolitic near the base. The Cheddar Oolite is 125 ft thick and includes in the middle, a thin bed of black splintery limestone with Lithostrotion and one or more impersistent 1-ft beds of calcite-mudstone near the top. Both the Cheddar Oolite and the overlying beds are displayed in crags [ST 502 516], about 0.25 mile south-south-west of Bristol Plain Farm, just south of the lane (New Road) where the succession beneath the Lithostrotion beds is as follows:

Between here and the Cheddar Head Farm area scattered outcrops show a similar succession. In crags [ST 495 522] 1050 yd S. 53° W. of Cheddar Head Farm the lowest bed above the Cheddar Oolite yields Composita and Davidsonina carbonaria. Between these crags and the mining ground north-east of Chancellor's Farm there are no good exposures. The Cheddar Limestone and Oolite cease to be separable from the overlying oolite-mudstone beds, and sections in the sides of mining trenches show only about 100 ft of calcite-mudstones, pale oolites, and dark splintery mudstones lying between the Burrington Oolite and the Lithostrotion beds (see above). At one point, 1050 yd N. 18° E. of Chancellor's Farm, abundant silicified specimens of D. carbonaria were noted at the base of the Lithostrotion beds. The Lithostrotion beds and the succeeding calcite-mudstone beds are typically developed and well exposed throughout the area and call for little comment. The former average some 180 ft in thickness and the latter vary from 200 ft in the south near the Priddy Fault to 275 ft in the east.

The Hotwells Limestone, which is well exposed throughout the area, consists of massive grey crinoidal limestones having a sharply defined base and with a chert level, associated with black thinly bedded limestones, some 50 to 60 ft above the base. The greatest exposed thickness of Hotwells Limestone is in the area south of Cheddar Cliffs where some 350 ft are present, bounded to the south by a thrust fault north of Bradley Cross which repeats the lower part of the succession. Good exposures of the lower beds of the Hotwells Limestone occur in crags [ST 479 534] 1000 yd N. of Carscliff Farm, immediately south of the thrust mentioned above, where the following section is exposed:

Many crag sections are seen on either side of the road leading to Cheddar for a distance of 1.5 miles north-west of Cheddar Head Farm. Crags 150 yd south-east of Wellington Farm, expose the chert horizon from which Caninia cf. subibicina and Lithostrotion junceum were collected. The beds are rather disturbed due to the proximity of a small fault. In an exposure [ST 479 540] 1100 yd W. 6° N. of the farm, Carcinophyllum vaughani was collected from limestone crags just above the chert horizon.

To the north of Priddy, the strata along the northern edge of the pericline are inverted so that the Hotwells Limestone underlies the Clifton Down Limestone. Small quarries [ST 527 537] about 600 yd south-south-east of Haydon Grange exposed grey massive crinoidal limestone, dipping at 65° to the south, from which Carcinophyllum aff. welchi Ryder was collected. G.W.G.

North Hill Pericline, Eastern Half

This area is bounded to the west by the Stock Hill Fault and to the south by the Emborough Thrust.

Castle of Comfort Inn–Niver Hill–Tor Hole

Though the Lower Limestone Shale is nowhere exposed it can in places be detected by augering and is seen as a yellowish-grey rather gritty clay.

About 75 ft below the top of the Black Rock Limestone in the West End area a chert-nodule horizon is developed but this cannot be traced east of the Eaker Hill Fault. It is seen in an old quarry [ST 567 529] 1000 yd west-south-west of Grove Farm, where the strata are inverted and dip at 70° to the south. The upper part of the Black Rock Limestone is discontinuously exposed in the valley running north-eastwards from Tor Hole, where it comprises black crinoidal limestones with Caniniid corals.

Both the Vallis Limestone and Burrington Oolite are well exposed in the Tor Hole valley, the junction between the two formations being arbitrarily taken at the point above which ooliths first make their appearance.

The usual threefold subdivision of the Clifton Down Limestone can be recognized in this area (Figure 3) though the subdivisions are less sharply defined than is the case farther west. The basal oolite–mudstone beds everywhere rest sharply on the Burrington Oolite. A borehole, 75 yd north-west of the Castle of Comfort Inn, proved Clifton Down Limestone to a depth of 200 ft under a cover of Lias limestone (see Appendix II, p. 203). The Lithostrotion beds are seen in a series of old quarries [ST 566 533], north of West End, and the succeeding calcite-mudstones in crags [ST 582 535] on either side of the valley 650 yd E.N.E. of Grove Farm. A quarry [ST 581 537] 650 yd N.E. of Grove Farm, on the east side of the road, exposes 45 ft of dark grey to black calcite-mudstone, algal limestone, and porcellanous oolite, with shale partings in the top four feet. These beds are overlain by massive Hotwells Limestone. A wedge-shaped infilling of (?) Triassic breccia is seen in the western wall of the quarry at the top of the Clifton Down Limestone. Another quarry, on the opposite side of the road exposes normal calcite–mudstones dipping at about 45° to the north-east, and cut by a series of small north-easterly trending vertical tear faults with associated folding.

The Hotwells Limestone is only exposed adjacent to the Biddle Fault, where in a small quarry, 900 yd W. 39° N. of Grove Farm, grey granular limestone yielded Palaeosmilia murchisoni. It was also reported by Welch (1929, p. 52) from a well at Greendown (see also p. 82). The northward extension of the Eaker Fault, shown on the maps 0.33 mile east of Green Down, is conjectural as are the outcrops of Hotwells Limestone on either side. In the quarry near Grove Farm [ST 581 537], mentioned above, about 100 ft of Hotwells Limestone are seen; the rocks are mostly massive, grey to dark grey, granular limestones, but 25 ft above the base lie some 35 ft of thin-bedded limestone with beds ranging from 2 in to under 2 ft in thickness. There are further good exposures of Hotwells Limestone in the valley sides 150 yd south-east of the quarry.

East Harptree

The small inlier of Carboniferous Limestone, about 0.5 mile south of East Harptree shows an inverted succession of Clifton Down Limestone. The upper calcite-mudstones are seen in small exposures in the eastern part, and beds with Lithostrotion in a small overgrown quarry [ST 568 548] in the western part. The dip of the beds is 40° to 55° to the west-south-west. G.W.G.

Bathway, Chewton Mendip

Good exposures of Clifton Down Limestone occur to the north-west of Bathway. In the ravine 850 yd north-west of Bathway there is a discontinuous section of splintery limestone, calcite-mudstone and oolite, the fullest section being in a large quarry (Willet's Lane Quarry) [ST 588 529] where the junction of the Clifton Down Limestone and Hotwells Limestone was exposed. Davidsonina septosa is here recorded the first time for the Mendip region. The section is as follows:

Hotwells Limestone, similar to that seen in the Willet's Lane Quarry, succeeds the Clifton Down Limestone at the north-east end of the ravine 850 yd north-west of Bathway. In a small inlier [ST 600 523], 950 yd south-east of Chewton Mendip church, dark fine-grained limestone with Productids dips at 25° to the north-east.

Nedge Hill–Red Hill–Green Ore

This area comprises the eastern nose of the pericline and that part of the southern limb of the pericline which extends westwards to the Biddle Fault.

The lower brachiopod-rich beds of the Black Rock Limestone are seen in several small quarries, on either side of the main road, at Nedge Hill. In an old quarry [ST 582 507] 640 yd north-east of Green Ore was seen much-veined dark grey crinoidal limestone dipping 40° to the south, and containing very large Caniniids, Fasciculophyllum densum (Carruthers) and Spirifer cf. konincki Dewalque.

Vallis Limestone, consisting of pale grey coarse crinoidal limestone, is seen in a number of shallow overgrown pits 725 yd south of East End. Burrington Oolite, comprising dark oolitic limestone, is exposed dipping at 35° to the east-north-east in a shallow pit 220 yd south-east of East End. White oolitic limestone of the same formation is seen dipping at 40° to 45° to the south-south-east in a series of crags [ST 592 505] 1500 yd east-north-east of Green Ore, and at 30° to the south 850 yd west of Green Ore.

The Clifton Down Limestone is rather poorly exposed. An old quarry [ST 600 512] showed splintery black limestone with chert and silicified Lithostrotion dipping at 38° to 40° to the north-east. The rock here is much fissured the joints being filled with limonite. In a copse adjoining the road, 550 yd south-east of Green Ore, old workings expose calcite-mudstone with bands of silicified Lithostrotion dipping at 30° to the south-south-east, although close to the Emborough Thrust, the beds show no signs of disturbance.

Hotwells Limestone, comprising coarse oolitic limestone and pseudobreccias, was seen dipping at 30° to the north-east in lane sides 550 yd west-south-west of Emborough church. A few very indifferent exposures of similar rocks dipping to the south-east occur 500 yd west-south-west of Emborough Grove.

Biddle Fault (near Tower Hill)–Stock Hill Fault

The Lower Limestone Shale was shown by augering to consist mainly of olive-grey shale.

The Black Rock Limestone has a general southerly dip of 18° to 20°, its junction with the underlying shales being marked by a line of swallets. A quarry [ST 563 497] on the south side of the road, 1500 yd east-south-east of Hunter's Lodge Inn showed compact, well-bedded grey crinoidal limestone dipping at 20° to the south-south-east and containing abundant large Caninia cylindrica and Bellerophon sp.F.B.A.W.

Stoke Woods Syncline

Bounded on the north by the South Western Overthrust, and on the east by the Stock Hill and Rookham faults, the area is limited on the south-west by the Ebbor Thrust complex. East of Ebbor Rocks and Higher Pitts Farm, the northern limb of the Stoke Woods Syncline merges into the western nose of the Pen Hill Pericline.

Bradley Cross

The small inlier of Carboniferous Limestone 0.25mile south-east of Bradley Cross is probably part of the southern limb of the Stoke Wood Syncline. The dips here are 36° to 42° in an east-north-easterly direction. At the western tip of the inlier, a small quarry 300 yd north-east of Lyde Farm, exposes coarse reddish crinoidal limestone at a stratigraphically low level in the Black Rock Limestone. Some 80 ft or more of tough, black, fine-grained dolomite, seen in crags 600 yd north-east of the farm, separate the Black Rock Limestone from exposures of Burrington Oolite in the north-eastern part of the inlier.

Westbury Beacon

This area lies west of the Priddy Fault. Black Rock Limestone occupies the core of the Stoke Woods Syncline north and east of Carscliff Farm where there are numerous exposures of the coarse crinoidal and brachiopod limestones typical of the lowest part of the succession. Farther south, the uppermost beds of the Black Rock Limestone consist of dolomitized granular crinoidal limestones. Although the beds lie close to the South Western Overthrust they show little sign of disturbance and dip at 20° to 30° to the south-west. Burrington Oolite forms the high craggy ground about 0.5 mile north-east of Draycott. Here the fold pitches 12° S.S.E. with dips of 20° to 40° on either side of the axis. G.W.G.

The greater part of the high ground around Westbury Beacon is occupied by Burrington Oolite which is seen in numerous rocky outcrops. The general dip of the rocks on the northern limb of the syncline in this area is 35° to 40° to the south-south-west. In the western part of Stoke Woods, an inlier of Carboniferous Limestone is separated from the main mass by a tongue of Dolomitic Conglomerate.

Here the rocks, which dip at 40° to 45° to the east-north-east, consist of Burrington Oolite underlain by some 300 ft of massive grey dolomitized limestone.

The Clifton Down Limestone is well exposed in natural outcrops south and west of Westbury Beacon. The lower beds are calcite-mudstones and oolites which are succeeded by splintery limestones full of silicified Lithostrotion.

Priddy to Rookham faults

A small ill-exposed outcrop of Lower Limestone Shale, consisting largely of shale, occurs at the foot of the limestone scarp on the western side of the Rookham Valley, west of the Rookham Fault. In the eastern part of the area, the Black Rock Limestone has a broad outcrop consequent upon low and undulatory dips. The lowest beds, consisting of thinly bedded, dark grey, crinoidal limestones with chert and silicified fossils are to be seen in a highly disturbed and faulted condition in the crags on the western side of the Rookham Fault. At a quarry [ST 551 492] 1050 yd S.S.E. of Hunter's Lodge Inn highly fossiliferous dark grey crinoidal limestone dipping at 40° to the north-east yielded large Caniniids, Zaphrentids and Aulophyllum sp.F.B.A.W.

South of Rookham, in the area north of the Ebbor Fault belt, exposures on either side of the trackway (Tyning Lane) leading from Upper to Lower Milton show the upward passage of dolomitized crinoidal limestones at the top of the Black Rock Limestone into oolitic-crinoidal limestones at the base of the Burrington Oolite. Farther west, crags [ST 540 478] in the hillside north of the lane expose white massive dolomite. This rock forms a prominent band some 80 to 90 ft thick lying about 170 ft above the base of the Burrington Oolite. G.W.G.

Elsewhere, the Burrington Oolite is poorly exposed. Weathered oolite is seen in the sides of the 50 to 60 ft deep solution cavity [ST 532 499] known as Sandpit Hole and situated 800 yd N.N.W. of Higher Pitts Farm.

The threefold division of the Clifton Down Limestone (see p. 19) is well developed in this area and the most complete section is seen in the gorge (Ebbor Gorge) east of Ebbor Rocks^‡15  (Figure 5). The exposures begin 450 yd west of Higher Pitts Farm in the lower division of calcite-mudstones and oolite, from which Davidsonina carbonado was collected. The top of the succeeding silicified Lithostrotion beds descends to the valley floor 800 yd W. 30° S. of Higher Pitts Farm and thereafter, the gorge is cut in gently folded rocks of the upper calcite-mudstone beds; these are traversed by a fault or very sharp monoclinal fold, 1300 yd south-west of Higher Pitts Farm which introduces Hotwells Limestone. This folding may be traced in the Clifton Down and Hotwells limestone north-westwards from the gorge for about a mile. The Lithostrotion beds are also well exposed in old quarries [ST 517 498], 1 mile south-west of Priddy, where they dip at 40° to 45° to the south-west. At Broadmead Quarry [ST 505 504], 1 mile north-north-east of Westbury, over 250 ft of the uppermost calcite-mudstone beds are exposed, the junction with the Hotwells Limestone lying 50 yd S. of the quarry.

The Hotwells Limestone is seen in a series of natural exposures in a small valley 350 yd south-east of Broadmead Quarry. The lowest beds are coarse oolitic-crinoidal limestones with Dibunophyllum and Palaeosmilia and include a well marked chert horizon 70 ft above the base (see p. 21). The higher beds consist of rather unfossiliferous, massive, slightly oolitic, splintery limestone associated with bands of 'rubbly limestone'. In the Ebbor Rocks area, the Hotwells Limestone has a wide outcrop due to the folding, but except for the cliff in Ebbor Gorge, sections are scanty. The uppermost beds of the Hotwells Limestone show a development of hard reddish oolite with a band of chert developed in the top.

Pen Hill Pericline (east of the Rookham Fault)

The northern limit of the pericline is marked by the line of the Emborough Thrust. To the south it passes into the Binegar Syncline and the folded ground west of the Slab House Fault.

Rookham Fault to Biddle Fault

Adjacent to the Rookham Fault a disturbed sequence through the Lower Limestone Shale was proved in the Rookham Borehole (see Appendix II, p. 206). The lowest beds include coarse-textured, red-purple crinoidal limestones which crop out at the bottom of Rookham Hill dipping at 40° to the north-west and in small exposures in Rookham Wood [ST 548 483], north-eastwards the limestones appear to be replaced by shale. Higher beds are well exposed in the banks of an old pond [ST 567 492], 950 yd W. 20° S. of Hill Grove, in which olive-yellow sandy shales show an inverted dip of 60° to the south-south-east.

The Black Rock Limestone occupies the remaining ground south of the Emborough Thrust and the Rookham Fault. Formerly exposed in old mining trenches the rocks showed inversion with dips of 60° to 82° to the south-south-east. F.B.A.W.

On the southern limb of the pericline, the Lower Limestone Shale outcrop is mostly concealed by Liassic rocks. About 0.25 mile west-south-west of West Horrington, the limestones in the middle of the sequence form a well-marked feature adjacent to the Biddle Fault. Farther south, the succeeding Black Rock Limestone is seen in scattered exposures along sides of the wooded valley leading towards the Mental Hospital. The southerly dips increase from 15° to 20° in the north to 75° or over in the south seen at points within about 100 yd of the Ebbor Fault (see p. 140). The main chert horizon is present a little to the north of the fault and is succeeded by some 150 to 200 ft of crinoidal limestones and dolomites. The Clifton Down Limestone outcrops south of the Ebbor Fault. D.R.A.P.

East of the Biddle Fault to Wells Road

On the north side of the pericline, a narrow band of Lower Limestone Shale extends eastwards as far as Hill Grove, beyond which, faulting has brought Old Red Sandstone against Black Rock Limestone, the contact being marked by a line of solution cavities. Still farther east, the Lower Limestone Shale reappears in a wide tract of swampy ground south-west of Whitnell Farm, where augering showed purplish sandy micaceous shale to be the main rock type. On the south side of the pericline, the Lower Limestone Shale occupies a wide depression west of the fault at Slab House, in which the dips are generally 30° to 40° to the south-south-east but, locally, sharp isoclinal folding is present in the softer beds. In places, as in Biddle Combe, thin crinoidal bands are developed towards the base.

Along the northern limb of the pericline all the strata are inverted, with dips of 60° to 80° to the south. Exposures of the different formations close to, and along the east side of, the main Wells–Bristol road may be seen at the following places. A small quarry [ST 604 495] 450 yd west of Whitnell Farm shows dark crinoidal Black Rock Limestone with abundant Orthotetids. Vallis Limestone resting on Burrington Oolite may be seen in a quarry [ST 603 498] 600 yd north-west of Whitnell Farm, whilst in the large roadside quarry [ST 604 501], known as Quar Tyning, 700 yd north-west of the farm, are to be seen well-bedded calcite-mudstones and dark grey fine-grained limestone with chert. These beds, in which many bands of Lithostrotion and papilionaceous Chonetids are present, occur in the top half of the Clifton Down Limestone. To the north of Quar Tyning, a narrow outcrop of Hotwells Limestone consisting of reddish, oolitic-crinoidal limestone and pseudobreccia is bounded by the Emborough Thrust which brings it into contact with Coal Measures shales.

On the southern limb of the pericline, the lowest beds of the Black Rock Limestone are seen in a series of low crags between the Haydon Farms and the Biddle Fault, where the rock consists of thinly-bedded, dark grey, crinoidal limestone rich in silicified Unispirifer tornacensis, The dip is 20° to 30° to the south-south-east. Higher beds of the same formation are to be seen in a quarry [ST 584 480] 220 yd south of Haydon Farms. Here, dark grey crinoidal limestone with Caniniid corals dip at 12° in a general southerly direction. F.B.A.W.

Many exposures of the lower beds in the Black Rock Limestone are seen in crags and shallow workings north and east of West Horrington. The main chert beds are exposed in the valley 750 yd north of the Mental Hospital chapel. Higher beds in the Black Rock Limestone may be observed in a roadside quarry 400 yd N. 30° E. of the chapel. Here, 10 ft of crinoidal limestone overlies 30 ft of dolomite, whilst valley-side exposures, 500 yd east-north-east of the chapel, show a passage of dark grey crinoidal and dolomitic limestones into the succeeding Burrington Oolite. The latter consists of oolitic-crinoidal limestone and dolomite. The dolomite attains a thickness of 200 ft in this area. D.R.A.P.

Emborough–Gurney Slade to Binegar

This area includes the eastern nose of the Pen Hill Pericline and the Binegar Syncline, to the south. The exposures are most conveniently described from south to north.

Cherty, crinoidal Black Rock Limestone dipping at 25° to 30° to the north-north-east is exposed along the lane (Rownmead Lane) 0.75 to 1 mile south of Binegar church. Old quarries in Binegar Bottom, a dry valley extending south-westwards from Gurney Slade, show good sections of typical Vallis Limestone, pale grey coarse-grained, oolitic-crinoidal limestone. Fossils collected from the quarry 1200 yd S. 3° W. of Binegar church included Caninia sp.cf. C. caninoides (Sibly); from the quarry 1000 yd S. 5° W. of the church Palaeosmilia cf. murchisoni, 'Zaphrentis' konincki Edwards and Haime; Productus (Pustula)cf. nodosus (I. Thomas), and from the quarry 900 yd S. 5° W. of the church Palaeosmilia murchisoni. From all three quarries, Chonetes (Megachonetes) sp.was also collected. The succeeding Burrington Oolite occurs in a series of crags [ST 617 487] on either side of Binegar Bottom where it dips at 28° to 30° to the north-east. The largest exposure, however, is in a long railway-side quarry [ST 621 500], 850 yd north-east of Binegar church, in which fine-grained oolite with Caninia dips at 30° to the east. It is also exposed in the Emborough (western) Quarry (see p. 137).

The lower part of the Clifton Down Limestone around Binegar includes a local but strong development of oolite. At Highcroft Quarry [ST 619 488], 900 yd south-east of Binegar church, the beds above the oolite are worked and consist of dark grey limestone and calcite-mudstone, locally cherty, with abundant sheets of Lithostrotion and papilionaceous Chonetids. The dip is 35° to the north-east. In the large, now abandoned, Binegar Quarry^‡16  0.33 mile east of Binegar church, similar beds of Clifton Down Limestone were exposed both above and below a strongly jointed oolite 150 ft thick. The dip is 38° to 40° to the south-east. The overburden, both here and at Highcroft Quarry include silicified Jurassic material (see p. 109). In Gurney Slade village, on the east side of the main road is a line of crags consisting of splintery limestone with Lithostrotion. Typical splintery Clifton Down Limestone rock types are also seen in Pearce's Hill Quarry [ST 623 495] and large abandoned quarries [ST 623 500], [ST 625 500] on either side of the road at the top of Marchant's Hill. The latter include huge wall-like Jurassic infillings left by quarrying (see p. 106).

Hotwells Limestone is exposed on either side of Gurneyslade Bottom, a valley leading north-east from Gurney Slade towards Blacker's Hill. On the south side are several deep disused quarries in Hotwells Limestone capped by Dolomitic Conglomerate. On the north side is the large Gurneyslade Bottom Quarry [ST 627 494] in which massive, oolitic-crinoidal limestone with bands of Productids is seen to dip at 30° to 48° to the south-east. Higher beds including bands of rubbly limestone are seen in the south-eastern part of the quarry. About 700 yd to the north is the large disused Cock Hill Quarry [ST 625 500] in which are exposed massive oolitic limestones rich in Productids of the hemisphaericus and giganteus type. Clisiophyllids are also numerous. The dip is 35° to the east. Parallel, close-spaced sets of fissures filled with Mesozoic material (see p. 106) form a striking feature of this exposure. They are probably part of a system of radial joints due to tension developed in the periphery of the Pen Hill periclinal fold.

In the north of the area are the great Emborough Quarries [ST 620 516], [ST 623 516] consisting of two roughly rectangular excavations on either side of the railway. The greater part of the quarries have been cut in the Clifton Down Limestone which is some 750 ft in total thickness, and consists of dark grey splintery limestone with Lithostrotion, oolites and calcite-mudstones. In the western quarry the Burrington Oolite is exposed in the southern face. The junction with the overlying Hotwells Limestone extends in a general north-westerly direction on either side of the railway viaduct south pier. The Hotwells Limestone itself comprises oolitic-crinoidal limestones with Palaeosmilia and Clisiophyllids, and, in its higher parts, beds of fossiliferous 'rubbly limestone'. These are well exposed in the sides of inclines leading from the quarry plant in the eastern quarry down into the western quarry. Bands of purplish shale with calcareous nodules also occur. Dips in the quarries are in a general north-easterly direction and range from 42° in the south to 60° to 70° farther north. In the western part of the quarried area, however, local inversion of the strata accompanied by tear faulting results in a southerly dip of 30° to 32°.

In the eastern side of the small valley, some 500 yd south-east of Emborough church the junction of inverted reddish Hotwells Limestone with the Quartzitic Sandstone Group can be seen; farther east this boundary appears to be faulted. F.B.A.W.

Beacon Hill Pericline

Shepton Mallet area

This area is largely covered by Liassic rocks, through which Carboniferous Limestone appears as small inliers.

About 2 miles east-north-east of Shepton Mallet the extreme narrowness of the Lower Limestone Shale outcrop north-east of Newman Street is presumed to be the result of structural disturbance. To the south, the lower crinoidal limestones and the main chert horizon of the Black Rock Limestone are exposed in a number of small quarries. The dip is about 40° to the south.

About 0.75 mile north-east of Shepton Mallet, an inlier of Black Rock Limestone appears through the Lias limestones, the exposures being confined to the immediate vicinity of the Fosse Way. In the north of the inlier fine-grained splintery limestones dip at 10° to the west, and, in the southern half of the inlier they are folded into an anticline with dips of 28° and 40°, to the north and south respectively. Dolomitic limestone brash occurs in the fields to the west. The horizon is probably high in the Black Rock Limestone sequence.

About 1100 yd N. 10° E. of Shepton Mallet church, a tiny inlier of Black Rock Limestone, consisting of about 70 ft of crinoidal limestone with much chert in its upper part, is exposed behind factory buildings. The dip is 65° to the south. About 0.25 mile to the south-south-east, immediately south-west of the railway viaduct, an old quarry [ST 620 443], west of the road, shows Black Rock Limestone comprising some 30 ft of crinoidal limestone, apparently dipping southwards at ? 80°. About 80 yd farther to the south-east, the cutting on the east side of the main road shows Lias and Rhaetic limestone and conglomerate resting on much fissured and veined coarse crinoidal limestone. The horizon lies at about the Black Rock Limestone–Burrington Oolite junction. At Bowlish, 200 yd north of the crossroads, an old quarry shows crinoidal limestones, near the top of the Black Rock succession, dipping at 35° to 60° southwards truncated by gently dipping Lias limestones. The presence of the small inlier on the map, 500 yd north-east of this quarry, is based on the evidence of dolomitic crinoidal limestone in field brash.

North and north-west of Darshin, a line of cliffs [ST 605 441] in the wood (Darshill Wood) exposes Burrington Oolite overlain by Clifton Down Limestone. The dips range from 55° to the south-south-west in the east, to a general south-westerly dip of 25° farther west in the Clifton Down Limestone.

Windsor Hill–Ham Woods–Thrupe

Exposures in the Lower Limestone Shale are infrequent. In the stream bed, 200 yd south of Burnt House Farm, silty shales with limestone bands are exposed dipping westwards. North-east of Thrupe, mudstones and shales dipping at 20° westwards are visible in a ditch [ST 606 461] on the west side of the railway immediately south of a bridge over the line. About 0.125 mile farther north, contorted shales are seen in the stream bed [ST 606 463] which passes under the railway embankment.

The lower half of the Black Rock Limestone is seen in large disused quarries in the Windsor Hill area. A composite section of the strata exposed in the four main quarries is as follows:

The lowest beds lie about 30 ft above the top of the Lower Limestone Shale.

In the Windsor Hill quarries [ST 614 452], adjoining the east side of the railway, the lowest 200 ft of this succession are exposed. The dip is in a general southerly direction at 24° to 34°. Fossils collected in the lowest 40 ft included Zaphrentites delanouei; Syringothyris cuspidata cyrtorhyncha North, Rhipidomella michelini (Leveille), and Unispirifer tornacensis. An almost identical section is exposed in a quarry (Downside Quarry) [ST 618 451] to the south-east, where there is much disturbance along a north-easterly trending fault, which also affects the nearby Lias rocks. On Windsor Hill, the Black Rock Limestone is folded into a syncline whose axis pitches westwards across the line of the railway tunnel through the top of the hill. The rocks on the southern limb are seen near the farm (Rosamond Green Farm) 1200 yd W. 7° N. of Downside crossroads, where the dip is 35° N.N.W. The top 200 ft of the succession given above are seen in the Ham Woods Quarry [ST 610 453], west of the railway. Occasional exposures of the limestones above the chert beds are seen in the Ham Woods valley. At Thrupe, the chert beds are exposed in small quarries [ST 604 457] east of the road. About 0.25 mile south-east of Washingpool Farm a wartime trench in a wood revealed the chert beds dipping at 25° to the south-east and overlain by decalcified limestone and dolomite. At one point the trench cut across part of an elongated north-westerly trending area of red ochreous clay; probably a fissure filling. Nearby, the country rock is locally much altered—a specimen (E26771) being described by Mr. R. W. Elliot as a quartz-chalcedony dolomite with locally common granules of haematite in the interstitial carbonate.

The best exposures of the Clifton Down Group are in Ham Woods, where the succession may be summarized as follows:

The Burrington Oolite totals some 750 ft but the lower part, consisting mainly of oolitic-crinoidal limestones, is poorly exposed. Minor faulting affects the Burrington Oolite and the lower part of the Clifton Down Limestone and some 20 ft of the latter may have been cut out. Two small outliers of Burrington Oolite have been mapped a mile west of Maesbury. They appear to have been introduced by faulting and folding, the details of which are obscure.

Maesbury Fault–Ashwick–Withybrook

The railway cutting west of Maesbury Castle provides an excellent exposure of the top two-thirds of the Lower Limestone Shale. The succession is as follows:

The lower part of the Black Rock Limestone is exposed in a number of small quarries between Maesbury Castle and one mile to the west. The approach cutting on the south side of a recent trial quarry [ST 597 475] 900 yd south-east of Slab House, drove through the Maesbury Fault (see Plate 4B) and 'an associated branch fault. In the quarry the beds are folded and much shattered, particularly so in the approach cutting which exposes a lead-zinc vein and Liassic fissure deposits (see pp. 106, 164). Fossils collected include Caninia aff. cornucopiae, Cyathaxonia cornu Michelin and Fasciculophyllum omaliusi. North of Maesbury Castle the railway cutting exposes the basal beds of Black Rock Limestone dipping at 25° northwards. Farther east, the Black Rock Limestone is poorly exposed and the base is taken below a line of swallets.

The Vallis Limestone, in this area, consists of light grey oolitic-crinoidal and crinoidal limestones. Old quarries, 350 yd north of Oakhill church, show 125 ft of these beds, which are also seen dipping at 50° to the north-north-west in a small quarry 100 yd east of the crossroads at Withybrook. The overlying Burrington Oolite is seen around Ashwick Grove and forms a prominent bluff 250 yd south-east of the mansion.

Clifton Down Limestone with dips of 70° to 85° to the north-north-east is found at intervals for a distance of 350 yd south-south-west of Badger's Cross. Farther east the lower part of the Clifton Down Limestone, including the beds with Lithostrotion (see p. 19) are poorly exposed, while the upper part, consisting predominantly of calcite-mudstone, is well exposed between Ashwick Grove and Batch Farm. Old quarries [ST 654 476] a little to the east-north-east of the farm expose about 50 ft of calcite-mudstone with bands of chert nodules. The beds dip at about 50° to the north.

Hotwells Limestone consisting of coarse crinoidal limestone dipping at 45° to 50° north-north-east is exposed in crags 350 yd east of Ashwick church and farther east in the valley sides in Harridge Wood. An old quarry [ST 654 477], 300 yd north-east of Batch Farm exposes the basal beds of the Hotwells Limestones dipping at 45° to the north-north-east and consisting of 95 ft of granular limestone and oolites. D.R.A.P.

South of the Ebbor Fault Belt

This belt of much disturbed Carboniferous strata lying to the south of the Ebbor Thrust and associated faults, includes all the Carboniferous Limestone outcrops not hitherto described. These rocks are exposed in discontinuous outcrops on the southwestern side of the Mendips and in a series of small inliers south of the main hill mass. The Ebbor Thrust itself is not seen west of the Ebbor Rocks area though it is presumed to continue north-westwards towards Cheddar. The Thrupe Fault probably represents the eastward continuation of either the Ebbor Fault or an associated fault. G.W.G.

Nyland Hill

West of Draycott, there is a group of small inliers of Carboniferous Limestone centred round the steep hill mass of Nyland Hill. The inliers are separated from one another by a thin mantle of Triassic rocks. The Carboniferous strata dip at 30° to 40° to the north-east except in the easternmost exposures where the dip is 30° to the north. Black Rock Limestone crops out at the south-west corner of Nyland Hill [ST 456 504] where fine-grained, slightly dolomitized, dark grey limestone with bands of coarse crinoidal limestone yielded Caninia and Michelinia. Burrington Oolite forms the central and highest part of the hill, while the basal beds of the Clifton Down Limestone, consisting of calcite-mudstone, are seen in the small inlier east of Nyland Farm. The most easterly outcrop, 700 yd south-west of Draycott railway station, consists of a narrow ridge of limestone rising a few feet above the moor level. The rock is very weathered coarse crinoidal oolite, probably Burrington Oolite.

Lodge Hill, Westbury

The crest of this ridge is formed of highly dolomitized crinoidal limestone dipping at 30° to 40° to the north. A very small inlier [ST 484 484] of similar rock protrudes through the Triassic marl 1580 yd S. 8° E. of Rodney Stoke church.

Ebbor Rocks to Easton

South of the Ebbor Thrust in the Ebbor Rocks area (Figure 5) a sharp synclinal fold is developed with a complementary anticline to the south. In the immediate vicinity of the fault line the rocks are intensely shattered and recrystallized. Black Rock Limestone occurs in the north-west and consists of black crinoidal limestone with Zaphrentids. Exposures at several places [ST 515 491], [ST 514 491] show a sharp synclinal fold with dips of 50° and 60°. The succeeding Burrington Oolite is badly exposed. Old roadside quarries, some 1010 yd north-west of Ebbor Farm, show intensely shattered rock which is largely recrystallized. Small outcrops of Clifton Down Limestone form crags [ST 522 484] 600 yd north-west of Ebbor Farm, in which shattered cherty limestone with Lithostrotion are exposed.

Separated from and just south of, the main limestone mass is a narrow elongated outcrop of Carboniferous Limestone. The rock is a dark fine-grained dolomitized limestone and may be Clifton Down Limestone. F.B.A.W.

About 0.75 mile south of Easton, there are two small inliers of dolomite with a little dolomitized crinoidal limestone. In the northern area, the rocks are apparently folded into an anticline with dips of 14° to 30° and a general N.W. to S.E. axial trend. Fossils collected from a small overgrown quarry [ST 513 468] in the north of the inlier include: Athyris cf. lamellosa (Lévéille), Productus (Dictyoclostus)cf. vaughani, Unispirifer tornacensis and Syringothyris cuspidata cf. cyrtorhyncha. Mr. M. Mitchell reports that this fauna suggests a horizon in the lower part of the Black Rock Limestone. The southern inlier shows light grey massive dolomite dipping in a general southerly direction at about 20° to 50°.

Lower Milton, Wells

The south-westerly continuation of the Ebbor Thrust is seen west of Lower Milton as a structurally complex N.W. to S.E. aligned fault belt some 250 to 400 yd wide (Figure 12). Within this fault belt, scattered exposures of the upper calcite-mudstone beds of the Clifton Down Limestone are seen in the northern part, and Hotwells Limestone in the southern part. A 'fault slice', 30 to 60 yd wide, of shattered and recrystallized limestone, probably Clifton Down Limestone, forms the southern margin of the fault belt. The highest exposed beds in the Hotwells Limestone sequence are seen in natural outcrops just north of this 'fault slice'. No detailed sequence can be made out because of the structural disturbance; the rock types include black oolites, massive granular limestone with cherty patches, thin blue shales, and at one place [ST 5386 4738], a bed of current-bedded quartzitic sandstone, seen to 3 ft, underlies massive grey granular limestone.

South of the fault belt, the ground rises to a wooded eminence (Milton Hill) formed of Barrington Oolite which is well seen in quarries and natural outcrops. This formation, which consists of very massive grey oolites with a varying, but usually small, admixture of crinoidal debris, totals some 750 ft in thickness. The dip of the beds is on average 18° to 20° to the south. The now disused Milton Quarry [ST 539 471] 0.5 mile south-east of Wookey Hole church, shows very massive limestones with vertical calcite-filled fissures and red-stained joints trending in a north-north-easterly direction. Fossils collected included Carcinophyllum aff. mendipense and Palaeosmdia murchisoni. Underwood Quarry, 0.25 mile to the south, provides a magnificent exposure of similar rocks. Just beyond the eastern limit of the workings (in 1955) faulting brings in Clifton Down Limestone to the east, where the best exposures are in the Lithostrotion beds.

Stoberry Park, Wells

This inlier, north of Wells, consists entirely of the uppermost calcite-mudstone beds of the Clifton Down Limestone. The dip is consistently to the south-west and ranges from 15° to 36°. There are good exposures in disused quarries [ST 553 469] on either side of the main Bristol road and [ST 557 469], [ST 557 467] on either side of the lane (Beryl Lane) 0.25 mile west of Beryl. G.W.G.

Knapp Hill, Wells

The Clifton Down Limestone is exposed on Knapp Hill and the wooded valley (Biddle Combe) on its east side, south of the Ebbor Fault (see p. 41). Here calcite-mudstones underlain by Lithostrotion beds dip southwards at 20° to 30°, with, adjacent to the fault, steepening of the dip to 60° to 70°, associated with local folding and faulting.

Lyatt to Dinder

Burrington Oolite forms the core of the Lyatt Anticline, the axis of which passes through Dinder Wood, where massive oolites with northerly and southerly dips are exposed. North of the overturned and faulted northern limb, some 700 yd south-west of Chilcote Manor, poorly exposed oolite and oolitic crinoidal limestone has been mapped as Burrington Oolite. At one point, these beds yielded Lithostrotion bristoliense Vaughan. Between East Horrington and Crapnell Farm, this formation appears in discontinuous outcrops from beneath a cover of Mesozoic rocks, but the exposures are poor and the structure complicated. The rocks consist of oolites associated with dolomites and, west of Crapnell Farm, dark fine-grained limestone. West of Dinder Wood, the core of the Lyatt Anticline is formed of Clifton Down Limestone, exposures on both limbs of the fold are frequent. On the south side of the dry valley, 250 yd south-east of Chilcote Manor, exposures of Lithostrotion limestones, oolites and calcite-mudstones are assigned to the Clifton Down Limestone. D.R.A.P.

The exposures at the western end of the Lyatt Anticline are in the upper calcite–mudstones of the Clifton Down Limestone. Tor Hill Quarry, 0.5 mile east-south-east of Wells Cathedral, exposes a section across the core of the anticline which here consists of a rather complex series of small folds. In this area, the rocks on the northern limb of the anticline are much disturbed and the northernmost exposures, 600 yd W. of the cathedral, show Hotwells Limestone. The strata, consisting of massive grey granular limestone, are inverted with dips of 70° to 80° to the south.

Croscombe

A small inlier of Carboniferous Limestone, about 0.5 mile west of Croscombe, consists entirely of the upper calcite-mudstone beds of the Clifton Down Limestone. The frequent exposures show a progressive increase in the dip from 38° to the south at the southern edge of the inlier to approximately vertical along the northern edge. G.W.G.

Another small inlier is present about 0.5 mile south-east of the village. Here, the succession, which is inverted with southerly dips of 40° to 55°, consists of Hotwells Limestone underlying Clifton Down Limestone. A stream bed and crags [ST 600 439] in the central part of the inlier showed the lowest 100 ft of the Hotwells Limestone. These beds yielded Carcinophyllum aff. vaughani, Lithostrotion cf. junceum, L. cf. martini, L. pauciradiale, Palaeosmilla murchisoni; Chonetes sp.(cf. C. aff. comoides Vaughan 1905 pl. 26, fig. 4), C. (Megachonetes) sp., Productus (Gigantoproductus) maximus and P. (Linoproductus) hemisphaericus. D.R.A.P.

Dulcote Hill–Church Hill

The Carboniferous succession in this elongated inlier is inverted, Burrington Oolite, usually much disturbed, being underlain by Clifton Down Limestone. This, where the sequence is complete, shows the usual tripartite subdivisions (see p. 19) of lower mudstone-oolite beds (some 25 ft in thickness), Lithostrotion beds (160 to 170 ft) and upper calcite-mudstones (180 to 190 ft). Up to 150 ft of Hotwells Limestone crops out in the lower slopes on the north side of Dulcote Hill.

At the western end of the inlier, Dulcote Quarry [ST 566 443] exposes the lower two-thirds of the Clifton Down Limestone and the upper part of the Burrington Oolite. The most accessible part of the section is on the west side of the quarry where the rocks are vertical. Here, Davidsonina carbonaria was collected from the bottom part of the Lithostrotion beds which, at this point, due to a small fault, come into direct contact with the Burrington Oolite. At the eastern end of the quarry (in 1955), some 15 to 20 ft of black calcite-mudstone and pale oolite intervened between the two formations, though here also the section was disturbed. The Burrington Oolite both here and farther east is in a highly shattered condition and cut by numerous calcite veins. The adjacent quarry to the east [ST 569 441] exposes Burrington Oolite, while a further 100 yd east, the old Railway Quarry [ST 573 440] shows very massive Burrington Oolite faulted against Lithostrotion beds. In the north-east corner of this quarry, the lower mudstone-oolite beds of the Clifton Down Limestone are seen, but the nature of their junction with the Burrington Oolite is not clear. A small roadside quarry [ST 578 439] 900 yd south south-east of Dinder church, in Clifton Down Limestone under a cover of Dolomitic Conglomerate shows the Lithostrotion beds resting on the upper calcite-mudstone beds. G.W.G.

Farther east, Paradise Quarry [ST 589 438], 0.5 mile west-south-west of Croscombe church, exposes the lowest 80 ft of the upper calcite-mudstones of the Clifton Down Limestone. The beds are much disturbed and dip at 45° to 80° to the south. D.R.A.P.

South of Church Hill

About 0.75 mile south of Dinder, an inlier of Carboniferous Limestone forms a small but well-marked hill (Corrington; summit 445 ft O.D.) and extends some way south of the Sheet boundary (Figure 13). Good natural exposures prove a succession of some 550 ft of Black Rock Limestone with a strong chert development in the top 80 ft (the main chert, see p. 17) overlain by over 300 ft of light grey dolomite which forms the upper part and the southern slopes of the hill. The latter is a very hard massive rock which owes its hilltop position to its resistance to erosion. Farther east, about 0.5 mile south-west of Stump Cross, a similar succession is seen in the area of Knowle Hill but only a small part of the outcrop [ST 594 431] comes within the area of the map. The strata in both these areas are gently folded with a general southerly dip of about 20° to 25°. G.W.G.

References

BAMBER, A. E. 1924. The Avonian of the Western Mendips from the Cheddar valley to the sea. Proc. Bristol Nat. Soc. (4), 6, 75–91.

DE LA BECHE, H. T. 1846. On the Formation of the Rocks of South Wales and South Western England. Mem. Geol. Surv.,1.

DIXON, E. E. L. in Strahan, A. 1907. The Geology of the South Wales Coalfield. Part VIII. The Country around Swansea. Mem. Geol. Surv.

DIXON, E. E. L. and VAUGHAN, A. 1911. The Carboniferous Succession in Gower, with Notes on its Fauna and Conditions of Deposition. Quart. J. Geol. Soc., 67, 477–571.

EVENS, E. D. 1959. Origin of Picrite Blocks near Wells. Geol. Mag., 95, 511.

FOURMARIER, P. and Others. 1954. Prodrome d'une description geologique de la Belgique. Liege.

GEORGE, T. N. 1954. Pre-Seminulan Main Limestone of the Avonian Series in Breconshire. Quart. J. Geol. Soc., 110, 283–322.

GEORGE, T. N. 1956. Carboniferous Main Limestone of the East Crop in South Wales. Quart. J. Geol. Soc., 111 for 1955,309–22.

GEORGE, T. N. 1958. Lower Carboniferous palaeogeography of the British Isles. Proc. Yorks. Geol. Soc., 31, pt. 3,227–318.

GREEN, G. W. 1953. in Sum. Prog. Geol. Surv. for 1952,20.

ILLING, L. V. 1954. Bahaman Calcareous Sands. Bull. Amer. Assoc. Petroleum Geol., 38, no. 1,1–95.

KELLAWAY, G. A. 1960. in Sum. Frog. Geol. Surv. for 1959,26.

KELLAWAY, G. A. and WELCH, F. B. A. 1955. The Upper Old Red Sandstone and Lower Carboniferous Rocks of Bristol and the Mendips compared with those of Chepstow and the Forest of Dean. Bull. Geol. Surv. Gt. Brit., No. 9,1–21.

MORGAN, C. LLOYD. 1890. Mendip Notes. Proc. Bristol Nat. Soc. (3), 6, 169–82.

NEWELL, N. D. and RIGBY, J. K. 1957. Geological Studies on the Great Bahaman Bank. Spec. Publ Soc. Econ. Paleont., Miner., 5, 15–79.

REYNOLDS, S. H. 1920. On the Occurrence of Picrite Serpentine at Ebbor near Wells, Somerset. Geol. Mag., 57, 224–6.

REYNOLDS, S. H. 1921. The Lithological Succession of the Avonian at Clifton. Quart. J. Geol. Soc., 77, 213–43.

REYNOLDS, S. H. and VAUGHAN, A. 1911. Faunal and Lithological Sequence in the Carboniferous Limestone Series (Avonian) of Burrington Combe (Somerset). Quart. J. Geol. Soc., 67, 342–92.

RICHARDSON, L. 1928. Wells and Springs of Somerset. Mem. Geol. Surv.

SIBLY, T. F. 1905. The Carboniferous Limestone of Burrington Combe. Proc. Bristol Nat. Soc. (4), 1, 14–41.

SIBLY, T. F. 1906. On the Carboniferous Limestone of the Mendip Area (Somerset) with especial reference to the Palaeontological Sequence. Quart. J. Geol. Soc., 62, 324–80.

VAUGHAN, A. 1905. The Palaeontological Sequence in the Carboniferous Limestone of the Bristol Area. Quart. J. Geol. Soc., 61, 181–305.

SIBLY, T. F. 1906. The Carboniferous Limestone Series (Avonian) of the Avon Gorge. Proc. Bristol Nat. Soc. (4), 1, 74–168.

WALLIS, F. S. 1935. Carboniferous Cephalopods from Shipham, Somerset. Proc. Bristol Nat. Soc. (4), 7, 538–41.

WALLIS, F. S. 1936. Further notes on Carboniferous Mollusca from Shipham. Proc. Bristol Nat. Soc. (4), 8, 217–9.

WELCH, F. B. A. 1929. The Geological Structure of the Central Mendips. Quart. J. Geol. Soc., 85, 45–76.

WELCH, F. B. A. 1932. The Geological Structure of the Blackdown Pericline. Proc. Bristol Nat. Soc. (4), 7, 388–96.

WELCH F. B A. 1933. The Geological Structure of the Eastern Mendips. Quart. J. Geol. Soc., 89, 14–52.

WHITTARD, W. F. 1949. Temporary Exposures and Borehole Records in the Bristol Area: IV. Boreholes on Mendip. Proc. Bristol Nat. Soc., 27, 479–82.

WOODWARD, H. B. 1876. Geology of East Somerset and Bristol Coal-Fields. Mem. Geol. Surv.

Chapter 5 Upper Carboniferous

Millstone Grit Series

BETWEEN the top of the Hotwells Limestone and the Ashton Vale Marine Band at the base of the Coal Measures is a group of mainly quartzitic sandstones with occasional bands of sandy shale, and to these rocks the name Quartzitic Sandstone Group has been applied (Dines 1954, p. 24). South of Bristol, at Winford, on the eastern side of Broadfield Down (Bristol Sheet 264), the presence of Eumorphoceras in the basal bed indicates that the group forms part of the Millstone Grit Series (Kellaway and Welch 1955, p. 18). So far, within the area covered by the Wells (280) Sheet no diagnostic fossils indicative of Millstone Grit age have been found.

The Quartzitic Sandstone Group has limited outcrops on both the north and south sides of the Mendips. The total thickness on the north side is about 150 ft. On the south side the full thickness is only seen west of Ebbor Rocks where it measures rather more than 175 ft. F.B.A.W.

Mr. R. W. Elliot reports on the microscopical nature of the sandstones as follows: Specimens of 'Millstone Grit' in our collections from Emborough (E6624), Ebbor Rocks (E25635), (E25636), near Wookey Hole (E26782), (E26783) and near Dinder (E27565) are in general . quartzitic sandstones, relatively free from impurity and in which secondary enlargement of the quartz grains is a noticeable feature. No feldspar is observed in thin section. Thin sections show that the rocks consist of subangular and angular grains (commonly around 0.3 mm diameter but ranging up to 3 mm) of quartz and granulitized quartz with subsidiary chert, siltstone and mudstone set in a matrix of angular grains (about 0.05 mm diameter) of quartz. Accessory minerals include tourmaline, zircon, leucoxene, and haematite.

Details

North of the Mendips

Compton Martin

Sanders's map (1865, Sheet 14) shows a small outcrop of Millstone Grit, 0.25 mile south-west of Compton Martin church. This may include the pinkish quartzite seen in the top of the nearby Cliff Quarry [ST 542 568], (see p. 27 above). The outcrop, however, is too small for insertion on the one-inch map.

Chewton Mendip

A trial pit (estimated position 0.1667 mile south by west of Chewton Mendip church) is shown on the Geological Survey Horizontal Section No. 104 (1875) as being in Trias resting on Millstone Grit. McMurtrie (1869, pp. 58–9) commenting on this pit states that 'after passing through some 35 fathoms of conglomerate, a boring was put down for 180 ft in a hard stone said to have been the mountain limestone'. McMurtrie (ibid.)considered, however, that the 'hard stone' was Millstone Grit. G.W.G.

Emborough

In Emborough Grove, to the south-west of Emborough church, a long line of overgrown workings show up to 20 ft of massive quartzitic sandstone dipping at about 30° to the south-east, and overlain by about 14 ft of grey-green shales. The latter are mapped as Lower Coal Series, although the Ashton Vale Marine Band has so far not been detected here. In an old quarry [ST 612 510], some 400 yd south-west of Emborough church, may be seen 16 ft of quartzite overlain by 7 ft of Rhaetic shales. At a distance of about 500 yd south-east of Emborough church the Quartzitic Sandstone Group occupies a small ill-exposed area extending eastwards for about 0.25 mile.

Gurney Slade

At a distance of 1 mile east of Binegar church there is a small outcrop of the Quartzitic Sandstone Group in which a quarry [ST 632 505] shows 10 ft of massive quartzite with red-stained shale bands overlain by 25 ft of Dolomitic Conglomerate.

Nettlebridge to Ashwick

From near Badger's Cross, the Millstone Grit occupies a narrow, but nearly continuous, outcrop extending eastwards to Mells (Frome Sheet 281). In a quarry [ST 649 481] in Harridge Wood, 1350 yd E. 18° S. of Ashwick church, about 30 ft of quartzite, locally pebbly, dip at 50° to the N.N.E. On the east side of Harridge Wood poorly exposed crags of quartzite may be seen in the ravine sides. F.B.A.W.

South of the Mendips

Apart from the surface outcrops, the presence of Millstone Grit at shallow depth beneath the Trias is inferred, on structural grounds, in a number of localities (see (Plate 5), and pp. 140–1). The reddish quartzitic sandstone exposed in an inlier, about 1 mile south-east of Cheddar, formerly considered to be of Millstone Grit age (Lloyd Morgan 1890, pp. 178–9; Welch 1929, p. 63) is now thought to belong to the Old Red Sandstone (see p. 12 above).

Dinder

About 0.25 mile south of Dinder church, a small outcrop of the Quartzitic Sandstone Group is indicated by abundant surface debris of quartzitic sandstone, while the presence of interbedded grey shales was proved by augering.

Wookey Hole

Small exposures [ST 539 475], 700 yd east of Wookey Hole church, show quartzitic sandstone dipping at 42° to the south-south-west. The structural relations of this outcrop are illustrated in (Figure 12). G.W.G.

Ebbor Rocks

The most extensive outcrop of the Quartzitic Sandstone Group south of the Mendips occurs in the area west of Ebbor Rocks (Figure 5). Quartzitic sandstone crags occur in the ravine (Ebbor Gorge) [ST 524 485] 1300 yd south-west of Higher Pitts Farm. From here a narrow band of quartzite extends north-westwards and this has been quarried at several places, the chief exposures being a quarry [ST 517 490] 2000 yd E. 14° N. of Westbury church and a line of quarries [ST 519 490] 2100 yd E. 9° N. of the church. In both exposures the quartzite is seen steeply dipping at angles of up to 70° in a south-westerly direction. A generalized section of the strata in this area is as follows:

At a locality [ST 5188 4905] 2110 yd E. 10° N. of Westbury church the grey shale of the above section yielded fossils, identified by Dr. W. H. C. Ramsbottom as Lingula mytilloides J. Sowerby and Bucanopsis sp.F.B.A.W.

Coal Measures

It is probable that within the area of the Wells (280) Sheet Coal Measures occupy an area of some 68 square miles on the north side of the Mendip Hills, but of this amount only 4.5 square miles are exposed at the surface, the remainder being concealed by later formations. On the southern flank of the Mendip Hills a small area of Coal Measures is exposed near Ebbor Rocks, north-west of Wells. The possibility of further areas of Coal Measures preserved beneath the Mesozoic rocks to the south of the Mendips is discussed elsewhere (see pp. 134, 155, also footnote on p. 12). East of a line drawn from Bishop Sutton southwards to Chilcompton and Nettlebridge the Coal Measures occupy the western limbs of the Pensford and Radstock basins where extensive mining has taken place, and here the succession and structure of the measures is tolerably well known. West of this line and north of the Mendips however the structure and stratigraphy of the concealed western extension of the coalfield are not known in detail. A few borings have been made in the Triassic-filled depression north of Black Down, notably at Banwell and Langford, but the complete succession in the Coal Measures has not been proved.

The largest structural disturbance affecting the coalfield is the great belt of faulting and folding known as the Farmborough Fault Belt which separates the Pensford Basin in the north from the Radstock Basin in the south. The presence of this structure was established by mining but it is believed to continue westwards into the Yeo valley north of the Mendips, being flanked on the north by the up-folded mass of Carboniferous Limestone and Millstone Grit composing Broadfield Down (Bristol Sheet 264). The southern end of the Radstock Basin is marked by a large thrust-faulted overfold in which the main thrust, known as the Southern Overthrust, bounds the workings in the Upper Coal Series. Smaller, but nonetheless important thrust faults, such as the Radstock Slide, also affect the Upper Coal Series of the Radstock Basin.

The development of coal mining in the southern part of the Somerset Coalfield has followed much the same general pattern as in other coalfields, and it was natural for the earliest workings to be situated in areas where the seams crop out at the surface, as between Clutton and High Littleton, and in the valley at Nettlebridge in the south. The earliest record of coal working goes back as far as the year 1305 (Hylton 1910), while Saxton's map of Somerset (1575) shows 'The Cole Pite' a little to the east of Nettlebridge. Mining in the concealed part of the coalfield was well established by 1800 as can be seen from the accounts by Strachey (1719) and Billingsley (1797). At the present time Old Mills and New Rock are the only two active collieries within the area of Sheet 280^‡17 .

Classification

The non-recognition of any marine bands other than the Ashton Vale Marine Band has made it impossible to subdivide the Coal Measures of Somerset in accordance with the latest official classification based on marine bands (Stubblefield and Trotter 1957). Under these circumstances it has been necessary to retain the old tripartite division of the Coal Measures into Lower Coal Series, Pennant Grit (or Series) and Upper Coal Series. This is based essentially on the classification of De la Beche (1846) and is the one used in practice by the mining industry. The sub-divisions now recognized by the Geological Survey in the Radstock Basin may be summarized as follows:

The Lower Coal Series comprises Lower, Middle and Upper Coal Measures; the Pennant Series and Upper Coal Series are Upper Coal Measures.

A brief mention must here be made regarding the position of the measures in the Pensford Basin in relation to the classification above. North of the Farmborough Fault Belt there are two groups of productive measures, separated by a belt of unproductive measures containing 'red measures', which have been worked from Bromley and Pensford collieries and are known as the Bromley and Pensford groups. It has usually been considered by mining engineers that the former is the equivalent of the Farrington Group and the latter equates with the Radstock Group, and this correlation has been adopted in the present survey (see (Figure 6)). Differing views have been expressed by Steart (1911) and by Moore and Trueman (1937). To the south-west of the Pensford Basin is the Bishop Sutton area in which coals have been worked for a very long time, one of the earliest descriptions being that of Strachey (1719). Here the 3 main seams resemble those of the Bromley succession with which a tentative correlation is suggested.

Since the succession of the Coal Measures will be fully described in a forthcoming memoir on the Bristol and Somerset coalfields only a short account will be given in the following pages.

Lower Coal Series

The Lower Coal Series comprises some 1400 ft of measures, largely shales and mudstones in which nodular ironstone is locally developed, and with sandstones developed sparsely near the base. It is thus a relatively soft and incompetent formation occupying a position between the competent Carboniferous Limestone–Millstone Grit block below and the massive Pennant Sandstone above, and is therefore liable to intense deformation when subjected to earth-movements and severe faulting. The Lower Coal Series contains a number of relatively thick coal seams, the more important of which are shown in section (Figure 6); these, when traced eastwards into the adjoining Frome (281) Sheet, are seen to improve in thickness and quality.

Except around Emborough church, in the valley at Nettlebridge, and near Ebbor Rocks, the outcrop of the Lower Coal Series is concealed. Its conjectured sub-Triassic outcrop, (Plate 5), strikes in a south-easterly direction from the Farmborough Fault Belt near Blagdon, through West Harptree to Nettlebridge. Only in its southern portion is the succession known, mainly from records of working at Moorewood Colliery [ST 642 495] 1000 yd north-west of Nettlebridge.

The base of the Lower Coal Series is defined by the Ashton Vale Marine Band characterized by Gastrioceras subcrenatum C. Schmidt. This band which occurs almost immediately above the quartzitic sandstone of the Millstone Grit has been recorded from a stream section [ST 653 480] 1 mile E. 18° S. of Ashwick church, and, on the south side of the Mendips adjacent to the Ebbor Thrust, at a locality [ST 519 490] 2070 yd E. 18° N. of Westbury church (see (Figure 5)) where fossils identified by Dr. C. J. Stubblefield include Anthracoceras sp.and Gastrioceras cf. subcrenatum. The lowest 400–450 ft of the Lower Coal Series forms a relatively barren group of measures, only two seams, the Red Axen and White Axen, having been worked to any extent. The Red Axen (2 ft) was probably won from small adits in Emborough Grove. The White Axen (2 ft 6 in), which is overlain by a water-bearing sandstone, appears to have been mined from the old pits immediately east of Emborough church, and from shallow pits in the valley south-west of Blacker's Hill. In the Ebbor Rocks region a small amount of surface working from a shaft 120 ft deep took place about 1835, but the seam here was said to be only 6 to 12 in thick (Greenwell, 1892). At the top of this barren group is the Perrink or Blackstone Vein (3 ft 6 in), the most constant seam in the Lower Coal Series of the area and one which has been extensively worked. A seam believed to be the Perrink was proved in the Ston Easton No. 1 and No. 2 boreholes [ST 6225 5174] and [ST 6211 5158] where it was highly disturbed; in the overlying measures Carbonicola of the pseudorobusta Trueman type are abundant. Separated from the Perrink by some 80 ft of mudstone is the Main Coal or Callows, a soft coal 4 to 6 ft in thickness, with a bad roof. These two seams were the chief coals worked at Moorewood Colliery [ST 642 495] and were used mainly for coking. Two seams, the Stone Rag (4 to 6 ft) and the Fern Rag (2 ft 6 in to 3 ft) lying about 60 and 120 ft respectively above the Main Coal were also worked to a limited degree at Moorewood. Above the Fern Rag there follows some 670 ft of measures in which no workable coals are developed within the Moore-wood area, though farther east, around Vobster (Frome Sheet 281) seams known as the Standing Coal and Dungy Drift have proved workable.

Pennant Series

The Pennant Series comprises some 3600 ft of mainly coarse sandstone or sub-greywacke with subordinate developments of sandy shale and mudstone, the largest of which, about 330 ft thick, overlies the seam known as the Garden Course. The base of the Pennant Series is drawn in the position at which dominantly argillaceous rocks are replaced by arenaceous strata. In the present region this boundary coincides with a thin coal seam known as the Little Course. This definition involves the inclusion within the Pennant Series of about 600 ft of mainly arenaceous measures with coal seams (part of the former New Rock Group) which were previously regarded as belonging to the Lower Coal Series (for instance, Prestwich 1871, Moore and Trueman, 1937). The upper limit of the Pennant series is now drawn at the base of the Rudge Vein. Within the district comprising Sheet 280 the Pennant Series is only exposed at four localities. Of these the most northerly are the two inliers east of Chew Stoke which lie at the western end of the Bromley Horst (p. 145). Here the exposed upper part of the Pennant Series consists of about 850 ft of sandstone and sub-greywacke with subordinate bands of shale and sandy mudstone. These rocks dip eastwards at about 18° towards the centre of the Pensford Basin. About 3 miles south-east of the Bromley Horst, within the angle formed by the Clutton and Timsbury faults, lies the largest exposed area of Pennant Series. On the south side of this sandstone tract the prominent south-facing scarp of the Timsbury Fault gives rise to an impressive and much quarried sandstone feature south of Temple Cloud. The only other exposure of the Pennant Series in this part of the coalfield is a small outcrop of basal Pennant sandstone emerging from beneath the Dolomitic Conglomerate about 0.5 mile north-west of Nettlebridge. Little is known about the distribution of the Pennant Series within the concealed part of the coalfield. Geophysical work (Bullerwell 1954) suggests that the sub-Triassic outcrop forms a belt extending in a north-westerly direction from Downside to Bishop Sutton where it is cut by the Farmborough Fault Belt, on the north side of which a very large westerly displacement of the outcrop is envisaged. In the extreme western part of Sheet 280 some 400 ft of sandstone, described as Pennant, were proved in a water bore [ST 3995 6087] 1900 yd north of Banwell church (p. 198).

Three important coal seams occur in the lowest third of the Pennant Series (sometimes known as the New Rock Group), but the overlying rocks contain no coal of any importance.

The lowest of the worked coals in the Pennant Series is the Great Course (4 ft) which lies about 90 ft above the base of the Pennant Series. It has been extensively mined at Nettlebridge Pit [ST 655 494] and is now worked at New Rock Colliery [ST 648506]. In the former locality the coal was in three leaves separated by thin dirt partings; traced westwards the partings thicken at the expense of the coal. Separated by some 460 ft of measures, predominantly sandstone, from the Great Course is the Garden Course, probably the most constant seam in the Pennant Series. At New Rock Colliery the seam section comprises: top (dirty) coal 1 to 2 ft, unbedded mudstone 1 to 2.5 ft, bottom coal 2 ft. This seam is followed by some 300 ft of measures, mainly sandy mudstone, in which two rather inferior seams, the Two Coal and the Small Coal have been worked to a limited extent at Nettlebridge Pit and New Rock Colliery. This sandy shale division is succeeded by a further 200 ft of sandstone at the top of which lies the Globe Vein, about 3 ft thick, reputed to be one of the best household coals in the district.

Little is known about the 2400 ft of measures which overlie the Globe Vein, but they appear to consist mainly of sandstone with thin bands of sandy shale and a few thin unworkable coal seams, some of which have been proved around Temple Cloud. Here are exposed some of the uppermost beds of the formation consisting of massive current-bedded sandstones which have been quarried for building to the east of Temple Cloud. Above these sandstones a group of shales, some 200 ft thick gives rise to the belt of low ground west of Highbury Hill. In the top part of these shales, which are overlain by the Rudge Vein, 'red measures' are developed.

Upper Coal Series

Farrington (and Bromley) Group

The Farrington Group varies in thickness from about 1100 ft around Farrington Gurney in the south to 800 ft at Bromley in the north. Massive Pennant-like sandstones are developed in the basal 300 ft, especially in the northern part of the district, whilst a certain amount of sandstone occurs above the No. 5 Vein, but otherwise the succession is largely composed of mudstone and sandy mudstone. The base of the Farrington Group is now taken by the Geological Survey at the Rudge Vein. The identification of this seam and its inclusion in the Farrington Group are supported by floral and faunal evidence which will be given in detail in a forthcoming memoir dealing with the Bristol and Somerset Coalfields. The Rudge Vein lies about 400 ft lower in the succession than the No. 9 of Radstock which was adopted by Moore and Trueman (1937, p. 228) as the base of the Farrington Group.

In contrast with the seams of the Lower Coal Series and the Pennant Series those of the Farrington Group are relatively thin. Only at Old Mills Colliery [ST 652 550] are the coals of this group now worked within the district of Sheet 280. The formation is entirely concealed except for a narrow sinuous outcrop extending from Farrington Gurney, through Hallatrow and Clutton to North End, and a small tract at Bromley, east of Chew Stoke.

The Rudge Vein (16 to 17 in) is a clean coal which has been worked in the past at Greyfield Colliery [ST 640 587] 1000 yd north-west of High Littleton church, at Clutton Ham or Rudge Pit [ST 629 585] 800 yd south-east of Clutton church, and in recent years at Marsh Lane Colliery [ST 632 552] Farrington Gurney. To the east, however, at Old Mills Colliery, the seam proved to be only 14 in and could not be worked. Between Clutton in the south and Bromley in the north, the Rudge Vein is concealed by Mesozoic rocks, but in the Bromley Horst a thin, dirty, and somewhat impersistent seam occurs at the top of the Pennant Series. This is thought to be the equivalent of the Rudge Vein. It is overlain by a fairly massive sandstone or sub-greywacke similar to that at the base of the Upper Coal Series at Highbury Hill near Clutton.

Above the Rudge Vein, typically seen in the western limb of the Radstock Basin, lie some 400 ft of arenaceous strata with a series of fireclays and thin impersistent coal seams developed locally in the middle. These beds unite to form a thick, rather dirty coal seam, part of which is worked at Old Mills Colliery as the No. 11 Seam. About 200 ft above this coal lies the No. 10 Seam, which is the most important coal in the Upper Coal Series of the Radstock Basin although the thickest development lies beyond the eastern margin of the Wells Sheet. When traced north-westwards from its area of maximum development this coal splits and in the Old Mills–Farrington area, on the Wells Sheet, consists of three distinct coals, as follows: coal (Night Vein) 18 in, measures 20 ft, coal (Dirty Duck) 16 in, measures 4.75 ft, bottom coal (Brights, or 21-inch Vein) 21 in. Above the No. 10, and separated from it by about 200 ft of measures, is the No. 7, New or Church Close Vein which crops out between Farrington church and Hallatrow. This is a soft brittle coal, about 33 in thick, which has been extensively worked at Old Mills, Farrington and Greyfield Collieries. The most constant seam of the Farrington Group however is the No. 5 Seam (or Middle Vein) which has an average thickness of 26 in and a good roof. It has been extensively worked at Old Mills and Farrington collieries, south of the Winterfield Fault, and from shallow pits northwards to near Paulton. In Farrington Colliery the seam was said to develop a median 4-in band of stone, which may indicate the beginning of a northward split in the seam, for at Greyfield Colliery two seams known as the Peacock and Dabchick occur in the position of No. 5 Seam, and have been worked throughout the Clutton area. Above No. 5 Seam there occur a number of relatively thin seams including the Top Vein or Streak but none of these are recognizable over any great distance.

The upper limit of the Farrington Series is usually taken at the Rock Vein of Radstock (Moore and Trueman 1937, p. 228). A seam known as the Rock Vein was also worked at Clutton but it is doubtful whether the two Rock Veins are the same coal seam. F.B.A.W.

North of Clutton the Farrington Series is concealed by Mesozoic rocks. Some doubt has been expressed as to the correlation of the Bromley seams of the Bromley Colliery (Moore and Trueman 1937, pp. 233–7). However, it is thought that the three principal seams worked in the Bromley area, Bromley No. 4, No. 5 and No. 6 equate with a part of the Farrington Group and the balance of the evidence favours the tentative correlation of Bromley No. 4 Seam with the No. 5 Seam of Farrington. None of the Bromley seams are thick. Bromley No. 4 the most persistent and widely worked of these coals consists of 20 to 24 in of clean coal with a relatively low ash content and high thermal value. The higher Bromley Seams are very erratic in their development and are succeeded by barren sandstone, fireclays and red measures similar to those separating the Farrington and Radstock groups at Clutton.

Barren Red Group

The Barren Red Group separates the Farrington and Radstock groups and is distinguished from them by the absence of coal seams and by the development of thick beds of sandstone, fireclay and red measures. The maximum thickness of the Barren Red Group is at least 750 ft. Of this, the red measures account for about 250 ft at Farrington, but are thinner in the north, being about 45 ft thick near the Bromley Horst. Although the Barren Red Group is not distinguished by colour on the one-inch map it has been recognized on the surface at a number of places. Thus it crops out between the Farrington and Radstock groups in the area between High Littleton and North End, and was formerly exposed in the railway cutting north of Clutton railway station where it is overlain by the Bull Vein at the base of the Radstock Group. North of this place the Coal Measures are concealed, but in the Bromley Horst red and mottled mudstones and fireclays crop out between the Bromley and Pensford groups of seams. When in a weathered condition the Barren Red Group consists of very weak clays which tend to flow or slip, even on gentle slopes, and its crop is marked by a small area of landslip lying about 0.125 mile east of Bromley Pit shaft. G.A.K., F.B.A.W.

Radstock (and Pensford) Group

The Radstock Group includes some 800 ft of measures between the Nine Inch Vein^‡18  and the Withy Mills Vein and forms the uppermost division of the productive measures. Although the seams are of no great thickness they include six easily worked, strongly caking, orthobituminous coals more or less evenly spaced through 300 feet of measures. These seams have been extensively worked in the past throughout the Radstock Basin, but reserves are now virtually exhausted and all extraction of the coal has ceased. The Radstock Group occurs in a small area in the east and north-east of Sheet 280, and here a considerable deterioration of the coal seams can be noticed when they are compared with their corresponding development at Radstock. Not only have some seams thinned, but splitting and development of dirt partings are very noticeable. The Bull Vein is here too thin to be worked, and the Bottom Little Vein (14 to 16 in) only capable of extraction in the High Littleton area. The Slyving and Middle veins appear to be the most constant, and have been worked on a considerable scale, though even in these coals attenuation and splitting is noticeable when traced northwards towards the Farmborough Fault Belt. The only seam which appears to remain unchanged as a single bed of coal is the Great Vein (24 to 26 in). Correlation between the Radstock and Pensford groups is not clear, but it is possible that the Slyving and Middle veins may be represented by two dirty coals, Pensford No. 3 and No. 2. F. B.A.W.

Publow Group

Since by definition the top of the Radstock Group is taken at the Withy Mills Vein, some 450 ft above the Great Vein, the name Publow Group (Kellaway and Welch 1955) is applied to all Coal Measures lying above this seam in the Radstock Basin, and its assumed equivalent in the Pensford area. In this group the rocks include plant and shell-bearing mudstones, bands of sandstone or grit and some thin dirty coal seams. Within the area under discussion most of the Publow Group is confined to a region north of the Farmborough Fault Belt. A sinking for coal at Farmborough [ST 657 602] 0.5 mile east of Hobb's Wall, (Anstie 1873, p. 92) probably penetrated Publow measures to a depth of 1320 ft. This shaft proved 15 coal seams of which only one attained a thickness of 13 in.

Only a relatively small area is occupied by the outcrop of the Publow Group. Highly disturbed measures lying within the Farmborough Fault Belt are seen in the floor of the valley east of Red Hill and extend northwards towards Chelwood House and the eastern part of the Bromley Horst. For the most part the rocks are very poorly exposed, but appear to consist of mudstone and fireclay associated with some sandstone. The latter contains much coaly debris including beds of coal pebbles. A few thin impersistent coal seams also occur and attempts have been made to work these in the vicinity of Chelwood and Hunstrete though all the individual workings appear to have been of limited extent. G.A.K.

References

ANSTIE, J. 1873. The Coal Fields of Gloucestershire and Somersetshire, and their Resources. London.

BILLINGSLEY, J. 1797. General view of the Agriculture of the County of Somerset with observations on the means of its Improvement. Bath.

BULLERWELL, W. 1954. A gravimeter survey of the Ston Easton–Harptree district, East Somerset. Bull. Geol. Surv. Gt. Brit., No. 6, 36–56.

DE LA BECHE, H. T. 1846. On the Formation of the Rocks of South Wales and South Western England. Mem. Geol. Surv., 1.

DINES, H. G. 1954. in Sum. Prog. Geol. Surv. for 1953, 24.

GREENWELL, G. C. 1892. Borehole for coal at Easton. Trans. Manch. Geol. Soc., 21, 596–604.

HYLTON, LORD. 1910. Notes on the History of the Parish of Kilmersdon. Taunton.

KELLAWAY, G. A. and WELCH, F. B. A. 1955. Upper Old Red Sandstone and Carboniferous, pp. 9–23 of Bristol and its Adjoining Counties, edited by C. M. MacInnes and W. F. Whittard. Bristol.

MCMURTRIE, J. 1869. A lecture on the Carboniferous Strata of Somersetshire. Proc. Bath Nat. Hist. and Antiq. Field Club, 1 (No. 2 for 1868), 45–60.

MOORE, L. R. and TRUEMAN, A. E. 1937. The Coal Measures of Bristol and Somerset. Quart. J. Geol. Soc., 93, 195–240.

MORGAN, C. LLOYD. 1890. Mendip Notes. Proc. Bristol Nat. Soc. (3), 6,169–82.

PRESTWICH, J. 1871. in Report of the Royal Coal Commission, 1, London.

SANDERS, W. 1865. Map of the Bristol Coal Fields, sheet no. 14. Bristol.

STEART, F. A. 1911. The North-western portion of the Somersetshire Coalfield and the Farmborough Fault. The Colliery Guardian, 102, 916–8.

STRACHEY, J. 1719. A curious Description of the Strata observed in the Coal-Mines of Mendip in Somersetshire. Phil. Trans. Roy. Soc., 30, 968–73.

STUBBLEFIELD, C. J. and TROTTER, F. M. 1957. Divisions of the Coal Measures on Geological Survey maps of England and Wales. Bull. Geol. Surv. Gt. Brit., No. 13, 1–5.

WELCH, F. B. A. 1929. The Geological Structure of the Central Mendips. Quart. J. Geol. Soc., 85, 45–76.

Chapter 6 Triassic

The oldest Mesozoic rocks proved in the area of the Wells Sheet are of Keuper age, and rest unconformably on an uneven surface composed of folded and eroded Palaeozoic rocks. No Permian or Bunter rocks have yet been recognized though it is possible that some may be present at depth in the south-western corner of the area which lies within 3.5 miles of the site of a borehole at Puriton [ST 320 408], where sandstones and marls below the Keuper were proved to a depth of 700 ft (Ussher 1911; McMurtrie 1912).

The principal local subdivisions of the Triassic are as follows:

Previous work

Whilst much has been written on the Keuper deposits of South-West England the amount of detailed information referring specifically to the district under discussion is small (see specially, Woodward 1876; Richardson 1928). In recent years Upper Triassic vertebrate faunas have been described from fissure deposits at several localities (Kane 1949 and 1956; Robinson 1957). Bristow (1864) introduced the term Penarth Beds to cover all the deposits between the top of the Keuper Marl and the base of the Blue Lias. The term however, fairly soon fell out of use, probably because it grouped together too many distinct formations. Much of our knowledge of the Rhaetic of the area is due to the detailed work of Richardson (1911), though previous workers, including Boyd Dawkins (1864) and Charles Moore (1867) have published accounts of various sections.

Keuper

The Keuper rocks are the most widespread group of strata present in the district and extend over some two-thirds of the Sheet area. The total thickness of the Keuper ranges between about 150 and 400 ft over much of the ground north of the Mendips but rises to over 1,500 ft south of the Mendips. The deposits consist mainly of red marls with occasional bands of greenish-grey sandstone and marl.

Adjacent to the Mendips the marls pass laterally into conglomerates and breccias (Dolomitic Conglomerate) which may attain over 300 ft in thickness.

The existing relief of the sub-Triassic surface (Plate 3) is an expression of pre-Triassic topography modified by subsequent earth-movements. The most prominent features of this surface are the Central Somerset Basin, a marked depression in the south-western and the north-western parts of the Sheet and the striking unevenness of the topography in the Mendips. The formation of the Central Somerset Basin took place in Triassic and later periods; the very large degree of post-Triassic movement is shown by the present-day inclination of the base of the Rhaetic (cp. (Plate 3) and (Figure 14)). The flanks of the Mendips, on the other hand, were clearly defined in Triassic times and were dissected by a number of steep-sided valleys, or gorges, radiating outwards from the central high ground. The degree of dissection of the Palaeozoic rocks is most marked in the western and southern parts of the Mendips. Every gradation is present between the relatively undissected central plateau area and a number of small steep-sided isolated hills of Carboniferous Limestone such as Nyland Hill, south of Cheddar. Unlike the Carboniferous Limestone, the Coal Measures normally gave rise to a relatively subdued relief. The pre-Triassic topography may bear a close relationship to the structure in the Palaeozoic rocks. Thus major Triassic valleys, now represented by thick Keuper deposits, are commonly aligned along anticlinal and synclinal fold axes and faults present in the underlying Palaeozoic rocks. Examples of the former are provided by the Triassic valleys occupying the cores of the Blackdown, North Hill and Pen Hill periclines; Armorican synclines associated with Triassic valleys include the Binegar, Stoke Woods and Cheddar synclines; Armorican faults along which Triassic erosion was active are numerous and include the Biddle, Stock Hill and Ebbor faults.

The Keuper rocks are of a continental facies probably laid down in arid or semi-arid conditions. The Dolomitic Conglomerate is composed of coarse debris derived from the adjacent Palaeozoic rocks whilst the Keuper Marl represents the finer grades much of which accumulated farther away in large, but possibly temporary, lakes or inland seas. The extensive dolomitization of the Dolomitic Conglomerate both at the western end and to the south of the Mendips may be due to the action of magnesium salts derived from these waters. The grey and green bands in the Keuper culminating in the Tea Green Marl at the top of the succession are thought to indicate more humid climatic periods.

Dolomitic Conglomerate

Typically, the Dolomitic Conglomerate is a breccia or conglomerate largely composed of fragments of Carboniferous Limestone cemented in a matrix of sandy marl or fine-grained limestone debris. The rock fragments are rounded, subangular, or angular and vary in size from that of a walnut, or even smaller, to large boulders measuring several feet across (Plate 4A). Locally debris derived from the Old Red Sandstone and Quartzitic Sandstone Group is present. The Dolomitic Conglomerate—like the Carboniferous Limestone—forms bold crags, may be eroded into gorges, supports an underground drainage system including caves and swallets and is the locus for an extensive lead-zinc mineralization (Green 1958). The formation represents Triassic scree and outwash fans adjacent to the ancient hills of Palaeozoic rocks, and it fills 'fossil' wadis or gorges cut into these hills. Recent erosion has partially re-excavated some of these gorges, as at Burrington Combe (see p. 70).

The Dolomitic Conglomerate has in many cases undergone considerable secondary changes, in particular silicification, haematitization and dolomitization.

Silicification is discussed elsewhere (p. 95). Haematitization is the conversion of the conglomerate into an 'earthy' iron ore known as red ochre. The presence of all gradations from unaltered conglomerate to impure haematite rock show that the formation of the latter is the result of metasomatic replacement of calcium carbonate by haematite (see also p. 165).

The most widespread form of alteration is dolomitization, hence the name 'Dolomitic Conglomerate'. All stages of dolomitization are present, both matrix and pebbles showing various degrees of alteration. Examination of thin sections by Mr. R. W. Elliot showed that the dolomite occurs as rhombs, plates and grains (ω =1.680) which vary between the limits of 0.017 and 0.2 mm diameter. Dolomitization is usually accompanied by hydration or partial hydration of the original disseminated haematite to limonite so that macroscopically the colour has changed from red to yellow and yellow-brown. Widespread dolomitization of the Dolomitic Conglomerate occurs in the Shipham–Rowberrow area and can be traced along the southern flanks of the Mendips from the western margin of the Wells Sheet to the Wells area. Dolomitization affects more than one level in the Dolomitic Conglomerate though it is most marked in the upper parts of the succession.

Keuper Marl, Tea Green Marl

The marls of Keuper age are divided into Keuper Marl (red marls) below and Tea Green Marl above. The former consist of red silty mudstones and siltstones with minor bands of sandstone and grey or green silty mudstones. The rocks may be calcareous or dolomitic. The Tea Green Marl consists of dominantly green and grey silty mudstones^‡19 . In parts of the area south of the Mendips the upper part of the beds shown on the map as Tea Green Marl is distinguished by the presence of thin bands of dark grey and almost black silty mudstone reminiscent of the black shales of the overlying Westbury Beds (Rhaetic).

The classification and nomenclature of the beds between the top of the Keuper Marl and the base of the Rhaetic have been subject to much variation. De la Beche (1839, p. 223) grouped them together as the Variegated Marls and put them at the top of the 'New Red Sandstone Series'. Most later authors have also grouped these beds together, but have given them various names descriptive of their colour and classified them as Rhaetic (Dawkins 1864) or Penarth Beds (Bristow 1864, Bristow, Etheridge and Woodward 1873, Woodward 1876) or Keuper (Moore 1867). Richardson (1905, 1911) separated the Tea Green Marl below from the beds, with dark grey mudstone bands, above which he named the Grey Marl. He regarded the latter as passage beds between the Keuper and the Rhaetic in West Somerset and the Glamorgan coastal belt, though he did not define their lower limit on any published section. In the Uphill cutting section (on Weston super Mare Sheet 279) Kellaway and Oakley (1935) were, however, able to draw the base of the Grey Marl at a well-defined breccia bed. Richardson (1905) further subdivided an upper fossiliferous part of the Grey Marl as Lower Rhaetic and named it 'Sully Beds'. In the area under review beds corresponding to the Grey Marl have only been recognized in the 'Isle' of Wedmore and adjacent areas and they appear to pass downwards without break into the Tea Green Marl with which they have, therefore, been grouped.

General stratigraphy

North of the Mendips

The general sequence of the Keuper rocks in the Chew Stoke area^‡20  worked out by Mr. G. A. Kellaway is as follows:

It is inferred that this general sequence is continued into the area to the west though the beds below the level of the Chillyhill Rock Bed are not exposed at the surface. The sandstone horizons become increasingly manly when traced westwards. Apart from local irregularities due to the form of the pre-Triassic surface, thickening appears to affect all members of the succession and the formation reaches a maximum of about 700 ft north of Banwell (Figure 7). By contrast, when traced eastwards and south-eastwards from Chew Stoke the Keuper diminishes in thickness to some 150 to 250 ft, sandstone horizons cease to be individually recognizable, and the lower half of the succession passes into a series of sandy marls and manly sandstones; this increased sandiness of the succession is probably due to the proximity of Pennant Sandstone outcrops. In the area of the Bromley Horst (p. 74) and around Pensford, the diminution in thickness is at least in part due to overlap and thinning out of the basal beds.

Celestine occurs sparingly as nodules mostly confined to the Woodford Hill Sandstone horizon in the area west of the River Chew. Gypsum in veins and thin bands has been noted in the lower part of the Keuper Marl at Banwell.

The Tea Green Marl in the area north of the Mendips varies in thickness from 10 to 12 ft whilst in the most westerly outcrops, at Butcombe and Stonebridge, a figure of 25 ft is reached. There is everywhere a rapid southward thinning close to the Mendips where the beds are locally overstepped by the Rhaetic, (Plate 3).

As fossils are very rare in the Keuper the discovery of fish remains and 'Modiola minima' (Valpy in Woodward 1876, p. 80) in what appears to be the lateral equivalent of the Butcombe Sandstone near Blagdon is of interest (see p. 71).

South of the Mendips

The Keuper is here much thicker than on the north side of the Mendips. Evidence suggests a maximum thickness of over 1500 ft and possibly as much as 2000 ft for the Keuper.

In the Wells–Westbury–Coxley area, a distinct horizon, named for convenience of description in this account the Westclose Hill Conglomerate (but not separately indicated on the one-inch map), consisting of grey, green, yellow and reddish conglomerate, sandstone and marl, totalling about 10 to 30 ft in thickness, lies at about 100 ft to 130 ft below the base of the Tea Green Marl (Figure 7). The beds are extensively dolomitized locally whilst occurrences of celestine and gypsum have been recorded from immediately below the Westclose Hill Conglomerate at Easton. When traced to the north and north-east this band merges into the thick Dolomitic Conglomerate sequence on the south flank of the Mendips and when it is traced in the opposite direction it becomes fine-grained and passes into grey marl. A lower band of yellow much dolomitized conglomerate is known from boreholes at Wells and in the Cheddar–Draycott area where it apparently passes laterally into greenish grey marl.

It will be noted (Figure 7) that over wide areas both north and south of the Mendips there are two main horizons within the Keuper Marl distinguished by the presence of calcareous sandstones, conglomerates or breccias which pass laterally into grey, green and variegated marls or, in places adjacent to the Mendips, may become strongly dolomitized. The uppermost of these horizons, which is also the most strongly marked, is represented by the Butcombe Sandstone north of the Mendips and the Westclose Hill Conglomerate south of the Mendips. The close similarity between their lithology and their position in the Keuper Succession suggests that the beds correlate with one another. It is possible that they have resulted from distinct climatic episodes in Keuper times (see p. 64).

The thickness changes in the Tea Green Marl are summarized in (Figure 2). The thickest development at outcrop is in the area south of Wedmore where the following succession may be recognized:

The upper two subdivisions probably constitute the Grey Marl (see above, p. 65).

Rhaetic

The Rhaetic rocks are usually divided into a lower group consisting predominantly of black or grey shale and known as the Westbury Beds (Wright 1860, emend. Richardson 1911) and an upper group of pale marls and thin limestones known as the Cotham Beds (Richardson, 1911). For purposes of mapping, however, they are treated as one unit. The pattern of thickness variations of the Rhaetic beds is similar to that of the Tea Green Marl (Figure 2). Thus the thickest development (up to 35 ft) occurs in the main area of the Central Somerset Basin, south of the Mendips with a less marked development (about 20 ft) in the Banwell area, north of the Mendips. This contrasts with the relatively thin Rhaetic succession (10 to 15 ft) present over the remaining areas north and south of the Mendips. The Rhaetic deposits overlap or overstep the Tea Green Marl onto the Palaeozoic rocks and are themselves overlapped and overstepped by the Lower Lias (Figure 14). The presence of Rhaetic fossils in fissure deposits in areas where the Rhaetic has been overstepped give some indication of the former extent of the Rhaetic deposits and show that the line of Lower Liassic overstep is an erosional feature and not a fossil shoreline. The local occurrence of Rhaetic land vertebrates and of conglomerates consisting of Carboniferous Limestone pebbles shows, however, that part of the Mendips formed an island, or islands, in Rhaetic times.

Westbury Beds

These consist mainly of black or dark grey pyritic shales, typically non-calcareous, though sometimes including layers of fibrous calcite known as 'beef'. The fauna, though often abundant in individuals, is poor in number of species. Typical species include Rhaetavicula [Pteria] contorta (Portlock), Schizodus' ewaldi (Bornemann), Chlamys valoniensis (Defrance), Pteromya crowcombeia Moore, Protocardia rhaetica (Merian), Mytilus cloacina Tutcher, Pleurophorus elongatus Moore, small gastropods and fish remains. The well-known bone-beds, which occur at or near the base, are conglomeratic sandy limestones or calcareous sandstones, usually only a few inches thick, and commonly packed with the scales, teeth, bones and spines of fish, together with the bones and teeth of aquatic saurians. Over much of the district there is a single conspicuous basal bone-bed, but as the Rhaetic thickens towards the middle of the Central Somerset Basin as many as four or five, separated by black shales, may be developed.

The base of the Westbury Beds is sharply defined even where the basal bone-bed is missing although the underlying Tea Green Marl includes beds of dark grey or black shale, as for instance, in the Wedmore area. The Keuper black shales can be distinguished from the relatively more plastic Rhaetic black shales by their harder nature and powdery texture. In the Wedmore area the lowest third of the Westbury Beds includes the Wedmore Stone, (Dawkins 1864) a hard limestone composed of shell fragments, which varies from 2.5 to 4.5 ft in thickness and which was formerly much used as a local building stone. This horizon, farther west, passes laterally into sand and sandrock. Impersistent dark grey argillaceous limestones, rarely more than a few inches in thickness occur at several levels in the Westbury Beds.

Cotham Beds

These consist of pale grey and greenish-grey calcareous mudstones with occasional thin calcite mudstones of which the thickest (1.5 to 9 in) lies at, or within a few inches of, the top and forms the well-known 'Cotham Marble'. The latter has a roughly mammilated top surface and a finely laminated basal part with an intermediate part containing curious, commonly arborescent, patterns which account for its popular name of 'Landscape Marble'. This patterned effect is now attributed to calcareous algae (Hamilton 1961). Though not continuous, this band has a wide lateral extension, more especially north of the Mendips. The Cotham Beds are otherwise poorly fossiliferous and were probably laid down in a lagoon or landlocked sea. Strahan (in Richardson 1911, p. 73) long ago suggested that these beds represent a temporary reappearance of late Keuper conditions of deposition. G.W.G.

Details

Keuper North of the Mendips

Banwell–Churchill–Wrington

The Banwell Moor Borehole [ST 3995 6087], 1900 yd N. 2° E. of Banwell church, estimated to start at about 75 to 100 ft below the base of the Rhaetic proved marls to a depth of 511 ft, underlain by hard red and grey sandy marls with thin sandstones to 570 ft, and finally conglomeratic rocks to 603 ft, at which depth Coal Measures were penetrated (see p. 198). The bands of grey marl between 201 and 296 ft may be the local representatives of the Chillyhill Rock Bed and the Castle Lodge Sandstone of the Chew Stoke area. Two other boreholes yield information on the shape of the pre-Triassic surface in this area. A bore [ST 437 603] adjacent to Churchill church, within 200 yd of, and 55 ft below the level of the nearest Carboniferous outcrop, proved over 203 ft of hard and soft red and mottled marls, while 1.5 miles to the north-east a borehole at Langford House [ST 459 611] showed only 72 ft of marls, mostly red, resting directly on Coal Measures (Richardson 1928, p. 44). G.W.G.

To the north-west of Churchill the Butcombe Sandstone is seen between Honey Hall, Brinsea, and Brean Bridge where somewhat indurated grey-green marls with subordinate sandstone give rise to small features, which are also to be seen farther east between Wrington and Lye Cross. North of Lye Cross the bed forms a small ridge composed largely of a hard conglomeratic rock. A section of this rock (E22992), described by Prof. K. C. Dunham, shows a grey fine conglomerate of limestone fragments (some partly chertified) with fragments and grains of quartz and quartzite and rare pieces of felsite in a matrix of calcite with scattered areas of fibrous celestine. F.B.A.W.

Burrington–Butcombe–Nempnett Thrubwell–Ubley

The Dolomitic Conglomerate is well exposed at many places along the northern flanks of the Mendips. The north-south part of Buffington Combe is cut partly in Carboniferous Limestone and partly in Dolomitic Conglomerate which here infills a Triassic valley some 300 to 400 ft in depth. There are numerous exposures of red conglomerate, a roadside section [ST 487 590] opposite the café (see (Figure 19)) shows about 20 ft of very coarse conglomerate, poorly bedded but apparently horizontal, while, about 200 yd to the south-east, 25 ft of conglomerate with a northerly dip of 10° is seen. The lateral passage from Dolomitic Conglomerate into Keuper Marl was well seen 175 yd south-west of Burrington church, in a road section 50 yd in length which showed about 10 ft of interbedded red conglomerate and sandy marl passing northwards into red sandy marl. South-east of Burrington, yellow conglomerate is well exposed in the sides of old mining trenches immediately north of the Carboniferous Limestone outcrop. The winding valley between Rickford and Blagdon is entirely cut in reddish Dolomitic Conglomerate which is well exposed in the woods on either side. Between Blagdon and Ubley, on the steep northern slopes of the Mendips, red Dolomitic Conglomerate is seen in numerous exposures while the lower slopes, around Ubley, include thick lenses of red marl. A lane section [ST 528 577] 550 yd S.10°W. of the church showed 10 ft of apparently horizontal red marl with many small pebbles of Carboniferous Limestone and an impersistent thin layer of sandy conglomerate. At Cleeve Hill, a roadside excavation [ST 532 576] showed 12 ft of red and green mottled marl dipping at about 4° to the north-north-east.

South of the River Yeo, between Bourne and the Blagdon reservoir, the Keuper succession is intermediate in character between the conglomerate sequence adjacent to the Carboniferous Limestone outcrop and the marl-sandstone sequence of the Yeo and Chew valleys (p. 66). A water bore at Bourne [ST 4874 5995] proved 'pale conglomerates' to a depth of 20 ft, followed by red and green marls to 154 ft, red conglomerate and sandstone to 178 ft and, finally, sandy red and green marls to the bottom at 200 ft. The upper and lower conglomerate in this boring are probably the lateral equivalents of the Butcombe Sandstone and the Chillyhill Rock Bed respectively. Mapping north and east of the borehole indicates that the driller has included with his upper 'pale conglomerates' a lower bed of hard grey and green marl totalling about 10 to 15 ft in thickness. The 'pale conglomerate' is successively overlain by 20–25 ft of variegated grey, green, purple and some red marl followed by a thin, light grey, pebbly conglomerate with fish remains. A single fragment of Saurichthys apicalis Agassiz was collected from a small exposure of the latter [ST 494 598] 70 yd south-south-east of a farm (Wadley Farm) and 1500 yd north-west of Blagdon church. This conglomerate is shown on O.S. Sheet 19 as Penarth Beds due to the discovery therein of 'fish-bones and Modiola minima' by R. H. Valpy (Woodward 1876, p. 80). As however, this sequence of pale conglomerate and variegated marls is similar to that mapped as Butcombe Sandstone north of the Yeo it has been classified as Keuper.

North of the River Yeo the Butcombe Sandstone which lies about 150 ft below the base of the Rhaetic, consists of grey and variegated marls with thin sandy marlstone and conglomerate bands totalling some 15 to 25 ft. It forms a well-marked series of plateaux throughout the area. The fullest proved Keuper Marl sequence is at the western end of the Yeo Reservoir where the section starts just below the top of the Butcombe Sandstone. The sequence, the top of which is exposed in the road cutting [ST 505 604] immediately north of the reservoir dam, is as follows:

The puddle trench of the reservoir was excavated to a maximum depth of 175 ft below ground level 'in red and variegated marls, and a layer of grey dolomitic conglomerate . onto a bed of watertight red marl' (Richardson 1928, p. 211). This conglomerate layer may represent the Castle Lodge Sandstone of the Chew Stoke area (see p. 66). Though the Butcombe Sandstone horizon has not been mapped south of the River Yeo, it is probable that the lower part of its sequence is represented in the upper part of a cutting [ST 503 597] at the disused railway station at the southern end of the reservoir dam, where some 8 ft of green marl overlies red marl. G.W.G.

Exposures near Butcombe of the Butcombe Sandstone horizon include a roadside section [ST 514 614] 540 yd S. 15° W. of the church in grey and red marls with two 1-ft beds of calcareous sandstone, and a stream bank [ST 519 621] 380 yd W. 6° S. of How-grove Farm where the following section was observed:

Farther to the east, the Butcombe Sandstone forms broad ledges around Regilbury Park Farm, Strode and Walnut Tree Farm. The succession which includes more sandstone than is the case farther west, is well seen in a bank section [ST 538 619] 100 yd south-west of Walnut Tree Farm: F.B.A.W.

F.B.A.W.

In the Nempnett Thrubwell area, there are numerous exposures in the Butcombe Sandstone and underlying red marls. The conglomeratic bed at the top of the latter which varies from 1 to 2 ft in thickness has been worked as a building stone. The grey marls underneath, which may become indurated, include thin sandy and conglomeratic layers and a persistent 2 to 3 ft red marl band in the lower part. The most complete section, which is in the sides of the lane, [ST 533 596] 0.5 mile south of Nempnett Thrubwell, is as follows:

In the stream bed [ST 546 607] north of Rookery Farm there are discontinuous exposures of grey marls with thin conglomeratic sandstone layers (Butcombe Sandstone). The Woodford Hill Sandstone, which lies some 60 ft below the Rhaetic, can only be recognized east of Butcombe. It consists of thin red mudstone and manly sandstone and forms the conspicuous shelf on which Rookery Farm stands. Close to the farm, drainage works revealed nodules of celestine, which were also seen, at about the same stratigraphical level, in a silage pit [ST 521 606] at West Town, 0.75 mile west-north-west of Nempnett Thrubwell church. G.W.G.

Chew Stoke area

A generalized Keuper sequence for this area has been given earlier (p. 66). The upper sandstone-marl part of the sequence applies mainly to the area west of the River Chew. The basal beds of the Keuper were proved in the trial borings and shafts of the Chew Stoke Reservoir (see pp. 145–6 and (Figure 15)). Borehole 'A' [ST 571 616] is said to have proved about 15 ft of red and grey Pennant-like sandstone at the base of the Triassic, overlain by some 30 ft of red and mottled marl (Moore 1940, p. 66). These beds are overlain by marls with thin bands of calcareous sandstones which were seen in excavations in the valley floor including the cut-off trench (Picken 1957, p. 341).

The Castle Lodge Sandstone, which locally attains a maximum thickness of 10 ft of sandstone and marl, forms the plateau at Castle Lodge, 600 yd east-north-east of Chew Stoke church at which point it is situated about 200 ft above the base of the Triassic. It is exposed in the stream bed [ST 554 616] 450 yd south-west of Chew Stoke church and in a waterfall [ST 557 621] 150 yd north of the church. South of Chew Stoke the identification of this sandstone is uncertain due to faulting. Its outcrop terminates against the fault which passes along the north side of Gravel Hill at a point about 200 yd south of the Inn at Chew Stoke. It may be represented by a faulted sandstone proved in the upper part of the section during the construction of the byewash at the Chew Stoke Reservoir [ST 570 616]. The Castle Lodge Sandstone is succeeded by about 40 ft of marls which are in turn overlain by red and grey marl and calcareous sandstone of variable thickness and often highly decalcified. These beds, which compose the Chillyhill Rock Bed, form a plateau at Chillyhill Farm west of Chew Magna (Bristol Sheet 264) and probably correlate with similar rocks forming the plateau west of Chew Stoke church. This is capped by sandstone which gives rise to a waterfall [ST 547 615] in the stream bed 750 yd east of Lower Strode Farm. The Chillyhill Rock Bed may also be represented by hard green marl and sandstone seen resting on red marl in the borrow pit [ST 567 608] situated 150 yd west of Woodford House (now demolished). These beds give rise to a well-marked feature extending south-westwards from the pit for a distance of about 0.75 mile.

The Butcombe Sandstone, which lies some 75 ft above the Chillyhill Rock Bed, continues eastwards from Butcombe to Chew Stoke where it gives rise to a well-marked feature. In the 60 ft of Keuper deposits which separate the Butcombe Sandstone from the Tea Green Marl the Woodford Hill Sandstone occupies a median position. The latter forms plateaux on the east side of The Knoll and on Woodford Hill, but has not been recognized east of the River Chew.

On Knowle Hill the Keuper succession differs from that described elsewhere in the Chew Stoke area. Here the upper part of the Keuper consists of at least 60 ft of red marl overlain by the Tea Green Marl and these beds are underlain by an indeterminate mass of red marl and sandstone which is presumed to be the equivalent of the Butcombe Sandstone and lower beds at Chew Stoke. The individual sandstone bands cannot generally be mapped. North of the fault which follows the northern flank of Knowle Hill two impersistent sandstones give rise to small plateaux, the youngest sandstone being adjacent to the fault. These may represent the Castle Lodge Sandstone and the Chillyhill Rock Bed of the area to the west. This correlation is supported by their distance above the Triassic base and a general resemblance in colour and texture.

Moorledge–Stanton Wick–Chelwood

A ridge of Coal Measures (the Bromley Horst p. 145) separates the outcrop of the Keuper rocks into two parts and in the northern portion only the Keuper basement beds are preserved. The latter consist of red sandstone and red sandy marl passing down into a ferruginous sandstone which generally has a vivid red colour. Around Stanton Wick individual sandstone bands can be distinguished locally but owing to the sandy nature of the intervening marls the distinction between sandstone and marl is less well marked than it is farther to the west. South of the Bromley Horst, however, it is seen that the upper part of the Keuper Marl, totaling about 100 ft, is mainly devoid of sandstone, while the lower portion about 150 ft thick consists of calcareous sandstone and marl passing down into the characteristic bright red basal sand and sandstone. Conglomeratic layers with a pebble content of sandstone, ironstone and quartz with occasional fragments of Carboniferous Limestone are sometimes found at the bottom of the basal sands. The whole character of the Keuper succession around Chelwood thus differs from the thick sandstone-marl sequence west of the Chew. Small exposures in red calcareous sandstone and vivid red sands near the base of the Triassic are seen in road cuttings, pits, and stream sections in the dissected sandstone plateau extending from Chelwood to Hunstrete House. East of Red Hill the lower sandy portion of the Keuper becomes progressively thinner whilst the upper manly portion retains or even increases its thickness. Thus at Widcombe Brake and Hunstrete House the Keuper (including the Tea Green Marl) is about 160 ft thick, of which the basal sandy portion totals only about 30 ft. At least part of this attenuation is due to the overlap and thinning out northwards of the lower sandy beds against the Coal Measures. G.A.K.

Compton Martin–East Harptree–Litton

In the wooded valley (Compton Combe) south-west of Compton Martin haematitized conglomerate has been mined for red ochre at a number of places. The best exposure is seen on the east side of a quarry (Cliff Quarry) [ST 542 567] 475 yd south-west of Compton Martin church, where a small level has been driven into the hillside along a strongly ferruginous bed of conglomerate. The rock includes large vughs filled with coarsely crystalline calcite. Between here and the Inn (Wells Way Inn) at the crossroads 0.667 mile south of Compton Martin the conglomerate, within some 300 yd of its junction with the Carboniferous Limestone, is a vivid red colour. A quarry [ST 546 559] situated in the eastern quadrant of this crossroads shows up to 15 ft of bright red, well-bedded conglomerate dipping at 10° to the north-east. The iron in the conglomerate is locally concentrated into stringers of haematite. The large area of Dolomitic Conglomerate, which extends south-eastwards from Compton Martin as far as the Biddle Fault, includes ill-defined lenses of red marl some of which have been mapped e.g. near Beaconsfield Farm; here a borehole [ST 5510 5626] proved 74 ft of red marl overlying 34 ft of red and yellowish, patchily dolomitized conglomerate with red sandy marl bands. Within this area numerous blocks of yellowish much silicified conglomerate (see also p. 95) are seen in many places on the hillside between Compton Martin and the wooded gorge (Harptree Combe) which extends south-westwards from between West and East Harptree. A thin section (E24610) of this rock from a small exposure [ST 5472 5676] 330 yd S. 43° E. of Compton Martin church is described by Mr. R. W. Elliot as follows: 'A sandy chert containing scattered rounded and subangular grains (often 0.06 to 0.12 mm diameter) of quartz with infrequent large fragments of silicified limestone set in a fine-grained interlocking mosaic of quartz and chalcedony. Ragged plates of fibrous baryte showing undulating extinction are not uncommon. Galena occurs as occasional cubes rimmed by haematite and in one instance occurs as an irregular stalagmitic form .'

Near West Harptree the Dolomitic Conglomerate interdigitates with and finally passes northwards into Keuper Marl. A roadside exposure [ST 558 568] 0.25 mile southwest of West Harptree church, within the zone of interdigitation, showed the following section:

An old quarry [ST 559 565] 0.25 mile south-south-west of West Harptree church shows 18 ft of conglomerate, whilst in the wooded gorge (Harptree Combe) extending from the mouth, midway between West and East Harptree, good exposures of conglomerate are seen for 1.5 miles in a south-westerly direction. Within Harptree Combe, near the site of Richmond Castle, up to 20 ft of yellow conglomerate are exposed in old workings [ST 561 557] for manganese and zinc ores. North of Smitham's Hill and eastwards to the Biddle Fault the Dolomitic Conglomerate boundary shown on the map is somewhat generalized since there is an irregular lateral passage southwards into red marl. Northwards this line of passage approximately coincides with the Biddle Fault and accounts for the striking difference in the extent of the conglomerate outcrop on either side of the fault. In this area, between East Harptree and the Biddle Fault, the conglomerate is extensively silicified. The silicification of the Triassic rocks appears to end abruptly along the west side of the fault, possibly due to the presence of marl on the east side of the fault. An exception is, however, provided by a small outcrop of conglomerate, adjoining the east side of the Biddle Fault, 700 yd S. of Eastwood Manor. Here a small quarry [ST 5746 5460] exposes 2 ft of unaltered red conglomerate overlying 5 ft (seen) of conglomerate completely altered to chert and impregnated with abundant baryte. Other exposures of this altered rock include two disused quarries nearer to East Harptree. The first [ST 576 551], 200 yd S. of Eastwood Manor showed the following section–Soil 1.5 ft, red micaceous rather fine-grained sandstone the lowest 6 in bleached, 3.25 ft, passing down into conglomeratic chert, seen 4 ft. Thin sections (E23832)–(E23833) of the sandstone, described by Prof. K. C. Dunham, showed that increased redness was correlated with a relatively greater degree of haematite impregnation, while the rocks showed numerous cavities apparently due to the leaching of original (?) carbonate. The lower cherty rock was seen in thin sections (E23834)–(E23835) to be composed of baryte (α = 1.636) showing a flamboyant fibrous texture, chalcedony and quartz. Another quarry [ST 576 560], 1150 yd E. of East Harptree church, reported to have been formerly over 12 ft deep, showed, at the time of the survey, 5 ft of chert and cherty sandstone.

Eastwards from the Biddle Fault to Litton, the Keuper Marl consists of an upper group of red marl some 80–100 ft thick, and a lower group (base not seen) of red marl with (usually) thin beds of red marlstone, red fine-grained calcareous sandstone, pebbly sandstone and conglomerate. The lower group is frequently exposed; an 8-ft conglomerate in a stream [ST 5795 5515] and a 12-ft pebbly red fine-grained calcareous sandstone in a roadside [ST 584 552], respectively 350 and 800 yd west of Eastwood Manor represent some of the thickest rock bands.

Bishop Sutton–Clutton–Farrington Gurney–Chilcompton

Around Bishop Sutton and Stowey and southwards along the Chew valley to Coley the Keuper sequence is approximately 400 ft in thickness. The succession starts with a basal conglomerate some 10 to 30 ft thick, then follows red marl with impersistent bands of red and pale grey calcareous sandstone, whilst the top 130 to 160 ft consists of red marl capped by the Tea Green Marl (about 10 ft). G.W.G.

In the Clutton–High Littleton area, the Keuper Marl, about 150 ft thick, includes much sandy red marl in the lower part and rests directly on Coal Measures. Between Temple Cloud and Farrington Gurney the feather edge of the Dolomitic Conglomerate is exposed in isolated masses at the base of the succession, whilst southwards from the Farrington Gurney area rapid lateral passage into conglomerate takes place until, adjacent to the Carboniferous Limestone outcrop, the sequence is entirely conglomeratic. Thus at Farrington Colliery [ST 641 555], 700 yd east-south-east of Farrington Gurney church, 12 ft of conglomerate is recorded at the base of the Triassic, but farther south a borehole [ST 6280 5526] 900 yd south-west of the church proved 90 ft of conglomerate. At New Rock Colliery [ST 648 506] 850 yd west-south-west of Downside Abbey, over 130 ft was recorded in the shaft. At Ston Easton No. 1 Borehole [ST 6225 5174] 900 yd north-east of Emborough church the lowest 146 ft of a total thickness of 240 ft of Keuper were conglomerate.

The Tea Green Marl averages about 10 ft in thickness throughout the area though it appears to be locally absent in the area between the Chewton Fault and Old Down. F .B. A.W.

Central Mendips

Christon–Shipham

Between Christon and Black Down the excavated core of the Blackdown Pericline is partially infilled with Triassic rocks, mainly Dolomitic Conglomerate. The latter consists of red conglomerate overlain by yellow dolomitized conglomerate with a zone of partly altered rock in between. West of Sidcot the red conglomerate occupies the centre of the basin on the southern margin of which an arcuate outcrop of yellow conglomerate extends intermittently from south of Winscombe eastwards to Hale Coombe, beyond which it broadens out to form the major part of the Dolomitic Conglomerate outcrop between Sidcot and Black Down. In the western half of the Triassic basin the red conglomerate is overlain by Keuper Marl. A borehole [ST 408 577] 0.7 mile north-north-west of Winscombe church proved the following succession: drift 10 ft, marl 20.5 ft, conglomerate 81.5 ft. In the Shipham–Rowberrow area the Dolomitic Conglomerate, which is well exposed, has been extensively exploited for zinc and lead ores. The depths of the workings (Green 1958, p. 83) indicate a thickness of over 300 ft of conglomerate in the central area. A quarry [ST 452 583] about 0.25 mile south-south-east of Rowberrow church exposes 15 ft of approximately horizontal conglomerate. In the valley (Rowberrow Bottom) 0.75 mile east of Shipham up to 20 ft of approximately horizontal red and yellow conglomerate are seen resting on Old Red Sandstone dipping at 40° to the south, while cliffs above the entrance to the cavern (Rowberrow Cavern) [ST 460 580] 1000 yd east-south-east of the inn at Rowberrow show 35 ft of conglomerate with very large boulders of Carboniferous Limestone. G.W.G., F.B.A.W.

Charterhouse–North Hill

Conglomerate is well seen in quarries alongside the Cheddar–Compton Martin road about 0.25 mile south-west of King Down Farm. Several boreholes (see Appendix II) in the neighbourhood of the Castle of Comfort Inn show over 300 ft of Triassic rocks filling up pre-Triassic gorges. Much of the Old Red Sandstone core of the North Hill Pericline is overlain by a large mass of conglomerate, probably over 200 ft in thickness, which has been strongly mineralized. A roadside swallet [ST 5473 5160], one mile south-south-east of the Castle of Comfort Inn shows over 28 ft of red conglomerate with boulders of Black Rock Limestone. Surface debris shows that the eastern edge of the Dolomitic Conglomerate outcrop on Eaker Hill, adjacent to the Biddle Fault, is much silicified. Comparable alteration of the conglomerate adjacent to the Biddle Fault farther north is described elsewhere (p. 75). The road leading eastwards from Red Quarr follows the line of an exhumed pre-Triassic valley for about 0.5 mile, crags of conglomerate up to 20 ft high being exposed on the south side.

Locally, red marls overlie the Dolomitic Conglomerate whilst the succeeding Tea Green Marl–never more than 5 to 6 ft in thickness–may be absent over wide areas due to the overstep of the Rhaetic rocks (Plate 3). Over a small area some 600 yd to the south-east of Red Quarr, the Tea Green Marl represented by a grey-green conglomerate, rests on the Old Red Sandstone. A silage pit [ST 571 511], 1400 yd south-east of Red Quarr, showed 4 ft of stony loam overlying 5.5 ft of grey-green marl with two 10-in red sandy marl bands, the uppermost of which contained baryte. G.W.G.

Keuper south of the Mendips

Isle of Wedmore

The Tea Green Marl in the Weare–Wedmore area, forms a well-marked north-east and north-facing scarp overlooking the moors drained by the River Axe.

In the lane leading south-west from Weare church, the Tea Green Marl includes a 3 ft 9 in black shale band about 20 ft from the top. F.B.A.W.

Wedmore itself, is largely built on Tea Green Marl–the best exposure being in Madwomen's Lane (p. 83). At Mudgley, the red marl noted by Woodward (1876, p. 63) occurs within the Tea Green Marl sequence. Beds near the top of the succession are seen in the side of Stitchling Lane [ST 450 457]-13 ft of banded dark grey-brown sandy shale and paler marls—and on Mudgley Hill (p. 83). A section, 0.5 mile east of Bagley, in the lane (Snake Lane) leading to Panborough was first observed by Dawkins (1864, p. 403) and redescribed by Richardson (1911, p. 53) who stated that 20 ft of red marl were overlain by 20.5 ft of Tea Green Marl, but the present survey suggests that the real thickness of the Tea Green Marl is of the order of 50 ft. The rocks here have a steady southward dip.

Loxton–Draycott

A great thickness of Dolomitic Conglomerate is present on the southern flanks of the Mendips often filling large gullies cut into the Carboniferous rocks. Numerous examples occur between Loxton and Draycott. Many boreholes (e.g. see Appendix H) show that the sub-Triassic surface commonly has a much steeper gradient than the present day Mendip hill slopes.

Sections of the puddle trench of the Bristol Waterworks Reservoir south of Axbridge showed about 75 ft of red marl with green bands 45 to 58 ft above the base. A few faults with throws of 1 to 5 ft were encountered. The dip of the strata is 2° to the south-west but veered to a more westerly direction on the west side of the reservoir (information per Mr. J. A. Picken). A cross-section of part of a pre-Triassic valley side is exposed in the northern face of the quarry [ST 451 555] at The Perch, 1.5 miles north-west of Cheddar, showing the steep-side and the infilling of yellow, much dolomitized, conglomerate with numerous calcite veins. Crags 25 ft high of similar rocks occur in the woods to the west. On both sides of the valley east of Bradley Cross are seen 15 to 20 ft of red and yellow dolomitized conglomerate. Between 0.25 and 0.5 mile east of Bradley Cross the dolomitized conglomerate has been partially silicified, the altered rocks bearing a strong resemblance to those in the Harptree area (pp. 74–6). Thin sections (E25052), (E25053), (E25054), (E25055), (E25056), (E25057), examined by Mr. R. W. Elliot, appear to show that organic debris composed of coarse calcite was relatively resistant to dolomitization (see also p. 22) whereas dolomite was more resistant to silicification than calcite. In the two valleys leading north-eastwards from Batcombe Farm and east-north-eastwards from Draycott church there are good exposures of conglomerate showing the red unaltered rock overlain by patchily altered red and yellow conglomerate and finally, in the higher parts of the valleys, dolomitized yellow conglomerate. This latter may be the equivalent of the Westclose Hill Conglomerate of the area to the south-east. A borehole [ST 4727 5159] 520 yd north-west of Draycott church, starting a little below the conglomerate sequence given above, proved 110 ft of red marls overlying over 90 ft of conglomerate including a 40 ft yellow dolomitized layer; the bottom of this borehole probably lies only a short distance above the base of the Triassic (Fig 7, and Appendix II, p. 202). It is estimated some 300 ft of unaltered red conglomerates and marls separate the two dolomitic horizons at Draycott. Midway between Draycott and Cheddar and about 0.5 mile to the north-west of the borehole last mentioned a recent borehole [ST 4664 5202] proved a thick marl sequence overlying 22 ft of red marly breccia at the base of the Triassic (Appendix II, p. 201). The lower yellow dolomitic horizon of the Draycott area is apparently here represented by a thick bed of green-grey marl 35 ft above the base of the Triassic. G.W. G.

Bands of thin-bedded conglomerate separated by marl were formerly quarried around Draycott under the name of 'Draycott Marble' and used for paving stones, chimney pieces etc. A quarry (Bryscomb Quarry) [ST 477 511] 550 yd north-east of the railway station showed the following section:

Rodney Stoke–Bleadney–Wookey

A borehole [ST 4844 4998] 270 yd E. 30° N. of Rodney Stoke church proved 315 ft of marl, with thin conglomerates between 231 and 251 ft depth (Richardson 1928, p. 119). To the south-east of Rodney Stoke yellow, hard rather massive conglomerate—the Westclose Hill Conglomerate (see p. 68)—gives rise to Westclose Hill; a borehole [ST 4962 4902], 520 yd north-west of Westbury church, proved 30 ft of conglomerate overlying 170 ft of Keuper Marl. To the north-east, the marl, which underlies the conglomerate, passes laterally into the predominantly reddish conglomerate flanking the Carboniferous outcrop of the Mendips (see Horizontal Section, Sheet 280). Westbury and Easton are built on the outcrop of the Westclose Hill Conglomerate of which 20 to 30 ft is seen in the railway cutting at Easton. F.B.A.W.

A borehole (Erlong Lane) [ST 5055 4780], about 0.25 mile south-west of Easton, starting at a level about 15 ft below the Westclose Hill Conglomerate, proved 350 ft of red marl including about 60 ft with red and green bands 100 ft from the bottom (Whittard 1949, p. 482). At Shortwood, south-eastwards from Easton, the conglomerate loses its distinctive yellow colour. Here, an old quarry [ST 518 469] shows 15 ft of approximately horizontal yellowish-red conglomerate, whilst the same band, now red in colour, caps Henley Hill, about 0.5 mile to the south-east. South and south-west of Wookey in the area of Hembury, Ben Knowle and Hay hills, two thin red marlstone bands, at about 100 and 125 ft below the base of the Tea Green Marl, form a discontinuous series of bench features. On the south-eastern side of Hembury Hill a roadside exposure [ST 5086 4414], 200 yd north-east of Castle, shows hard red marl dipping at about 7° to the south-west, with many gypsum veins sub-parallel to the bedding. A borehole [ST 510 449] 0.5 mile to the north, starting at a level of about 100 ft below the Tea Green Marl, proved 250 ft of red marl with, according to the driller, gypsum veins. About 1.5 miles farther to the north-east, a borehole [ST 5285 4645] 180 yd west-north-west of Wookey Station proved a similar depth of marl. The top of this boring is estimated to lie about 150 ft below the base of the Rhaetic.

The Tea Green Marl varies from 45 to 65 ft over most of the area, though adjacent to the Mendips, it thins rapidly and on Wattles Hill, east of Easton, it is only 13 to 15 ft thick. Locally, in the Bleadney–Wookey area, the Tea Green Marl contains nodules of celestine (see p. 167).

Wookey Hole–Wells–Croscombe

Many exposures of conglomerate occur in and around Wookey Hole, those bordering the path to, and in the cave itself, are notable. The rock is of reddish colour and includes many sandstone as well as limestone pebbles. In the south-western face of a large quarry (Underwood Quarry) [ST 539 468] 1000 yd north-east of Wookey Station, can be seen a continuous section of about 5 to 15 ft of red conglomerate resting on Carboniferous Limestone. Near the entrance to the quarry the rapid lateral passage of conglomerate into red marl is clearly seen; the beds have a depositional dip of about 15°.

Around Coxley, a local development of sandstone and conglomerate, estimated to lie about 125 ft below the base of the Tea Green Marl, may be the lateral equivalent of the Westclose Hill Conglomerate of the Easton area. The (abandoned) railway cutting [ST 530 440] at Coxley Wick exposes 6 ft of red marl, containing a 1 ft 4 in-band of greenish-white calcareous sandstone, resting on 4.5 ft of [ST 530 440] conglomerate dipping at 6° to the south-west. This conglomerate or pebbly sandstone is again seen in the laneside 100 yd to the north-north-east, and on the opposite side of the River Sheppey where it forms a well-marked plateau and has been quarried. The best exposure is in a pond [ST 535 439] 710 yd east-north-east of the Inn at Coxley, where 7 ft of grey pebbly sandstone is seen. Mr. R. W. Elliot reporting on a thin section (E27569) of this rock describes the matrix as of crystalline calcite, often coarse, with rhombs of dolomite (ω = 1.679). Red and grey marl and manly sandstone at about the same level forms discontinuous features for two miles eastwards from Coxley. The same horizon is probably represented by the conglomerate in the bottom of the railway cutting, 1000 yd west of Wells Cathedral, where the following section is visible:

Within the City of Wells details of the Keuper succession have been obtained from boreholes (see pp. 204–5). The Westclose Hill Conglomerate and a lower dolomitized horizon are present, the latter being comparable with a similar occurence at Draycott (pp. 77–8). In the area north and west of Wells indurated red marlstone and sandy marls lying above the level of Westclose Hill Conglomerate give rise to low plateaux such as the ridge along which the road to Burcot lies. These marlstones become increasingly but very irregularly developed in the area extending to 0.5 mile north of the railway cutting mentioned above, and may locally pass into red manly sandstone or include conglomeratic layers as in a road section [ST 544 460] some 900 yd west-north-west of Wells Cathedral:

East of Wells, conglomerate underlies much of the rising ground north of the road to East Horrington and was formerly quarried in the grounds of the Mental Hospital. In the wooded valley [ST 570 467] west and north-west of the Mental Hospital up to 18 ft of conglomerate can be seen in the valley sides. D.R.A.P.

A road cutting 350 yd south-east of the Cathedral on the Shepton Mallet road shows 7 ft of red marl overlain by gritty red marlstone with tiny limestone pebbles and numerous calcite-filled cavities. About 0.25 mile farther east a similar section is seen at the entrance to a quarry (Tor Hill Quarry) [ST 559 455]. Here the Keuper rocks show what may be a depositional dip of about 15° to 20° to the south. A thin bright red marly development of the Dolomitic Conglomerate fringes the Carboniferous Limestone inliers in the area south of Dinder, the best exposure being a pit [ST 573 442] on Dulcote Hill where 7 ft of finely conglomeratic marlstone are seen.

A roadbank exposure [ST 5768 4433], 350 yd south-east of Dinder church, shows up to 15 ft of hard whitish conglomerate with a southerly dip of 15° to 24°. The unusually high dips are probably depositional and related to the westerly continuation under shallow Triassic cover of the Carboniferous Limestone ridge exposed 400 yd to the east. G.W.G.

Conglomerate is well exposed between Dinder and Darshill, as in the Sheppey valley where up to 40 ft is seen in bold cliffs east of Croscombe. Good exposures of up to 40 ft of red marl are seen in the railway cutting about 0.75 mile west-north-west of Stump Cross. In the cutting [ST 594 441] 350 yd west-south-west of Stump Cross (on Weston super Mare Sheet 279) some 30 ft of red marl with thin grey bands containing salt pseudomorphs are to be seen. D.R.A.P.

The Tea Green Marl is seen in two good lane sections north of Wells. One is the well-known 'Milton Lane Section' [ST 549 473] 320 yd east-south-east of Upper Milton, of which the latest detailed description is by Richardson (1911, p. 57), where the 11 ft of Tea Green Marl consists of grey-green and whitish marl and marlstone. Woodward (in Bristow, Etheridge and Woodward 1873) records fish scales from two of the marlstone horizons. In the laneside [ST 556 472] 600 yd north-east of Stoberry Park the following section, including 17 ft 9 in of Tea Green Marl was observed:

The pebbles in the conglomerate are mostly derived from the Carboniferous Limestone Series and include much chert. Mr. R. W. Elliot has examined a thin section (E27520) of the hard conglomerate 3 ft below the top of the Tea Green Marl and reports that partial silicification of the rock has occurred and that the matrix includes dolomite (ω = 1.679).

South-east of Wells the Tea Green Marl is confined to isolated hills, and on the hill [ST 569 437] 1200 yd south-west of Dinder church, augering has proved a thickness of 15 to 20 ft; at a knoll (The Roundabout) [ST 585 435] 1600 yd south-east of Dinder church, some 28 to 30 ft were proved. On the plateau formed by Tea Green Marl, 1000 yd south-east of Wellesey Farm, the surface brash includes celestine-calcite masses (see also p. 168) and some hard whitish marl containing small cube-shaped cavities, possibly after salt, and rare speckles of galena. In this area, the thickness of the Tea Green Marl is 45 to 50 ft.

South-west of Shepton Mallet, the section in the railway cutting by the Three–Arch Bridge [ST 602 429] 550 yd south-east of Stump Cross (just beyond the southern edge of Sheet 280) was described by Richardson (1911, p. 60). Here, 14 ft 3 in of Tea Green Marl overlie 12 ft of red and variegated marls containing a little celestine and salt pseudomorphs. G.W.G.

Rhaetic north of the Mendips

Christon–Banwell

There are no sections. A small patch of blue-grey Rhaetic clay fills a depression in the Carboniferous Limestone 1 mile north-west of Christon. A trial shaft [ST 387 602] was sunk at Park Farm, north-west of Banwell in the mistaken idea that the dark Rhaetic shales were of Coal Measures age.

Butcombe–Nempnett Thrubwell

Here there is a local development of massive conglomerate formed mainly of pebbles of Carboniferous Limestone, up to 6 in across, set in a calcareous cement which contains broken Rhaetic lamellibranch shells. It was first mentioned by Woodward (1876, p. 80) and subesquently by Richardson (1911, pp. 70–1) who named it the 'Butcombe Conglomerate'. He recorded a section (at Cuckoo's Nest) 450 yd south-east of Nempnett Farm, of conglomerate overlying thin black Rhaetic shale and Tea Green Marl, but at the time of the present survey only 4 ft of conglomerate was visible. The rock was also seen in the top of the steep bank immediately west of Nempnett Farm, and at a point 700 yd east of the farm where it was thinly developed. The most extensive outcrop stretches round the hilltop east of Butcombe church, where it is about 8 ft thick. F.B.A.W.

Chew Stoke–Chelwood

On Gravel Hill and the Knoll (Stonedge Common) 0.5 mile south-south-west of Chew Stoke church, the Rhaetic totals some 10 to 12 ft in thickness, the Westbury and Cotham Beds being present in equal proportions. Isolated outliers of the lower beds cap The Knoll, Woodford Hill, and Knowle Hill. Farther east, good exposures of the Rhaetic were seen during the construction of the reservoir (Stanton Wick Reservoir) [ST 612 609] 635 yd S. of Stanton Wick Farm. Black shale (Westbury Beds) was proved by augering in the reservoir floor. The succeeding Cotham Beds were 5 to 6 ft thick and included a strongly lenticular development of Cotham Marble (see p. 96 for the section). South and east of Chel wood the Rhaetic is estimated to be about 12 ft thick, but there are no exposures. G.A.K.

Paulton–Emborough–Chilcompton

The section of Paulton Hill Pit [ST 659 562] on Frome (281) Sheet, 1100 yd south-east of Paulton church recorded by Buckland and Conybeare (1824, p. 279) shows 3 ft of 'knotty claystone' ( ? Cotham Beds) on 6 ft of 'black marl' ( ? Westbury Beds). The best section is in the railway cutting 600 yd east-south-east of Chilcompton church, described in detail by Richardson (1911, p. 66). Here may be seen 6 ft 6 in of Westbury Beds overlain by 4 ft 10 in of Cotham Beds. Road widening in March 1961 at Old Down, between the Inn and the railway bridge [ST 626 510] to the south, exposed a complete succession of the Rhaetic beds, which were described by Savage (1962). The Westbury Beds were 10 ft 4 in thick of which the lowest 6 ft 4 in comprised thinly-bedded sandstones and shales with vertebrate remains. The basal bone bed was a nodular quartz-pebble conglomerate up to 5 inches in thickness. The Cotham Beds included nodules of Cotham Marble up to 5 inches thick and 3 ft across. In an old quarry [ST 612 510] 400 yd south-west of Emborough church some 7 ft of dark Rhaetic shale overlie Millstone Grit. A formerly exposed section [ST 622 510] 1000 yd south-east of Emborough church, which showed 10 ft 7 in of Westbury Beds overlain by about 5 ft of Cotham Beds, was described by Richardson (1911, p. 68). Over half of the Westbury Beds consists of laminated sandstone and there are three distinct layers, including the basal bone-bed, with abundant vertebrate remains. F.B.A.W.

Castle of Comfort Inn–Chewton Mendip

Over much of this area the Rhaetic beds form a well-marked belt of boggy ground due to seepages thrown out from the overlying Harptree Beds (Lower Lias). The base of the Rhaetic may be marked by a line of swallets. The thickness of the Rhaetic beds decreases southwards from a maximum of 15 to 18 ft in the Smitham Hill area to the line (Figure 14) along which they are overstepped by the Lower Lias. No Rhaetic rocks were mapped west of the Castle of Comfort Inn but it is probable traces may locally be present as Woodward (1876, pp. 108–9) records loose sandstone blocks with 'Pullastra arenicola' at Haydon Grange, and traces of shale with 'Pecten (? Valoniensis)' 0.5 mile to the east (Lodmore Farm), while a borehole near the Inn proved a few inches of sandstone with fish remains (see Appendix II, p. 203). In the Castle of Comfort area the lowest few feet of the Rhaetic may locally consist of yellow sand; this is especially so where the Rhaetic succession is thinnest. The upper part of the Rhaetic may also locally pass into a sandy facies but in this case it has not been found possible to separate it from the overlying Harptree Beds (see p. 108). In the sides of a swallet [ST 550 528] 750 yd south-east of the Castle of Comfort Inn over 4 ft of sand near the base of the Rhaetic were seen overlain by 2.5 ft of grey rather silty clay. In the Red Quarr area, 10 to 15 ft of Rhaetic clay and sand are present. To the south-west of the road junction a down-faulted mass of Rhaetic clay gives rise to an extensive patch of boggy ground. About 0.25 mile south of Red Quarr where the Rhaetic overlaps eastwards onto the Old Red Sandstone it passes laterally into pale conglomerate in the same manner as the Tea Green Marl described above (p. 77).

At Greendown, a well [ST 5727 5393] is said to have proved only about 1 foot or so of black shale between the Keuper Marl and White Lias. South-east of Greendown two thin spreads of Rhaetic clay rest directly on a strongly planed surface of Carboniferous Limestone. A swallet section [ST 577 535] 350 yd north-north-east of Grove Farm showed 4 ft of blue-black micaceous shale with some sandy layers and a 1-ft sandy limestone band near the top. The strongly planed surface of Carboniferous Limestone in this area extends west and south of the Rhaetic–Lower Liassic outcrops and terminates in a relatively steep slope of Palaeozoic rocks, which extends in a general south-easterly direction from Greendown to Nedge Hill, and which may have formed a cliff in Rhaetic–Lower Liassic times. G.W.G.

Rhaetic south of the Mendips

Isle of Wedmore

An excavation at the farm (Stream Farm) [ST 419 524] showed abnormally thin Rhaetic Beds sandwiched between the Tea Green Marl and White Lias (see p. 98). The attenuation is probably caused by squeezing out of the beds close to the Weare Fault.

The most northerly point at which Wedmore Stone appears to have been worked is at the south end of Lascot Hill, 500 yd north-north-west of Wedmore church. F.B.A.W.

The road leading from Wedmore to Sand marks the line west of which the Wedmore Stone passes into sand and sandrock. The oncoming of the sandy facies is seen in an old quarry [ST 432 470] 1000 yd south-south-west of Wedmore church where about 2 ft of the typical shelly limestone facies is interbedded with calcareous sandstone layers. The hamlet of Sand is situated on the outcrop of these sands. A thin section (E26584) of sandstone from an outcrop 150 yd N. of the Hall at Sand was described by Mr. R. W. Elliot as 'consisting of subangular and angular grains (0 .06–0 .2 mm diam.) of quartz, granulitized quartz, microcline, orthoclase, microperthite, oligoclase and chert with quite common fragments of organic phosphate. Accessory minerals include tourmaline, zircon, and brookite'. The best section in these sands is in a lane (Madwomen's Lane) [ST 427 471], 1220 yd south-west of Wedmore church, details of which are as follows:

North-westwards from this exposure the sands, some 7 to 8 ft thick, give rise to a belt of sandy ground extending to Blackford where they pass into a very local development of Wedmore Stone, seen in the brook [ST 415 478] 650 yd east of the church.

Between Wedmore and Mudgley, east of the Wedmore Fault, the Wedmore Stone has been worked in many shallow quarries. It is a hard, slightly sandy, shell-fragment limestone averaging about 3 ft in thickness and occurring in three or four beds. The stone, grey when fresh, weathers to a brownish colour. The lenticular nature of the Wedmore Stone is well shown in a stream section [ST 454 473] 700 yd south of Wedmore church, which shows Tea Green Marl underlying and sharply marked off from 1 ft 3 in of black shale, which is succeeded by 2 ft 4 in of typical Wedmore Stone. Within 25 yd to the south-east, the Wedmore Stone passes into black shale only to reappear still farther to the south-east, first as a series of hard grey cementstone nodules up to 1.5 ft thick, then finally as typical shelly limestone 3 to 4 ft thick. On Mudgley Hill the Wedmore Stone gives rise to a well-marked dip slope falling to the south-west. Here, a good stream section [ST 441 460], showing the thickest development of Wedmore Stone in the district, is as follows:

West of the road (B3151) leading south from Wedmore, the black shales overlying the Wedmore Stone form a long dip slope in which a stream section [ST 438 469] exposes several inches of tough blue-hearted sandstone with abundant vertebrate remains.

A section [ST 464 460] showing 21 ft 6 in of Westbury Beds is exposed in a lane side south of Theale, 700 yd south-west of Northload Farm. It is the 'Snake Lane Section' of Dawkins (1864, p. 403), and of Richardson (1911, p. 53) on whose account the following summary is based:

This is the most easterly point at which the Wedmore Stone is recorded and from here westwards to Mudgley its development is patchy.

Yarley area

North of Yarley small outcrops of Rhaetic sand and clay cap Knowle and Chalcroft hills. On the hills south of the Wookey–Panborough road the Rhaetic beds, which are 25 to 30 ft thick, form well-marked dip slopes of dark heavy clay ground in many places. About 1.25 miles south-south-east of Yarley on the northern side of Castle Hill black Rhaetic shale is seen faulted against Keuper Marl.

Wookey Hole–Shepton Mallet

On Wattles Hill, west of Wookey Hole, the Rhaetic beds are about 15 ft thick. North of Wells, the 'Milton Lane Section' [ST 549 473] 320 yd east-south-east of Upper Milton, described in detail by Richardson (1911, p. 57), showed 8 ft 7 in of black shales of the Westbury Beds overlain by 4 ft 8 in of Cotham Beds comprising greenish-yellow shale with 10-in of grey limestone in three beds near the bottom. About 0.5 mile to the east, in the lane 600 yd north-east of Stoberry Park, the Rhaetic beds may be 15 ft thick (see also p. 80). A thin section (E27572) of the basal bone-bed is described by Mr. R. W. Elliot as consisting of irregular plates of calcite and rhombs of dolomite (ω = 1.679) with scattered quartz grains and chert set in a fine matric of calcite and dolomite and phosphatic material. To the south-east of Wells, on two small hills respectively some 1200 yd south-south-west and 1600 yd south-east of Dinder church, the Rhaetic attains a thickness of 12 to 15 ft, whilst farther south, on the southern margin of Sheet 280, the figure has increased to 25 to 30 ft. North and east of Wells the Rhaetic is overlapped by the Lower Lias along a gently sinuous line (Figure 14). G.W.G.

In the Shepton Mallet area the Rhaetic beds may reflect shore-line conditions, the usual clay sequence passing into sandy and conglomeratic beds near the line of overstep by the Lower Lias. Cotham Marble fragments were noted south-west of East Horrington, but so far no Westbury Beds have been identified here. North of Croscombe, sand is developed in the Rhaetic, and a temporary exposure [ST 592 455] 600 yd west-south-west of Crapnell Farm, showed 4 ft of sand and sandstone piping down into the Carboniferous Limestone, and containing Mytilus sp.and Hybodus ? cloacinus Quenstedt, overlain by 4 ft of clay with shelly limestone lenses containing Cardinia sp., Chlamys valoniensis (Defrance), ? Schizodus' ewaldi, Gyrolepis sp.scales, Saurichthys sp.Below the railway viaduct 700 yd N. of Shepton Mallet church at a classic locality (De la Beche 1846, p. 268; Moore 1867, p. 507) the Rhaetic is represented by a conglomerate about 1 ft thick, with fish remains, whilst in a quarry (Bowlish Quarry) [ST 612 442] 1000 yd north-west of Shepton Mallet church, no Rhaetic deposits intervene between the Carboniferous Limestone and the overlying Lower Lias, although Richardson (1911, p. 59) recorded Rhaetic material infilling fissures in the Carboniferous Limestone.

In a shaft [ST 628 431] 100 yd north-west of the railway station at Charlton, Shepton Mallet, black shale recorded at depths between 70 ft 7 in and 76 ft 4 in, is presumed to be of Rhaetic age. About 1 mile south-west of Shepton Mallet, the section in the railway cutting by the Three Arch Bridge (see p. 81) showed 11 ft 4 in of Westbury Beds underlying 4 ft 8 in of Cotham Beds which included, near the base, a hard greenish-grey limestone band (1.5 to 2 ft) with a conglomeratic base containing fish remains and Chlamys valoniensis. Some 2 miles east of Shepton Mallet, thin dark brown sand and sandy clays overlain by clay with fragments of limestone resembling Cotham Marble outcrop near Newman Street. D.R.A.P.

References

BRISTOW, H. W. 1864. On the Rhaetic or Penarth Beds of the neighbourhood of Bristol and the South-West of England. Geol. Mag., 1, 236–9.

BRISTOW, H. W., ETHERIDGE, R. and WOODWARD, H. B. 1873. Vertical Sections of the Lower Lias and Rhaetic or Penarth Beds of Somerset and Gloucestershire. Sheet 46, No. 14. Vert. Sect. Geol. Surv.

BUCKLAND, W. and CONYBEARE, W. D. 1824. Observations on the South-Western coal district of England. Trans. Geol. Soc. (2), 1, 210–316.

DAWKINS, W. BOYD. 1864. On the Rhaetic Beds and White Lias of Western and Central Somerset. Quart. J. Geol. Soc., 20, 396–409.

DE LA BECHE, H. T. 1839. Geology of Cornwall, Devon and West Somerset. Mem. Geol. Surv.

DE LA BECHE, H. T.1846. On the Formation of the Rocks of South Wales and South Western England. Mem. Geol. Surv. 1.

GREEN, G. W. 1958. The Central Mendip Lead-Zinc Orefield. Bull. Geol. Surv. Gt. Brit., No. 14,70–90.

HAMILTON, D. 1961. Algal Growths in the Rhaetic Cotham Marble of Southern England. Palaeontology, 4, pt. 3,324–33.

KELLAWAY, G. A. and OAKLEY, K. P. 1933. Notes on the Keuper and Rhaetic exposed in a Road Cutting at Uphill, Somerset. Proc. Bristol Nat. Soc. (4), 7, 470–88.

KOHNE, W. G. 1949. Exhibited specimens of Mesozoic terrestrial vertebrates. Proc. Geol. Soc., No. 1448,53–4.

KOHNE, W. G. 1956. The Liassic Therapsid Oligokyphus. London.

MCMURTRIE, J. 1912. On a Boring at Puritan, near Bridgwater, in Search of Coal South of the Mendip Hills. Proc. Somerset Arch. Nat. Hist. Soc., 57, pt. 2,25–53.

MOORE, C. 1867. On Abnormal Conditions of Secondary Deposits when connected with the Somersetshire and South Wales Coal-Basin; and on the age of the Sutton and Southerndown Series. Quart. J. Geol. Soc., 23, 449–568.

MOORE, L. R. 1940. Recent boring reaching the Coal Measures at Chew Stoke, near Bristol. Proc. Bristol Nat. Soc. (4), 9, 66–8.

PICKEN, J. A. 1957. The Chew Stoke Reservoir Scheme. J. Inst. Water Engin., 2, No. 4,333–82.

RICHARDSON, L. 1905. The Rhaetic and Contiguous Deposits of Glamorganshire. Quart. J. Geol. Soc., 61, 385–424.

RICHARDSON, L. 1911. The Rhaetic and Contiguous Deposits of West, Mid, and part of East Somerset. Quart. J. Geol. Soc., 67, 1–74.

RICHARDSON, L. 1928. Wells and Springs of Somerset. Mem. Geol. Surv.

ROBINSON, PAMELA L. 1957. The Mesozoic Fissures of the Bristol Channel area and their Vertebrate Faunas. J. Linn. Soc. (tool.), 43, 260–82.

SAVAGE, R. J. G. 1962. Rhaetic Exposures at Emborough. Proc. Bristol Nat. Soc., 30, 275–8.

USSHER, W. A. E. 1911. Excursion to Dunball, Burlescombe, Ilminster, Chard, Ham Hill, and Bradford Abbas (Part I). Proc. Geol. Assoc., 22, 246–54.

WHITTARD, W. F. 1949. Temporary Exposures and Borehole Records in the Bristol Area, IV. Boreholes on Mendip. Proc. Bristol Nat. Soc., 27, 479–82.

WOODWARD, H. B. 1876. Geology of East Somerset and Bristol Coal-Fields. Mem. Geol. Surv.

WRIGHT, T. 1860. On the Zone of Avicula contorta, and the Lower Lias of the South of England. Quart. T. Geol. Soc., 16, 374–411.

Chapter 7 Jurassic

In comparison with the development seen in areas to the east and south, the Jurassic succession in the greater part of the Wells Sheet is thin and incomplete, only the strata up to the top of the Inferior Oolite being preserved. With the exception of the region south and south-west of the Mendips, the Lower Lias is generally thin and condensed, with numerous non-sequences. The Middle Lias is not known and only a remnant of Upper Lias is preserved. The Inferior Oolite is represented only by the Upper Inferior Oolite which has a strongly transgressive base and rests on strata ranging from Carboniferous to Upper Lias.

During Lias and Inferior Oolite times the area over, and to the north of, the Mendips was one of structural instability, in which slight periodic uplift, often accompanied by erosion, took place and locally interrupted deposition. These movements were most marked along the Jurassic 'Mendip Axis'—an area which in post-Middle Jurassic times was arched into a broad anticlinal fold ((Figure 14) and p. 143). South and south-west of the Mendips, however, the deposition of the Lias probably continued throughout the period without any major interruption and a thick succession, totalling some 2000 ft of strata, was laid down.

LOWER LIAS

The following divisions of the Lower Lias have been mapped:

• Lower Lias Clay, including locally the Belemnite Marls^‡21 
• Blue Lias — and their littoral facies
• White Lias - and their littoral facies^‡22 

Different authors have expressed varying opinions as to whether the White Lias should be included as the uppermost division of the Rhaetic or the lowest part of the Lias. Present Survey practice, following long-established custom in the Bristol–Bath area (e.g. Wilson 1891, Vaughan and Tutcher 1903, Tutcher and Trueman 1925), has been to classify the White Lias with the Lias.

The Ammonite Zones of the Lower Lias

The substage names and zones of the Lower Lias, shown overleaf, are those in current use by the Geological Survey. The old Schlotheimia angulata Zone has, however, been divided into a S. angulata Zone above and an Alsatites liasicus Zone below as proposed by Dean, Donovan and Howarth (1961). The subzones so far recognized in the district covered by the Wells (280) Sheet are also included in the table.

White Lias

This is a uniform group of white, pale grey or cream-coloured calcite-mudstones, usually porcellanous in texture, with, in the lower part, interbedded buff-coloured marls and clays. The individual limestone beds vary from a few inches to about 2 ft in thickness. There is a thick and persistent limestone at the top of the White Lias which is known as the Sun Bed or Jew Stone; this may have a bored or fluted upper surface indicating a slightly unconformable junction with the Blue Lias. The White Lias attains a maximum thickness of 20 ft in the Paulton area north of the Mendips, but elsewhere it ranges from a few feet to 8 ft in thickness. A generalized map showing thickness variations in the White Lias of Somerset and Gloucestershire is given by Kellaway and Welch (1948, fig. 15). The somewhat sparse fauna consists largely of small marine lamellibranchs of which the commonest genera include Modiolus, Pteromya [Pleuromya]and Protocardia.

Blue Lias and Lower Lias Clay

The Lower Liassic succession can most conveniently be described in relation to its three main areas of sedimentation (Kellaway and Welch 1948, fig. 15, and pp. 53–4): (1) The Radstock Shelf (2) Mendip Littoral Area (3) Central Somerset Basin.

(1) Radstock Shelf

This covers most of the area north of the Mendips and includes part of the Radstock District, well known from the classic work of Tutcher and Trueman (1925); unfortunately, most of the sections there described are no longer visible. Blue Lias. Tutcher and Trueman's account refers to the Lias outcrops south of the Barrow Vale Fault and at Pau1ton. The succession in these areas will be described in some detail and then a brief account of the Blue Lias stratigraphy in the remaining parts of the Radstock Shelf will be given. The sequence described by Tutcher and Trueman is as follows:

In general this succession shows at the base, limestones with clayey partings representing the Hettangian; in the middle, an attenuated sequence less than 5 ft thick of clays, nodular limestones, and remanie fossil beds representing the Sinemurian; at the top, bioclastic ironshot limestones representing the lowest zones of the Pliensbachian. At about the close of Hettangian times and continuing up into the Sinemurian and lower Pliensbachian, the area was subjected to gentle earth-movements^‡24 , which not only had the effect of restricting sedimentation but also locally resulted in the erosion or condensation of the preexisting Liassic sediments.

The 'Corn Grits' mark the base of the Hettangian limestone group; they consist of three regular beds of pinkish-brown, gritty limestone with ironshot granules, separated from each other by thin clayey partings and reach a thickness of between 1.5 ft and 3 ft. The succeeding limestones are grey in colour, more finely grained, and thinner bedded and pass up into rather irregular nodular limestones alternating with thin shales. Ammonites are common only in and above the so-called 'Planorbis'or Psiloceras Bed which lies a few feet above the base. Hitherto the rocks below this bed have been variously termed the Pleuromya and Ostrea liassica Beds, the Ostrea Beds or the pre-planorbis Beds; but following the discovery of ammonites in them, Donovan (1956, p. 198) suggested that the lower boundary of the P. planorbis Zone should be extended down to the base of the Blue Lias.

Within the area described by Tutcher and Trueman, the Hettangian limestones attain a thickness of some 12 ft in the south, near Hobbs Wall, but from thence thin both westwards, to 5 ft on Clutton Hill, and also southwards until, in the southern part of the Paulton area, they are absent altogether.

The various components of the Sinemurian were described in great detail by Tutcher and Trueman. It must be remembered, however, that all of the individual beds of the following account are seldom present in a single exposure; some may be absent or represented by derived nodules in the overlying stratum.

Resting upon various horizons of the Hettangian and locally even upon the White Lias, is the hard bed of limestone, up to 1 ft thick, which was called by Tutcher and Trueman the Tucklandi Bed'. It contains many derived fragments and phosphatized fossils including Spiriferina walcotti (J. Sowerby) and it represents parts of the Arietites bucklandi and Arnioceras semicostatum zones.

The 'Bucklandi Bed' is followed by the 'Turneri'and Raricostatum clays, separated from one another by a band of hard, blue, yellow-weathering, nodular limestone termed the 'Obtusum Bed' or Obtusum Nodules'. The 'Turneri Clay', usually 1 ft thick but occasionally reaching 4 ft, has at its base the 'Spiriferina Bed'. The latter is about 3 in thick, and in addition to Spiriferina walcotti contains many fossils and much derived material and represents the part of the A. semicostatum Zone not already included in the underlying 'Bucklandi Bed'. The 'Taricostatum Clay' varies in thickness from 2 to 8 in.

At the base of the Pliensbachian is the band of limestone 1.5 to 2 ft thick known as the ‘Armatum Bed'. It appears to be developed only in the neighbourhood of Paulton. It is usually a pale coloured rock with derived fragments of a darker limestone and also gastropods in its basal part. According to Tutcher and Trueman (ibid., p. 604) it 'contains many large specimens of Apoderoceras leckenbyi and other Derocerates, while the derived nodules include innumerable specimens of Echioceras raricostatoides and other species of Echioceras. The bed also contains many specimens of large oxycone Arietids, chiefly Victoriceras victoris and allied species'. In the Paulton area, the remaining part of the U. jamesoni Zone and the T. ibex Zone are represented by the Valdani'and ',Tamesoni'limestones, lithologically very different from the typical fine-grained slightly argillaceous limestones of the Lower Lias, being largely made up of shell and crinoidal debris. These limestones, locally known as the 'Ruckle Beds'^‡25  frequently contain limonitic specks and weather brown and rubbly, so that their field brash resembles that derived from the Inferior Oolite. The greatest thickness of the 'Valdani-Jamesoni'limestones is in the Paulton area, where in Ham Quarry 6 ft was recorded by Tutcher and Trueman (1925, p. 615) and this thick development can be traced eastwards into the area of the Frome (281) Sheet. North of Paulton, however, only 1 ft 6 in were recorded by Tutcher and Trueman from Timsbury Sleight, while to the west, at Hunters Rest, Clutton Hill they may be represented by phosphatized nodules containing belemnites at the base of the Lower Lias Clay (here of P. davoei Zone age according to Tutcher and Trueman) and resting directly on limestones of S. angulata Zone age.

Turning now to the parts of the Radstock Shelf not covered by Tutcher and Trueman's account, the Blue Lias stratigraphy is not known in comparable detail. North of the Barrow Vale Fault in the north-western corner of the Wells Sheet, in the Stowey–Hinton Blewett area, and in the Chewton Mendip–Ston Easton area the Blue Lias is not less than 20 ft in thickness and may locally attain 25 to 30 ft or more. The bulk of the limestones are of P. planorbis to S. angulata zonal age but include at the top thin condensed successions belonging to the Arietites bucklandi Zone (Stowey area) or the Arnioceras semicostatum Zone (Chewton Mendip) similar to the Tucklandi Bed' of the areas to the east. Farther west in the Butcombe–Nempnett Thrubwell area, the Blue Lias has diminished in thickness (6 to 10 ft) possibly due to lateral passage of the upper beds into a clayey facies.

Lower Lias Clay. The full thickness of the Lower Lias Clay is only present north-east of High Littleton, where Tutcher (in Tutcher and Trueman 1925, p. 619) recorded a complete section comprising 120 ft mainly of blue clays and micaceous clays at Timsbury Sleight [ST 654 594]. He stated (p. 622) that the Middle Lias was absent and that the Upper Lias rested directly upon the clays of the Lower Lias. The Lower Lias Clay was assigned by Tutcher to the 'Capricornum'and 'Striatum'zones (in present day usage parts of the P. davoei Zone). South-east of Paulton, just outside the boundary of the Wells Sheet, 180 ft of clay were proved between the Inferior Oolite and the Blue Lias in a colliery shaft (p. 101). Farther south at Chewton Mendip (Coles Lane Quarry) the base of the Lower Lias Clay belongs to the U. jamesoni Zone. Elsewhere the zonal age of the clays overlying the Blue Lias is not known.

(2) Mendip Littoral Area

In the Mendip Littoral Area the Lower Lias rests for the most part on Carboniferous Limestone. Here the Liassic rocks consist mainly of sub-littoral facies comprising pale cream-grey, coarse-grained, bioclastic limestones, and pebbly limestones and conglomerates; the pebbles in the latter consist mainly of limestone and chert derived from the Carboniferous Limestone together with occasional quartz pebbles from the Old Red Sandstone. The weathering out of the pebbles often produces a curious honeycomb rock, first noted by Buckland and Conybeare (1824, p. 303). Fossils in these sub-littoral deposits are locally abundant and include lamellibranchs, gastropods and belemnites.

The sub-littoral facies reaches its thickest development, 50–100 ft, in the Shepton Mallet area where it has been quarried in the past and was known locally as 'Downside Stone'^‡26  (Richardson 1911, p. 58). The bulk of this rock represents the P. planorbis to A. bucklandi zones. In the town of Shepton Mallet it is overlain by clay of the E. raricostatum or U. jamesoni zones. Near Maesbury Camp, 1.5 to 2 miles to the north, the littoral limestone facies has extended upwards to include the U. jamesoni Zone and underlies clay of the P. davoei Zone. West of West Horrington the pebbly limestones pass laterally into 30 to 40 ft of well-bedded limestones, which differ from those of the normal Blue Lias succession only in the smaller proportion of interbedded clay bands. These limestones which appear to range up to and to include the A. bucklandi Zone are overlain by Lower Lias Clay of unknown zonal age.

The thick sub-littoral facies seen on the south side of the Mendips has a less thickly developed counterpart on the north side in the area around the Castle of Comfort Inn. To the east, however, between Smitham's Hill and Emborough, the Lower Lias limestones are silicified and these silicified rocks have been mapped as the Harptree Beds (seep. 94).

The presence of Lower Lias, both as clay and pebbly limestone, in fissures in the Carboniferous Limestone indicates the former wide extension of Lias sedimentation over the area of the present Mendip plateau (Figure 14). Most of these fissure deposits are probably of Lower Pliensbachian age.

The history of the Lower Lias in the Mendip Littoral Area appears first to have been one of rapid limestone deposition on either side of a Mendip island which was diminishing in area through gradual subsidence and the transgression of the sea across its sides. This phase ended with the deposition of the limestones of the A. bucklandi Zone. In later Sinemurian times there appears to have been a pause in sedimentation similar to that found in the Radstock Shelf; this was followed by subsidence and widespread transgression during which clays, representing the U. jamesoni and P. davoei zones were laid down. Locally, however, as in the Beacon Hill area, sub-littoral sediments of Downside Stone facies were being deposited as late as U. jamesoni Zone times.

A noticeable feature of these sub-littoral deposits is that their thickness greatly exceeds that of their correlatives which were formed farther offshore, and this change of thickness is accompanied by a marked change of facies. Thus at Shepton Mallet, the 80 to 100 ft of Downside Stone (P. planorbis to A. bucklandi zones) is represented at Cannard's Grave, about 1 mile to the south (on Glastonbury Sheet 296) by only 24 ft of limestone with subordinate shale (P. planorbis to A. bucklandi, and ?A. semicostatum zones) (Donovan 1958a, pp. 393–8). The relatively great thickness of pebbly bioclastic limestone may probably be attributed to the banking-up effects of offshore currents.

(3) Central Somerset Basin

This area embraces most of the ground south-west of the Mendips and includes a smaller area in the north-western corner of the map, north and west of Banwell. In contrast to the shallow water limestone successions of the Radstock Shelf and Mendip Littoral Area the deposits in the basin areas are mainly argillaceous and accumulated in relatively deep calm water. The Blue Lias includes much shale and clay, and north and west of Wedmore, the main body of this formation passes into a series of clays with bands of argillaceous limestone which are included, for mapping purposes, with the Lower Lias Clay. West of Wookey the upper part of the Lower Lias Clay has been removed by post-Jurassic erosion and the highest beds exposed consist of pale grey or whitish clays and marls with occasional manly limestones, up to 150 ft thick. These are correlated with the Belemnite Marls of the Dorset coast (U. jamesoni and T. ibex zones) and probably with the Belemnitiferous Marls (Green in Wilson and others 1958, p. 23) of the Yeovil area.

In the central parts of the basin, such as in the Highbridge area (Kellaway and Welch 1948, p. 52) the Lower Lias reaches a very great thickness. There is evidence indicating a maximum thickness of 200 to 300 ft of banded limestone and shale (including Blue Lias) at the base, followed by 700 to 800 ft of dark grey shale and mudstone, which in turn is overlain by 300 ft of marl and shale of which at least 150 ft may be referred to the Belemnite Marls.

The gradual thickening of the strata towards the centre of the basin can be traced in the exposed Liassic rocks south of the Mendips. At Cannard's Grave about 1.25 miles south-south-east of Shepton Mallet church (on Glastonbury Sheet 296) Donovan (1958a, p. 396) gives the succession above the White Lias as: Blue Lias 24 ft (P. planorbis to ?A. semicostatum zones), Lower Lias Clay (O. oxynotum and ?E. raricostatum zones) 30 ft, pale manly clays and limestone 6 ft seen. The pale marly clays at the top are probably Belemnite Marls. About 9 miles farther west, in the Bleadney area, the succession is as follows: Blue Lias 25 to 30 ft, Lower Lias Clay 100 to 120 ft, Belemnite Marls seen to 150 ft. Still farther west, in the Westham area, sediments ranging from the top of the White Lias to the middle of the A. obtusum Zone total 150 ft.

It will be seen that while the mid-Sinemurian period of non-sequence and diminished sedimentation already noted in the 'littoral area' at Shepton Mallet affects the 'basin facies' south of the town, more continuous sedimentation must have occurred farther to the west.

Upper Lias

The Upper Lias, which consists of sands and ferruginous cephalopod limestones up to 10 ft thick, has only a very limited outcrop north and south of the Mendips on Sheet 280. In the Timsbury area and near Doulting it overlies the Lower Lias unconformably and in both cases is itself overlain unconformably by the Upper Inferior Oolite. Typically the beds contain abundant ammonites which represent a condensed succession of the several zones which normally characterize the Upper Lias part of the Junction Bed in the areas south of the Mendips (Wilson and others 1958, pp. 58–9)^‡27 . G.W.G.

Inferior Oolite

Only the Upper Inferior Oolite is present on the Wells Sheet. Its base is everywhere strongly transgressive and when exposed is seen to be conglomeratic. In the Central Mendips, in the Oakhill–Binegar area, it rests on a strongly planed surface of Carboniferous Limestone, while to the north and south of the Mendips, in the Timsbury and Doulting areas, it rests on a much condensed Upper Lias sequence.

The formation is best developed in the Doulting area, where it consists of a thin basal conglomerate (the 'Cockly Bed' of the quarrymen), overlain first by about 20 ft of rather rubbly and shelly limestone, then by the well-known Doulting Stone freestone horizon 25 to 30 ft thick (S. truelli Subzone)^‡28  and finally, at the top, by about 6 to 10 ft of lighter-coloured rubbly oolitic limestones, (P. schloenbachi Subzone)—the Anabacia Limestone and Rubbly Beds of Richardson (1907, p. 388). A complete section is recorded from the railway cutting 0.5 mile S. of Doulting church (on Glastonbury Sheet 296) by Richardson (1907, p. 390), where the total thickness is 56.5 ft of which the basal conglomerate accounts for 16 in and the top beds 11 ft. Richardson correlated the basal conglomerate with the Upper Trigonia Grit (G. garantiana Subzone) of the Cotswolds.

The Doulting Stone, which has been quarried for building since the early Middle Ages, consists of massive, granular, wedge-bedded bioclastic limestones locally slightly oolitic and weathering to a rubbly limestone near the surface; whole fossils are uncommon. In places the Doulting Stone approaches in character the Downside Stone facies of the Lower Lias (p. 91) on to which formation it presumably overstepped not far north of Doulting. It is therefore possible that the Doulting Stone has been formed by the erosion of littoral Liassic deposits, rather than the Carboniferous Limestone as suggested by Woodward (1894, p. 89; see also Donovan 1958b, p. 135). D.R.A.P.

Harptree Beds

The Harptree Beds comprise a group of highly siliceous rocks of Lower Lias and Inferior Oolite age occurring on the northern side of the Mendip plateau between East Harptree and Oakhill. Silicified Keuper rocks similar in general appearance to the Harptree Beds occur in the Harptree area (see pp. 74–7) but these have not been distinguished on the one-inch map from unsilicified Keuper. A local 'sandrock' development at the top of the Rhaetic in parts of the Castle of Comfort area has been included with the Harptree Beds.

Attention was first drawn to the Harptree Beds by Weaver in 1819 (1824, p. 364) who tentatively dated them as Upper Greensand in age. Subsequently in 1823 Buckland and Conybeare (1824, p. 294) drew attention to cherty rocks of undoubted Triassic age and concluded (with hesitation) that all the strata belonged to the Dolomitic Conglomerate (Triassic). De la Beche (1846, p. 277) recognized the Liassic age of most of the deposits. Woodward in 1868 distinguished both the Triassic and Liassic cherty beds in the revision mapping of Old Series Sheet 18. The same author (1876, p. 105; 1893, p. 123), drawing in part on the accounts of the earlier workers, gave a general account of the strata. Finally, the Inferior Oolite age of some of the deposits was independently recognized by Robinson (1957, pp. 274, 281) and during the recent 6-inch survey.

The most extensive spreads of Harptree Beds occur in two areas, one between the Castle of Comfort Inn on the west and Smitham's Hill on the east, and the other extending from south-west of Chewton Mendip eastwards as far as Emborough. Over the greater part of these areas the beds overlie Rhaetic clays and shales while lateral passage occurs into unaltered White and Blue Lias or littoral Liassic limestone of Downside Stone facies. If the thickness continues without diminution from one facies to the other then the Harptree Beds must range from 20 to over 30 ft in thickness. Sections are confined to the more northerly outcrop where the rocks consist of massive bedded buff coloured cherts with yellowish clayey partings. The best section is at Wurt Pit [ST 559 539] (see also below, p. 108) where the cherts have yielded a fauna of Blue Lias and Downside Stone aspect (Woodward 1876, p. 108; Donovan 1958b, p. 134). Elsewhere the Harptree Bed outcrops are marked by thick spreads of sandy loam with chert debris. There is no evidence for the age of the outliers between Emborough and Binegar while fossil evidence (see below, p. 109) indicates that the Beds between Binegar and Badger's Cross are of Inferior Oolite age.

Thin sections (E23828)–(E23829) of the massive chert from the Castle of Comfort–Wurt Pit area were described by Professor K. C. Dunham as 'secondary silica-rock composed of chalcedony and quartz'. Baryte and idiomorphic crystals of quartz are of common occurrence in the chert, while Dr. R. Dearnley reports the presence of a sphalerite veinlet in chert (E29239) thrown out from old ochre shafts [ST 611 504] 0.5 mile south-south-west of Emborough. The Harptree Beds of Inferior Oolite age in the Binegar–Badger's Cross area are less strongly silicified than those of Liassic age to the north-west. Thin slices (E3004)–(E3005) of rocks collected from near Roemead Farm [ST 617 477] were described by Dr. R. Dearnley as 'silicified orthoquartzite consisting of a mosaic of subangular quartz grains (0.25 mm diameter) showing optically continuous secondary grain enlargement. Small patches of microcrystalline silica (chert) occur apparently replacing quartz grains and associated with secondary silica between the grains'. Fibrous aggregates of baryte are present in both slices.

The Harptree Beds appear to be the result of metasomatic replacement by chert of Lower Liassic limestones and of Inferior Oolite calcareous shelly sandstones. It has long been known (Buckland and Conybeare 1824, p. 294) that similar alteration has affected the Triassic Dolomitic Conglomerate (see pp. 74–7). Professor K. C. Dunham remarks that silicified Triassic rocks from near Eastwood Manor, East Harptree (see p. 75) 'bear a striking resemblance' in thin section (E23834), (E23835) to the Liassic cherts. In the neighbouring areas of Carboniferous Limestone, at the eastern end of the Blackdown Pericline (see p. 166) silicification along the joints may also be related to the same metasomatic processes. The age range of the rocks affected by silicification led Woodward (1893, p. 126) to suggest that the alteration was hydrothermal in origin. The common association of baryte with the chert and the occasional occurrences of galena (see p. 74) and sphalerite support this conclusion.

The alteration is of late or post-Inferior Oolite age and probably later than the main Mendip Lead-Zinc mineralization (see pp. 164–5). The concentration of silicification, particularly marked in the Dolomitic Conglomerate, adjacent to and on either side of the northern half of the Biddle Fault suggests that the latter fault was already in existence at the time of metasomatism. This fault may be a manifestation of pre-Albian earth movements (see p. 143). G.W.G.

Details

White Lias

Butcombe–Nempnett Thrubwell

The White Lias decreases in thickness from between 7 and 8 ft in the east to under 4 ft in the west. It is best seen in an old quarry [ST 540 598] 550 yd north-north-west of Ubley Park Farm where the section as follows:

Old workings adjacent to the Rectory at Nempnett Thrubwell show very similar sections of White Lias with the Sun Bed thicker (about 1 ft 10 in) and showing a tendency to split into two beds. G.W.G.

Chew Stoke

Outliers of Lower Lias limestone, including White Lias, are present on Gravel Hill, 0.75 mile south-west of Chew Stoke and at Stonedge Common 0.5 mile to the south-south-west. The White Lias, measuring some 5 to 6 ft in total thickness, consists of a manly fossiliferous lower portion 2 to 3 ft thick, and a massive unfossiliferous upper part with bands of hard calcite mudstone of the 'Sun-Bed' type. G.A.K.

Hinton Blewett–Stowey–Clutton

South-east of Hinton Blewett the White Lias has an extensive outcrop but is probably nowhere more than 6 ft in thickness. The top part of the formation was exposed beneath the Blue Lias at Stowey (formerly Limekiln) Quarry south of Stowey (see p. 99). At Clutton, according to Bristow and Etheridge (in Bristow and others 1873) the White Lias is represented by only 10 in of marl capped by the Sun Bed 1 ft 6 in. Along the north side of Knapp Hill the top 3 ft of the White Lias may be seen beneath Blue Lias in several old quarries. F.B.A.W.

Stanton Wick–Red Hill

An almost complete section through the White Lias was seen during the construction of the reservoir (Stanton Wick Reservoir) [ST 612 609] 635 yd S. of Stanton Wick Farm. The section, recorded by Mr. R. B. Wilson in 1953, is as follows:

Traces of Blue Lias limestone occur in the soil above the White Lias here. Farther to the east, the extensive limestone plateau of Red Hill is formed by the White Lias.

Chelwood

In the area west of Widcombe Brake the White Lias appears to be about 6 to 7 ft thick. Surface brash of the uppermost 'Sun-Bed' limestone shows U-shaped borings infilled with what appears to be a Blue Lias limestone matrix. Additional evidence for the presence of a non-sequence at the top of the White Lias is seen in Harris's Quarry (see below).

East of Widcombe Brake, the best section of the White Lias is afforded by a quarry (Harris's Quarry) [ST 655 616] 760 yd east-south-east of Hunstrete House, where the section is as follows:

It is estimated that 3 ft of rubbly limestone and marl lying above the base of the White Lias are not exposed in this section. The top of the White Lias is marked by an erosion surface which truncates the internal structures of the Sun Bed. This surface consists of rounded sub-parallel ridges 8 in. in width, 3 in. in height, and a wavelength of 2 ft. Compaction effects due to these ridges and troughs persist for 1 to 2 ft into the overlying Blue Lias, thus proving that the White Lias had become consolidated and eroded before the deposition of the Blue Lias. G.A.K.

Paulton

In the Paulton Hill Colliery shaft section [ST 659 562], (on Sheet 281) Buckland and Conybeare (1824, p. 279) recorded Sun Bed 1 ft 6 in, White Lias 12 ft, Blue marl 6 ft. The Blue marl may correspond to that part of the White Lias below the massive stone normally quarried. Tutcher and Trueman (1925, p. 615) recorded a persistent fossiliferous clay band 7 to 8 ft below the top of the White Lias in the Paulton area.

Chewton Mendip–Ston Easton–Clapton

The only section visible was in a quarry [ST 641 535] at Clapton where beneath 9 ft of Blue Lias was seen Sun Bed 1 ft 1 in overlying 10 ft of White Lias limestones. Chewton Mendip Nos. 1 and 2 boreholes [ST 6091 5343] and [ST 6072 5316] east-north-east of the village proved 13 ft of White Lias.

About 570 yd S. of Chewton Mendip cross-roads a small exposure showed soft white limestone with patches of chert, while about 0.25 mile farther south the whole succession has apparently passed into chert (Harptree Beds). A similar southwards change takes place between Old Down and Emborough. A recent exposure near the railway bridge south of Old Down Inn proved the lowest beds of the White Lias which, here, consist of fossiliferous rubbly limestone 2 ft, overlain by well-bedded limestone seen to 1.5 ft (Savage 1962). In the Chilcompton railway cutting [ST 653 522] the lowest 3 ft of the White Lias, consisting of fossiliferous more or less rubbly limestone and marl, are exposed (Richardson 1911, p. 66). F.B.A.W.

Greendown

The White Lias has formerly been quarried in many places. The total thickness probably does not exceed 5 ft, and south of Greendown the White Lias abuts against an ancient cliff-line cut in Carboniferous Limestone (see also p. 82).

Wells area

West of Wookey Hole, on Wattles Hill, the White Lias forms a well-marked south-westerly sloping dip slope. The highest bed, possibly the Sun Bed, is a rather massive, somewhat bored, pale limestone with Modiolus langportensis (Richardson and Tutcher). North of Wells at Milton [ST 549 473] in the well-known 'Milton Lane Section' the White Lias totals 7 ft (Brodie 1866, pp. 93–5, Woodward in Bristow and others 1873, Richardson 1911, p. 57). Here the top is poorly defined. Woodward included in the bottom few feet of the White Lias, beds that are now considered to belong to the Rhaetic. In a lane section about 0.5 mile to the east, about 5 ft of White Lias is seen (p. 80). South-east of Wells, near the southern margin of the sheet, there are three small White Lias outliers. In the most easterly outlier, some 1600 yd south-east of Dinder church, the White Lias forms a well-marked dip slope falling to the south-west, and here, in an old quarry [ST 5850 4355], a section comprising 5 ft of well-bedded limestone may be seen. An exposure [ST 566 433] 1750 yd S. 35° W. of Dinder church, starting near the base of the White Lias, shows 4 ft of banded yellow and grey clay and limestone (3- to 4-in beds) overlain by 2 ft of thin-bedded calcite mudstone.

Yarley area

The White Lias is everywhere present but probably does not exceed 5 ft in thickness.

Isle of Wedmore

In the Bagley area the White Lias is about 5.5 ft thick (Boyd Dawkins 1864, p. 403) whilst farther west between Sand and Blackford the average thickness is 6 ft (see also p. 105). G.W.G.

Shallow quarries around the Stoughtons noted by Richardson (1911, pp. 54–5) are now abandoned (see also p. 106). The top bed (here known as the Jew Stone) varies from 4.5 in at Stone Allerton to a maximum of 2 ft (two beds) at Chapel Allerton (Woodward 1876, p. 85). An excavation [ST 419 524] 630 yd south-east of Weare church exposed the following section:

The anomalous thickness of the Rhaetic beds is probably due to the proximity of the Weare Fault. It is possible that the white marl recorded as Rhaetic in this section is basal White Lias. A stream section [ST 418 519] 840 yd south-south-east of Weare church, showed in ascending sequence: blue-buff shale 2 ft; hard grey limestone 2 to 3 in; blue-grey shale 1 ft 3 in; and massive splintery limestone 1 ft 2 in. It is possible that the massive limestone at the top of this section may be the Jew Stone, and the total thickness of the White Lias is about 3 ft. F.B.A.W.

Lower Lias (excluding White Lias and fissure deposits) - details

Stonebridge

North-west of Banwell typical 'basin facies' Blue Lias is developed. In an old quarry [ST 387 600] 650 yd north-west of Stonebridge was seen 4 ft of thin bedded limestone alternating with paper shale containing Psiloceras, dipping at 15° to the south-west, and overlain by blue-grey clay. At 400 yd west-north-west of the above, the following section [ST 3836 6006] was observed:

Butcombe–Nempnett Thrubwell–Chew Stoke

East of Butcombe, the basal Blue Lias ('Corn Grits') are exposed. In a now overgrown quarry, some 6 ft deep, 250 yd south-east of Nempnett Farm, Richardson (1911, p. 71) recorded topmost White Lias underlying rocks ranging up to the Psiloceras (C.) johnstoni Subzone. The total thickness of the Blue Lias here does not exceed 6 to 7 ft. Lower Lias Clay, about 20 ft thick, caps the Blue Lias west of Butcombe. F. B. A.W.

South-eastwards the Blue Lias increases to 10 to 12 ft in thickness. It is succeeded by a few feet of banded clays and nodular limestones some 300 yd west-south-west of Nempnett Thrubwell rectory where ammonites found loose on the outcrop include Alsatites sp., Schlotheimia sp. (s.l.) (A. liasicus Zone). In the area south-west of the rectory, old quarries show up to 3 ft of Blue Lias limestone overlying the White Lias. From a quarry [ST 533 600] immediately east of the rectory, a bed 1.75 ft above the White Lias yielded Astarte camertonensis Moore, Isognomon sp., Liostrea hisingeri (Nilsson), Mactromya sp., Parallelodon hettangiensis (Terquem), Pteromya tatei (Richardson & Tutcher), and P. tatei altior (Richardson & Tutcher). Blue Lias limestone forms an outlier about 0.5 mile north of Nempnett Thrubwell where abundant specimens of Psiloceras (Caloceras)are seen in the surface brash. A quarry section in this area is given above (p. 96). G.W.G.

Some 5 to 6 ft of Blue Lias limestone cap the Lower Lias outliers of Gravel Hill and Stonedge Common, respectively 0.75 mile to the south-west and0.5 mile to south-south-west of Chew Stoke. G.A.K.

Hinton Blewett

An old quarry [ST 585 563] 1250 yd south-west of Hinton Blewett church shows 7.5 ft of thin bedded limestones with shale partings, and including the Psiloceras Bed 3 ft below the top. Fossils collected in situ included Pleuromya aff. costata Young & Bird, Psiloceras planorbis (J. de C. Sowerby) (evolute form = 'P. sampsoni'Portlock) and Psiloceras plicatulum (Quenstedt) indicating the P. planorbis Subzone. A specimen of Psiloceras (Caloceras) johnstoni was found in a loose block thus indicating the presence of the overlying P. (C.) johnstoni Subzone. The total thickness of Blue Lias in this area is probably about 15 to 16 ft and this is overlain by thin Lower Lias Clay. G.W.G., F.B.A.W.

Stowey–Clutton

Here the Blue Lias includes the A. bucklandi Zone and is relatively thick, except in the east near Clutton. A recent (1960) water bore [ST 5916 5840], 200 yd S.E. of White Cross, starting at about the top of the Blue Lias proved a total of 36 ft of 'stone' and 'blue clay' of which at least 26 to 28 ft can be assigned to the Blue Lias (see Appendix II, p. 200). The best exposure, described by Donovan (1956, pp. 191–2), is at Stowey Quarry (formerly Limekiln or Sutton Hill Quarry) [ST 597 586] of the Stowey Quarry and Lime Co.0.5 mile south of Stowey church, where the uppermost 3 ft of White Lias are succeeded by 13 ft of banded limestone and shale in roughly equal proportions. The P. planorbis Subzone, capped by the Psiloceras Bed is 7 ft thick, the rest belonging to the P. (C.) johnstoni Subzone. About 0.25 mile south-east of Stowey church, partial re-excavation of an old quarry [ST 603 593] by Donovan (1956, pp. 192–3) showed in descending order: blue clay 1 ft, black phosphatic material (condensed Coroniceras rotiforme Subzone) resting in hollows in top of next bed below, hard limestone in two beds totalling 1 ft 10 in (Coroniceras conybeari Subzone), limestone with shale partings 3 ft seen (Schlotheimia angulata Zone). Donovan records several species of Coronicerate ammonites from the lower bed of hard limestone; the upper bed includes abundant Vermiceras scylla stoweyense Donovan, and the black phosphatic material, Coronicerate ammonites and Charmasseiceras sp.The top hard limestone is again seen in a cliff, 100 yd east of the quarry, where it overlies nearly 20 ft of Blue Lias, and in a stream bed [ST 608 594] 1000 yd east of Stowey church. The zonal position of the clays overlying the Blue Lias is unknown. The site of Bristow and Etheridge's Clutton section (Bristow and others 1873) is probably at the, now overgrown, quarries 0.5 mile north-north-west of Clutton church, east of the main road. Here the White Lias was succeeded by 6 ft 9 in of banded limestone and shale, followed by a 10-in massive limestone (probably the Bucklandi Bed') at the top. F. B.A. W.

The Lias limestone plateau of Red Hill is mostly capped by White Lias but farther west, south of Stanton Wick Farm, a capping of some 5 ft of Blue Lias is present. Most of this can be attributed to the P. planorbis Zone.

Chelwood–Widcombe Brake

The full thickness of the Blue Lias is present on the east side of the Hunstrete Fault where it attains about 20 to 25 ft and is overlain by about 50 ft of Lower Lias Clay. A section in the basal Blue Lias in Harris' Quarry is given above (p. 97). In the Hunstrete area, the upper part of the Blue Lias is phosphatic and yields Coronicerate ammonites. The lower part of the overlying Lower Lias clay is dark grey or black in colour, similar in lithology to the 'Turneri Clay' and Raricostatum Clay' of Timsbury (Tutcher and Trueman 1925, p. 605).

Some nodular limestone also occurs near the base of the clay. G.A.K.

Clutton Hill area

The Blue Lias thins rapidly when traced southwards from the Widcombe Brake area. An old quarry [ST 651 600] at Hobbs Wall (Tutcher and True-man 1925, p. 620) formerly showed White Lias succeeded by P. planorbis Zone 4.5 ft, overlain by the 'angulata Zone' of which 7 ft was seen. In another old quarry [ST 653 601], near the Hunters Rest Inn, 0.75 mile north-east of Clutton, the P. planorbis Zone, 3 ft 9 in, is overlain by the 'angulata Zone' 1 ft which is followed by 2 ft of blue clay with phosphatic nodules at the base containing 'Belemnites longissimus'(Tutcher and Trueman 1925, pp. 620–1). The uppermost clay of this section represents the base of the Lower Lias Clay and was regarded by Tutcher and Trueman as forming part of the 'Striatum Clays' (i.e. P. davoei Zone). Some 70 ft of Lower Lias Clay, with ironstone nodules, are seen in the sides of the lane leading to the farm (Barrow Hill Farm) at the east end of Blackberry Hill. Tutcher and Trueman (1925, p. 621) recorded a ?1-ft fossiliferous blue limestone band 20 ft above the base with ammonites which indicate the P. davoei Zone. The most important section in the area was that recorded by Tutcher (in Tutcher and Trueman 1925, pp. 621–2; fig. 7) during the construction of the reservoir [ST 654 594] on the hill (Timsbury Sleight) 1 mile north-east of High Littleton. The abridged section for the Lower Lias given below was scaled off from fig. 7 of Tutcher's account:

Paulton

Around Paulton, sections of a number of now overgrown or filled in quarries have been recorded by Tutcher and Trueman (1925, pp. 614–6). These show the progressive overstep of the Spiriferina Bed' (A. semicostatum Zone) at the base of the 'Turneri Clay' from the top of the P. planorbis Zone in the north onto the White Lias in the south. The overstep is accompanied by a southwards thinning of the 'Turneri Clay' itself. In the north, at Ham Quarry [ST 658 563], 600 yd south-east of Paulton church, a summary of the succession is as follows: clay (P. davoei Zone) 1 ft+, ironshot limestones (T. ibex–U. jamesoni zones) 8 ft, clay (E. raricostatum Zone) 2 in, limestone nodules (A. obtusum Zone) 5 to 8 in, clay (C. turneri Zone) 4 ft, limestones (P. planorbis Zone) 4.5 ft on White Lias. In the south, at Phyllis Hill Quarry [ST 656 556], 150 yd south of the hospital, the section, which started at the U. jamesoni Zone, was similar but the 'Turneri Clay' measured only 10 in and rested directly on the White Lias. In this area, the Lower Lias Clay is thicker than at Timsbury Sleight, 180 ft being recorded in the Paulton Hill Colliery shaft [ST 659 562] (on Frome Sheet 281) by Buckland and Conybeare (1824, p. 279).

Clapton–Ston Easton–Chewton Mendip

The Blue Lias forms a plateau over much of this ground and is overlain in places by patches of Lower Lias Clay. The average thickness of the Blue Lias present is 20 to 25 ft but north-east of Chewton Mendip it attains some 40 ft, underlying more than 30 ft of Lower Lias Clay. The Spiriferina Bed' and its associated hard limestone with phosphatic nodules lies at the top of the Blue Lias and comprises much of the field-brash south of Ston Easton.

In a quarry at Clapton (see p. 97) 9 ft of Blue Lias overlies the White Lias. Another quarry [ST 623 541] 700 yd north-north-west of Ston Easton church exposes 20 ft of Blue Lias overlying the Sun Bed (White Lias). The best exposure is in a quarry [ST 601 529], (Cole's Lane Quarry) 650 yd south-east of Chewton Mendip church in which 22 ft of limestones, the base of which is not seen, are overlain by Lower Lias Clay. The section is as follows:

F.B.A.W.

Castle of Comfort area

A borehole adjacent to the Castle of Comfort Inn proved about 40 ft of limestone of the Downside Stone facies, apparently resting on thin (?) Rhaetic sandstone which, in turn, rests on the Carboniferous Limestone (see p. 203). The rocks, which show incipient silicification, yielded Sarcinella (Serpula) socialis (Goldfuss); Camptonectes jamoignensis (Terquem and Piette), Chlamys pollux (d'Orbigny), C. aff. subulata (Munster), Lima (Plagiostoma)punctata (J. Sowerby), Modiolus sp., Myoconcha psilonoti (Quenstedt), Placunopsis pellucida (Terquem) and Terquemia arietis (Quenstedt). About 1000 yd east-south-east of the Castle of Comfort Inn, spoil from a well consisted of blue shale and unfossiliferous limestone intermediate in character between Downside Stone and normal Blue Lias. A large swallet [ST 561 517], 200 yd south of Red Quarr, shows in its sides Downside Stone with Isognomon [Perna] infraliasicus (Quenstedt). Blue Lias limestones are seen in a small area around Greendown adjacent to the Biddle Fault, where about 20 ft were proved in a well [ST 5727 5393]. G.W.G.

Shepton Mallet area

As the Lias is traced northwards towards the Mendips successively higher zones pass into the sub-littoral facies known as the Downside Stone and the lower beds become increasingly pebbly in content. Concurrently with these facies changes, the Lias overlaps the Rhaetic and comes to rest directly on Palaeozoic rocks over wide areas. The Shepton Mallet area is conveniently considered in two parts separated by a large east-west trending fault (Bodden Fault), which runs from Darshill in the west to Bodden and Chelynch in the east, whence it is traced into the adjoining Frome (281) Sheet.

North of the Bodden Fault

Near Newman Street, where there was a marine embayment in Rhaetic times, a normal Blue Lias facies is developed. A shallow roadside quarry [ST 648 447] 500 yd north-west of Newman Street, shows about 10 ft of limestone with clay bands, whilst an active swallet [ST 651 441] 200 yd north-east of the Inn (New Inn) at Chelynch exposes 3 ft of grey and purplish clays with limestone bands yielding Liostrea irregularis (Munster) and Psiloceras (Caloceras) sp.(P. johnstoni Subzone). Farther north, the rocks pass into Downside Stone facies, and some 100 yd south-west of Beacon Farm, a conglomerate band yielded pebbles of Palaeozoic sandstone, limestone and igneous rock.

The Downside Stone facies is well seen in a working quarry (Beacon Farm Quarry) [ST 635 448], 600 yd south-west of Beacon Farm, which yielded Cardinia sp., Chlamys pollux, Lima (Plagiostoma) punctata, Myoconcha psilonoti, Pseudolimea hettangiensis (Terquem) and Terquemia arietis. Donovan (1958b, p. 135) has recorded a detailed section here, showing about 23 ft of strata, mostly buff shelly coarse-grained limestone but with 2.5 ft of limestone approaching White Lias calcite mudstone facies in lithology in the middle of the section. He also reports 'Caloceras' from a loose block (P. planorbis Zone).

The basal beds of the Downside Stone resting with strong unconformity on the Carboniferous Limestone can be seen in a disused quarry [ST 622 446] behind Messrs. Hobbs Bros. factory on the west side of the Bristol road 270 yd south-west of the Downside cross-roads, and at the quarry (Bowlish Quarry) [ST 6125 4420] 1000 yd north-west of Shepton Mallet church. At the former locality fossils collected included Cercomya deshayesi (Terquem), Ctenostreon tuberculatus (Terquem) and Dimyopsis intusstriata (Emmrich) to which may be added Lima (Plagiostoma) valoniensis (Defrance), and Liostrea cf. laevis (J. Sowerby) recorded by Donovan (1958b, p. 135). In the intervening ground thin conglomeratic Rhaetic is present between the Carboniferous Limestone and the Lower Lias (p. 104). Up to 30 ft of Lower Lias (Downside Stone) are well seen in old quarries near the railway viaduct, 700 yd north of Shepton Mallet church, and alongside the line to the south-east. Farther north, the Downside Stone includes higher zones, as in the road cutting [ST 617 449] 10 yd north-east of the railway bridge and 800 yd W. of the Downside cross-roads, where sandy flaggy limestones yielded Platypleuroceras spp. (U. jamesoni Zone). About 2 miles north-north-west of Shepton Mallet, large blocks of Downside Stone obtained from a trench [ST 605 469], 450 yd south-east of the railway halt (Maesbury Halt), also yielded fossils indicative of the U. jamesoni P. davoei zones including: Cirpa fronto (Quenstedt), Gibbirhynchia sp., Lobothyris sp., Piarorhynchia juvenis (Quenstedt), Zeilleria cf. Z. ellioti Ager; Chlamys calva (Goldfuss), Chlamys aff. rollei (Stoliczka), Lima (Plagiostoma) gigantea, Pseudolimea pectinoides (J. Sowerby) and belemnites. In this area, the Downside Stone is succeeded by Lower Lias Clay which was proved in a borehole at Thrupemarsh Farm to a depth of over 50 ft below the surface and can be seen in a stream section [ST 615 462] 700 yd south-east of the farm. Here, near the base of the Lower Lias Clay, 12 ft of dark grey clays with calcareous bands and lenses yielded Androgynoceras sp. juv.cf. maculatum (Young & Bird), indicating the P. davoei Zone.

South of the Bodden Fault

The Lias adjacent to the fault is downfolded and a big thickness of Downside Stone is present. West of Shepton Mallet, a recent (1960) water bore [ST 6090 4396] 380 yd W. 4° S. of the cross-roads at Bowlish, (see Appendix, p. 204) proved 54 ft of Downside Stone (base at 315 ft above O.D.); a trial well at the cross-roads showed 80 ft of Lias limestone (base at 275 ft above O.D.) followed by 18.5 ft of 'very hard stone' [? Lias] which according to Richardson (1928, p. 173) may be Carboniferous Limestone. Farther east, boreholes 650 yd east of Shepton Mallet church [ST 626 436] and 1050 yd south-east of the church [ST 628 431] both proved 65 to 70 ft of Lias limestone overlying thin Rhaetic shales which rested directly on Carboniferous Limestone at 358 and 423 ft above O.D. respectively.

On the east side of Shepton Mallet there are good exposures in the upper part of the Downside Stone. A section on the Fosse Way 880 yd west-south-west of Bodden yielded an A. bucklandi Zone fauna including Spiriferina walcotti (J. Sowerby); Gryphaea dumortieri Joly, Pleuromya cf. angusta L. Agassiz, Coroniceras (Metephioceras) sp.(aff. conybeari J. Sowerby). Another section on the Fosse Way, 150 yd farther north than the last, yielded Lima (Plagiostoma) gigantea, Pseudolimea hettangiensis, and Pseudopecten priscus (Schlotheim). Similar horizons are well seen in an old quarry [ST 629 431] on the east side of the railway, 1250 yd south-east of Shepton Mallet church, and again in the railway cutting south of the church, extending for 0.5 mile eastwards from the Western Region station (see also Moore 1867, p. 506; Woodward 1893, p. 87). In the latter section the limestones are of a sandy manly, irregularly-bedded type intermediate between the Downside Stone and the Blue Lias. Farther north, a lower horizon in the Downside Stone is seen in the long railway cutting about 0.25 mile north-east of Shepton Mallet church, where debris from a lineside quarry [ST 623 440], at a [ST 623 440]point 525 yd N.E. of the church, included Psiloceras (Caloceras) intermedium (Portlock) proving the P. johnstoni Subzone. Nearby, a classic section [ST 620 441] in the old Viaduct Quarry on the east side of the Bristol road (A. 37), shows 30ft of Downside Stone down faulted to the south (Bodden Fault) against Carboniferous Limestone which is capped by thin Rhaetic and Liassic deposits (De la Beche 1846, p. 278; Moore 1867, p. 508, fig. 5, and others). Farther west the Downside Stone is overlain directly by thin Lower Lias Clay. A section [ST 619 440] for a new gasholder, 400 yd N. 18° W. of Shepton Mallet church, revealed 10 ft of grey and black clays with limestone bands with Furcirhynchia sp., Pseudopecten priscus and belemnites, whilst debris from the cemetery adjoining the gasworks, on the west side, yielded Gleviceras sp.aff. subguibalianum S. S. Buckman. These fossils indicate a late E. raricostatum or early U. jamesoni Zone age for the clays, and suggest that the discontinuity above the A. bucklandi Zone, noted farther south (p. 93), may extend into this area. Farther north these beds pass into Downside Stone facies (p. 92) as the Liassic shore-line is approached.

In the Bodden–Doulting area, some 60 ft of grey clay of presumed Lower Lias age intervene between the top of the Downside Stone and the base of the Upper Liassic cephalopod-limestones.

Ham Woods, East and West Horrington

In the southern part of this area the sub-littoral Lias limestones were formerly quarried as 'Chilcote Stone'. The best exposures are along the valley sides of Ham Woods where up to 55 ft of marly and sandy limestones with frequent pebble bands consisting of Carboniferous Limestone and, more rarely, chert fragments are seen. The relatively great abundance of pebbles here may be due to erosion of local deposits of the Triassic Dolomitic Conglomerate.

In the west of the area the basal Lias is in a Blue Lias facies, though including an unusually small thickness of interbedded shale. A small quarry [ST 5656 4725] 1000 yd N. 35° W. of the Mental Hospital (Wells) chapel shows 10 ft of sandy and nodular limestone with Calcirhynchia calcaria S. S. Buckman, Spiriferina tumida (von Buch), Zeilleria sp.and Schlotheimia cf. similis Spath (high S. angulata Zone). The rock includes, in the middle part, many thin white chert bands, whilst some of the brachiopods have been silicified. The chert in a thin rock slice (E26767) is described by Mr. R. W. Elliot as consisting of radiating fibrous chalcedony with irregular patches of quartz. D.R.A.P.

North of Wells

The Blue Lias in this area consists of 32 to 40 ft of mainly medium to fine-grained limestones, ranging up into the A. bucklandi Zone, and this is overlain by up to 15 ft of Lower Lias Clay.

The lane section [ST 556 473] 700 yd north-east of Stoberry Park described above (p. 80), showed that the White Lias is succeeded by about 12 ft of poorly exposed limestones which are in turn followed by 15 ft of thin-bedded somewhat nodular limestones with some clay partings, well seen in a large quarry [ST 557 474] adjacent to the lane. A specimen of Schlotheimia sp.(S. angulata Zone) was collected from a position 10 ft below the Lower Lias Clay, which crops out to the north of and at a level of 5 ft above the top of the quarry. About 1.25 miles north of Wells, the well-known 'Milton Lane Section' (for references, see p. 92) still affords the best exposure in the Lias. Here the White Lias is overlain by a bed 5 to 8 in thick of partially indurated marl, followed by 1 ft 11 in of massive pale grey limestone, and finally by about 35 ft of limestones mostly in beds 2 to 3 in thick, but in places 6 to 10 in. Some of the limestones are nodular and include partings and thin beds of manly shale. Brodie (1866, p. 94) recorded 'Ammonites Bucklandi'and 'A. Conybeari'in bed 1 of his account, 22 to 27 ft above the top of the White Lias. About 550 yd to the north-west, in a quarry [ST 546 477] 720 yd north of Milton Lodge, an exposure of 12 ft of limestones with clay partings yielded, from the top 2 to 3 ft, Psiloceras (Caloceras) intermedium (Portlock) indicative of the P. (C.) johnstoni Subzone. Formerly, higher beds may have been exposed, as Richardson (1909, p. 225) recorded 19 ft of limestones with 'Schlotheimia angulata'in the top 5 ft of this, or a neighbouring, quarry. The junction of the Blue Lias with the overlying Lower Lias Clay is locally marked by swallets, an example [ST 550 477] in which 30 ft of limestone are exposed, being seen 900 yd N. 33° E. of Milton Lodge. G.W.G. At a place [ST 541 482] 1200 yd south-east of Higher Pitts Farm a small exposure of Blue Lias resting upon Black Rock Limestone yielded Psiloceras (Caloceras) sp. cf. reynesi Spath indicating the P. johnstoni Subzone.

North of the main outcrop, about 2 miles north-east of Wells, a small patch of sandy clay, doubtfully referred to the Lower Lias, occupies a shallow depression in the Old Red Sandstone. F.B.A.W.

Yarley area

The Blue Lias was formerly quarried in many places, and on Hembury Hill hollows up to 8 ft deep marked the site of partly filled-in quarries which now expose up to 4 ft of limestone (with thin marl bands). The line of demarcation between the Blue Lias and the overlying Lower Lias Clay is poorly defined. South of the Mudgley Fault the Lower Lias Clay gives rise to heavy wet ground with the overlying Belemnite Marls forming a belt of higher dry ground adjacent to the fault. The marls are seen in a stream section [ST 497 447] 650 yd north-north-east of Hum Farm where 9 ft of whitish marl and marlstone yielded Tropiorhynchia spp.;Chlamys sp.(group of C. rollei Stoliczka), Liparoceras sp., Phricodoceras cf. lamellosum (d'Orbigny), Platypleuroceras spp.and belemnites. Prof. D. T. Donovan comments 'Apart from the Liparoceras which suggests the T. ibex or P. davoei zones, this fauna indicates the U. jamesoni Zone.' In the Yarley area, the Blue Lias measures 30 to 35 ft, the Lower Lias Clay 100 to 110 ft, and the Belemnite Marls more than 150 ft in thickness.

Isle of Wedmore

South of Blackford, the Blue Lias which reaches some 30 to 40 ft in thickness consists of limestones, from 2 to 16 in thick, separated by much interbedded shale and hard marl; a single band of marl measuring some 10 to 12 ft is present at the base. In the Bagley–Panborough area there are scattered exposures of the P. planorbis-P. (C.) johnstoni subzones, the best of which is in an old quarry [ST 469 457] 450 yd south-south-east of Panborough church in which are seen 10 ft of banded (3 to 8 in) limestones and shales in approximately equal proportions. In addition to lamellibranchs, Psiloceras (Caloceras) johnstoni was collected from the highest beds. West of Sand, temporary exposures [ST 427 465] south of the Blackford road showed 5 ft of White Lias limestones and clays overlain by 3 ft of hard fawn-coloured shale; while a pond [ST 418 469], 200 yd W.N.W. of the cross-roads north of Heath House, exposed 2.5 ft of White Lias overlain by 11 in of thinly bedded limestone and shale, followed by 2.5 ft of fawn-coloured shale and hard marl. A well [ST 418 464] 400 yd E.N.E. of Land's End Farm showed, above the water level, 15 ft of banded limestone and shale in the top part of the Blue Lias. Just south of this point, Woodward (1876, p. 102) recorded a section near the base of the Lower Lias Clay consisting of 60 ft of clay and shale with six limestones (individual bands 3 in to 1.5 ft).

Nearer Blackford, in two old quarries [ST 4160 4682], [ST 413 469] respectively 1500 yd and 1900 yd west by north of Sand Hall, and excavated in beds some 10 to 15 ft below the top of the Blue Lias, specimens of Lima (Plagiostoma) gigantea (group), Modiolus minimus (J. Sowerby), Pteromya tatei, Pseudopecten priscus, Protocardia truncata (J. de C. Sowerby) and Psiloceras (Caloceras) sp.(probably johnstoni)were collected.

The southern side of the 'Isle of Wedmore' is formed by Lower Lias Clay the outcrop of which is marked by heavy clay soil with scattered limestone fragments derived from limestone bands and nodules, and the small hills in this area may be due to a local increase in the number of such bands. At Westham, spoil from a septic tank [ST 407 465] 1420 yd S. 7° W. of Blackford church consisted of shales and clay with a cementstone nodule containing Cymbites neglectus Spath, Promicroceras perplanicosta (Spath) and P. aff. planicosta (J. Sowerby) indicative of the A. obtusum Zone. G.W.G.

North of Blackford the lower part of the Lias passes into a clay-limestone sequence in which only a thin basal limestone part, probably under 5 ft thick, can be differentiated as Blue Lias. All quarries in the area are now abandoned. An old working [ST 418 508] 800 yd north-north-west of Washbrook showed some 5 ft of paper shale with a 6-in limestone containing Psiloceras. At a horizon, 200 yd to the north-east, occurred hard splintery limestone with an irony top (possibly the White Lias Jew Stone) which yielded Liostrea sp., Myoconcha psilonoti, Parallelodon hettangiensis, Modiolus sp. ? hillanus and in an adjacent exposure Lima (Plagiostoma) valoniensis. A quarry [ST 410 487] 800 yd south-west of West Stoughton showed 6 ft of shales with two limestone bands 6 in and 12 in thick. Richardson (1911, pp. 54–5) recorded sections (now overgrown) of quarries in the area between Ashton and Stoughton Cross of up to 4 ft of Blue Lias with three limestones (5 in to 11 in) overlying up to 2.5 ft of White Lias limestone in three beds.

A borehole [ST 363 525] (on Weston super Mare Sheet 279), 570 yd west-north-west of Rooksbridge, proved Lower Lias shales and clays with, in the lower part, limestone bands, to a depth of more than 363 ft. The record of 'Ammonites striatus' from the core is indicative of the P. davoei Zone (Richardson 1928, pp. 44–5). F.B.A.W.

Fissure deposits (Central Mendips)

The former wide extension of the Lower Lias beyond the present outcrop is shown by the presence of deposits preserved in fissures in the Carboniferous Limestone of the Mendips (Figure 14). The mode of infilling of such fissures has been recently discussed by Robinson (1957, pp. 266–9).

Charterhouse–Castle of Comfort Inn

Abandoned workings along the line of an east-west trending mineral vein 700 yd S. of Ubley Warren Farm show blue clay at intervals. This may be the locality described by Moore (1867, pp. 491–5).

Amongst surface rubble apparently derived from a fissure [ST 527 563] 600 yd E. 32° N. of Home Farm, Nordrach, was seen rubbly brown limestone with brachiopods including Tropiorhynchia cf. T. deffneri (Oppel) and Zeilleria sp.,probably indicative of a Lower Pliensbachian age. Scattered debris from old workings (now mostly levelled) 1 mile north-west of the Castle of Comfort Inn included whitish clay suggestive of the Belemnite Marls. In this area, spoil from an old shaft, [ST 533 544] 1130 yd E. 37° S. of the cross-roads south-west of Fernhill Farm, included hard brown and reddish limestone of Downside Stone type with scattered Rhynchonelloids, while 1350 yd E. of the same cross-roads mining spoil [ST 538 551] contained partially silicified Downside Stone. Farther east, hard brown Downside Stone can still be seen lining the walls of a mineral stope [ST 545 550] 1000 yd N. of Spring Farm.

Green Ore–Binegar–Shepton Mallet

Mining tip [ST 567 506] 1400 yd W.N.W. of Green Ore, along the line of the Biddle Fault, included rather marly limestone containing Cincta ? Cirpa cf. fronto (Quenstedt) and Piarorhynchia sp.indicative of a post-Hettangian date, possibly Lower Pliensbachian.

At Gurney Slade, the classic Liassic 'dyke' [ST 623 500] on the top of Marchants Hill described by Moore (1867, p. 491), and recently studied by Robinson (1957, pp. 267–9, 280) is best seen on the west side of the Bristol road where it occurs as large wall-like masses left by the quarrying of the surrounding Carboniferous Limestone. The infilling largely consists of typical Downside Stone facies limestone and the contained fauna is dated as 'probably Upper Hettangian'. G.W. G.

The approach cutting to a disused quarry [ST 597 475] 900 yd south-east of Slab House exposes a fissure measuring some 15 ft across and of a type similar to that at Gurney Slade. The south side of the fissure is lined with several inches of vertically bedded Downside Stone welded onto the Carboniferous Limestone. This is succeeded by a jumbled mass of Downside Stone and Carboniferous Limestone fragments up to 5 ft in thickness and, lastly, on the north side of the fissure is a thick mass of light greenish-grey manly clay which penetrates cracks in all the other rocks. A mixture of U. jamesoni and T. ibex Zone fossils occur in both the limestone and the clay. Fossils from the limestone include Cirpa front° ? and Astarte sp. camertonensis –gueuxii group, Acanthopleuroceras sp.,and fossils from the clay include Piarorhynchia juvenis (Quenstedt), P. radstockiensis (Davidson); Acanthopleuroceras aff. binotatum (Oppel) and Platypleuroceras cf. rotundum (Quenstedt), while belemnite fragments are common to both.

The quarry (Windsor Hill Quarry) [ST 614 452] 1850 yd N.N.W. of Shepton Mallet church, includes a fissure filled with rather pebbly, sandy limestone recently described by Ktihne (1956) who dated the filling as 'Charmouthian' ( ?Lower Pliensbachian), although derived material of earlier date is present. G.W.G., D.R.A.P.

Upper Lias

North of the Mendips

A thin development of Midford Sands occurs in three small outliers north of High Littleton, on Blackberry and on Burrow hills, where about 10 ft of yellow loamy sand is exposed. A fourth occurrence is on the hill (Timsbury Sleight) 1 mile north-east of High Littleton, where excavations for a reservoir showed a complete succession which was recorded in detail by Richardson (1907, table III, p. 416) and J. W. Tutcher (in Tutcher and Trueman 1925, pp. 621–2). Here beneath a 3-in band of sandy micaceous limestone with Dumortieria subsolaris S. Buckman' (Dumortieria hemera'), there are 5 ft of yellow sand with some belemnites ('dispansum hemera'), followed by 4 ft of ironshot limestone ('Cephalopod Bed' of Richardson) containing ammonites representative of the 'struckmanni, 'pedicum', 'eseri', 'striatulum'and '?lilli'hemerae. At the base is an impersistent brown marl up to 3 in. in thickness with dactylioceratids ('anguinum'hemera). This sequence of ammonite hemerae in terms of modern zonal terminology extends from the lower part of the Dumortieria levesquei Zone down to the upper part of the Hildoceras bifrons Zone. According to Tutcher the Upper Lias here rests directly on blue clay of Lower Liassic age. F.B.A.W.

South of the Mendips

An attenuated Upper Lias sequence similar to that described north of the Mendips is present in the Doulting area. Here, some 5 to 10 ft of sandy and ironshot limestones with abundant ammonites intervene between (presumed) Lower Lias Clay and the Inferior Oolite. An exposure in a copse [ST 643 434] 500 yd north-west of Doulting church yielded Dactylioceras sp., Grammoceras audax S. S. Buckman, Harpoceras falcifer (J. Sowerby), Haugia variabilis (d'Orbigny), Hildoceras bifrons (Bruguière), Orthildaites sp.and Pachylytoceras jurense (Zieten). Another exposure in a ditch [ST 645 434], 300 yd north-north-west of Doulting church yielded Dactylioceras sp.,Grammoceras penestriatulum S. S. Buckman, and Hildoceras bifrons. The horizons represented by these ammonites, range from the Harpoceras falcifer Subzone through the Hildoceras bifrons Zone up into the Grammoceras striatulum Subzone. D.R.A.P.

Inferior Oolite

Doulting

The basal conglomeratic limestone is seen in surface brash south of Bodden and it was proved in trial shafts on Ingsdon Hill (Richardson 1915, pp. 487–9). At the present time, the main quarry in the Doulting Stone lies 400 yd north-north-east of Doulting church (Plate 2B) with a smaller working quarry some 950 yd north-east of the church. Working faces show 15 to 20 ft of mainly massive stone whilst a pit in the floor of the main quarry proved 24 ft of rubbly limestone underneath. There are no good exposures on the Wells Sheet of the beds above the freestones which, however, produce a prolific brash in the fields east of Doulting.

Richardson (1907, p. 396) estimated that some 6 ft of white fossiliferous oolitic limestone, rubbly above and flaggy below ('Anabacia Limestones') intervened between the freestone and the Fuller's Earth in the quarries between Chelynch and Doulting, but these beds have now been removed by quarrying.

Oakhill

On lithological grounds, the Inferior Oolite outcrop north-west of Oakhill has been included with the Harptree Beds which are described separately below. D.R.A.P.

Clutton area

North and north-east of High Littleton, there are three small Inferior Oolite outliers. Excavations for a reservoir on the hill (Timsbury Sleight) 1 mile north-east of High Littleton exposed a complete section (Richardson 1907, pp. 413–6 and table III; J. W. Tutcher in Tutcher and Trueman 1925, p. 621). Here the basal bed (the 'Conglomerate Bed') is a grey-brown limestone, 6 to 8 in thick, with derived lumps of oolite and earthy limestone and is overlain by 3.5 to 4 ft of pale grey rather sandy limestone with a bored top surface; this is followed by 3 ft of white limestone mixed with grey marl, containing abundant Isastrea and micro-brachiopods and named by Richardson the 'Upper Coral Bed'. Tutcher assigned the 'Conglomerate Bed' to the 'garantianum hemera' and the strata above to the truelli hemera'. F.B.A.W.

Harptree Beds

Castle of Comfort–Greendown

Somewhat inconclusive surface evidence suggests the presence of an outcrop of Harptree Beds 0.5 mile south-west of the Castle of Comfort Inn. This outcrop is separated from the main Harptree Beds outcrop, which lies to the east and north-east of the Inn, by a belt of Lower Lias limestone of Downside Stone facies. Cores from a borehole adjacent to the Inn (Appendix II, p. 203) showed some vaguely defined, pale grey cherty patches in the Downside Stone. The junction between the Downside Stone and the Harptree Beds has been tentatively drawn along a line of swallets.

The main outcrop of the Harptree Beds corresponds to that of the Harptree Hill area of Woodward (1876, fig. 11; pp. 106–8), who gave a generalized section of the strata comprising some 20 ft of mainly cherty beds overlying some 10 ft of sands and red-brown 'sandrock' with Rhaetic fossils. The sands may be developed only over a limited part of the area, and are nowhere exposed at present (see also p. 94). Woodward drew the upper limit of the White Lias at about 7 ft above the base of the cherts.

Spoil from numerous old shafts and pits containing abundant chert debris occurs in the area about 0.33 mile to the north-east of the Castle of Comfort Inn, and probably indicates the site of yellow ochre workings mentioned by early writers. Weaver (1824, p. 364) stated that yellow ochre occurred in clay beds 1 to 3 ft thick and was wrought by means of pits 2 to 3 fathoms in depth. Buckland and Conybeare (1824, p. 294), however, reported that the ochre was worked in sand below the cherts, while De la Beche (1846, p. 277) recorded that it was obtained by washing the friable weathered parts of the chert beds. Woodward (1876, p. 106 footnote) implied that the weathered beds mentioned by De la Beche are probably the same as the clay beds of Weaver's account.

Several shallow quarries [ST 550 542], about 550 yd north-east of Spring Farm show 5 to 7 ft of massive bedded chert dipping at various angles to the north-east. The Harptree Beds are, however, best seen in the sides of a large circular depression, or swallet, known as Wurt Pit [ST 559 539] situated about 1.125 miles east-north-east of the Castle of Comfort Inn. Here about 10 ft of massive bedded chert occur near the top of the pit and the upper beds have yielded a fairly extensive shelly fauna (for the fullest fossil list, see Donovan 1958b, p. 134) including Psiloceras sp.[group of P. plicatulum Quenstedt] indicating the P. planorbis Zone of the Lower Lias.

Eaker Hill

Small outliers of Harptree Beds occur 0.5 mile north-west and north-east, respectively, of the road fork at Red Quarr. There are no exposures. G.W.G.

Bathway–Emborough–Binegar

Harptree Beds cover an extensive area between Bathway and Nedge Hill in the west to near Emborough in the east. There are, however, no sections; the deposit gives rise to flats of brown loam full of large blocks of chert. Swallets are a characteristic feature of this ground. The presence of cherty patches seen in exposures in the White and Blue Lias south of Chewton Mendip has already been mentioned (pp. 97, 102).

Another spread of Harptree Beds lies between Emborough Grove and Binegar. Near Emborough Grove the bands of chest are separated by bands of ochreous clay which were formerly dug for use in the Emborough Fullers Earth and Ochre Works (see p. 168). A short distance to the west of Binegar church the Harptree Beds appear to be overlain by an oval mass (400 yd by 200 yd) of grey clay in which evidence of old workings is still visible. It is said this clay was dug for fuller's earth, and was ground in a mill close by.

Gurney Slade–Badger's Cross

Between Binegar and Gurney Slade the Harptree Beds appear as a coarse granular grit forming a thin, broken-up skin on the surface of the Carboniferous Limestone; but to the west and south of Badger's Cross the deposit is sufficiently continuous to be mapped. Fossils collected by Robinson (1957, p. 281) strongly suggest an Inferior Oolite age for this part of the formation. During the course of the survey a block of Harptree Beds from the top of the quarry [ST 620 493] 450 yd east-south-east of Binegar church yielded a specimen of Garantiana sp.indicating that part of the Upper Inferior Oolite is present.

South-west of Badger's Cross a patch of grey soapy clay up to 4–5 ft thick rests on the Harptree Beds. It is possible that this may represent a much degraded patch of Fuller's Earth preserved on top of the Inferior Oolite. F.B.A.W.

Blocks of silicified sandy limestone or sandstone [ST 617 477], 100 yd south-west of Roemead Farm, yielded specimens of Ctenostreon pectiniforme (Schlotheim). From similar blocks [ST 631 477], 400 yd north-west of Little London chapel (Oakhill), an Inferior Oolite faunal assemblage was collected including Lymnorella sp.and parts of other sponges; Sarcinella socialis (Goldfuss); Lopha marshii (J. Sowerby), Entolium corneolum (Young & Bird), Eopecten sp.(? abjectus Phillips sp.), Oxytoma inequivalve (J. Sowerby) and Pseudolimea sp.There are no exposures in this area but augering reveals yellowish-brown gritty loam. The total thickness of the rocks in the Oakhill area is probably a little over 10 ft. D.R.A.P.

References

BRISTOW, H. W., ETHERIDGE, R. and WOODWARD, H. B. 1873. Vertical Secticns of the Lower Lias and Rhaetic or Penarth Beds of Somerset and Gloucestershire. Sheet 46, No. 14. Wells; No. 16. Clutton. Vert. Sect. Geol. Surv.

BRODIE, P. B. 1866. Notes on a Section of Lower Lias and Rhaetic Beds, near Wells, Somerset. Quart. J. Geol. Soc., 22, 93–5.

BUCKLAND, W. and CONYBEARE, W. D. 1824. Observations on the South-Western coal district of England. Trans. Geol. Soc. (2), 1, 210–316.

DAWKINS, W. BOYD. 1864. On the Rhaetic Beds and White Lias of Western and Central Somerset. Quart. J. Geol. Soc., 20, 396–409.

DEAN, W. T., DONOVAN, D. T. and HOWARTH, M. K. 1961. The Liassic Ammonite Zones and Subzones of the North-West European Province. Bull. Brit. Mus. (Nat. Hist.), 4, No. 10.

DE LA BECHE, H. T. 1846. On the Formation of the Rocks of South Wales and South Western England. Mem. Geol. Surv.,1.

DONOVAN, D. T. 1956. The Zonal Stratigraphy of the Blue Lias around Keynsham, Somerset. Proc. Geol. Assoc., 66, 182–212.

DONOVAN, D. T. 1958a. The Lower Lias Section at Cannard's Grave, Shepton Mallet, Somerset. Proc. Bristol Nat. Soc., 29, 393–8.

DONOVAN, D. T. 1958b. Easter Field Meeting: The Lower and Middle Jurassic Rocks of the Bristol District. Proc. Geol. Assoc., 69, 130–40.

KELLAWAY, G. A. and WELCH, F. B. A. 1948. Bristol and Gloucester District, 2nd edit. British Regional Geology, Geol. Surv.

KÜHNE, W. G. 1956. The Liassic Therapsid Oligokyphus. London.

MOORE, C. 1867. On Abnormal Conditions of Secondary Deposits when connected with the Somersetshire and South Wales Coal-Basin; and on the age of the Sutton and Southerndown Series. Quart. J. Geol. Soc., 23,449–568.

RICHARDSON, L. 1907. The Inferior Oolite and Contiguous Deposits of the Bath–Doulting District. Quart. J. Geol. Soc., 63, 383–444.

RICHARDSON, L. 1909. Excursion to the Frome District, Somerset. Proc. Geol. Assoc., 21, 209–28.

RICHARDSON, L. 1911. The Rhaetic and Contiguous Deposits of West, Mid and part of East Somerset. Quart. J. Geol. Soc., 67, 1–74.

RICHARDSON, L. 1915. The Inferior Oolite and Contiguous Deposits of the Doulting- Milbourne Port District (Somerset). Quart. J. Geol. Soc., 71, 473–520.

RICHARDSON, L. 1928. Wells and Springs of Somerset. Mem. Geol. Surv.

ROBINSON, PAMELA L. 1957. The Mesozoic Fissures of the Bristol Channel area and their Vertebrate Faunas. J. Linn. Soc. (zool.)43, 260–82.

SAVAGE, R. J. G. 1962. Rhaetic Exposures at Emborough. Proc. Bristol Nat. Soc., 30, 275–8.

TUTCHER, J. W. and TRUEMAN, A. E. 1925. The Liassic Rocks of the Radstock District (Somerset). Quart. J. Geol. Soc., 81, 595–666.

VAUGHAN, A. and TUTCHER, J. W. 1903. The Lower Lias of Keynsham. Proc. Bristol Nat. Soc. (3), 10, 2–55.

WEAVER, T. 1824. Geological Observations on Part of Gloucestershire and Somersetshire. Trans. Geol. Soc. (2), 1, 317–68.

WILSON, E. 1891. On a Section of the Rhaetic Rocks at Pylle Hill (Totterdown), Bristol. Quart. J. Geol. Soc., 47, 545–9.

WILSON, V., WELCH, F. B. A., ROBBIE, J. A. and GREEN, G. W. 1958. Geology of the Country around Bridport and Yeovil. Mem. Geol. Surv.

WOODWARD, H. B. 1876. Geology of East Somerset and Bristol Coal-Fields. Mem. Geol. Surv.

WOODWARD, H. B. 1893. The Jurassic Rocks of Britain. 3. The Lias of England and Wales (Yorkshire excepted). Mein. Geol. Surv.

WOODWARD, H. B. 1894. The Jurassic Rocks of Britain. 4. The Lower Oolitic Rocks of England (Yorkshire excepted). Mem. Geol. Surv.

Chapter 8 Pleistocene and Recent

Introduction

Deposits belonging to this time are grouped together as Drift, and can be classified according to origin as periglacial, fluviatile, and estuarine or marine. The area under consideration (Wells Sheet) lay near the southern limits of the major Pleistocene ice-sheets and the glacial deposits of the areas to the north are here represented by periglacial drifts termed Head. Fluviatile deposits are represented by river terraces of gravel and loam, alluvial fans, and freshwater alluvium and in this category may be included peat and calcareous tufa which accumulated under freshwater conditions. Marine or estuarine deposits comprise estuarine alluvium and marine sand (Burtle Beds).

The oldest drift deposits in the area are the Burtle Beds, which contain a fauna of 'warm' and 'temperate' aspects, and were deposited during a period of relatively high sea level. They are now thought (Apsimon, Donovan and Taylor 1961, p. 102) to have been deposited during the Last Interglacial. Some stage in the Last Interglacial may also be represented by a fauna including Palaeoloxodon [Elephas] antiquus (Falconer), hippopotamus, and hyaena, discovered in a fissure in the Carboniferous Limestone at Milton Hill, 0.75 mile north-east of Wells (Balch 1948, p. 143; Donovan 1955, p. 28).

The Last Glaciation is represented by extensive head deposits, particularly at the foot of the Mendips, while the bulk of the Mendip cave faunas, including such forms as mammoth and reindeer, are referable to this period, though detailed correlation with the various Continental phases of this glaciation is not possible. At the close of the glaciation (about 8000 B.c.) the sea level around Southern England stood at least 100 ft below present sea level, but between 6300 and 3500 B.C. there was a relatively rapid eustatic rise in the sea level (the Flandrian Transgression, cf. Zeuner 1959, p. 290) to about the present level. These movements were marked by the formation of deep valleys graded to the 'Glacial sea level', followed by their infilling with alluvial deposits—sandy at first, then grey estuarine clay, with minor peat beds denoting pauses in the general rise in sea level. Godwin (references p. 126) has established the detailed chronology. The main marine incursion was complete during late Boreal times, by about 5000 B.C, and was followed by a long period of peat accumulation over a wide area. The base of this peat in Somerset lies at about Ordnance Datum. A second marine incursion dated at about A.D. 250 resulted in the erosion of some of the peat and the deposition of a younger estuarine clay (the Romano–British Clay) with a surface lying at about the present mean highwater level of spring tides (+18 to 20 ft O.D.). Subsequent sedimentation has been confined to limited peat formation and the deposition of freshwater alluvium and calcareous tufa, further incursions of the sea having been substantially prevented by man-made works (Avery 1955, p. 8).

Burtle Beds

These marine deposits were first noted by Buckland and Conybeare (1824, p. 309), the name Burtle Beds being first employed in the Index of Colours issued for official use by the Geological Survey in about 1844 (Woodward 1908, p. 161). The fullest description is by Bulleid and Jackson (1938, 1942). The Burtle Beds, which may exceed a thickness of 15 ft, rise above the surrounding alluvium of the Somerset levels forming low though sometimes quite extensive mounds up to a height of 25 to 30 ft O.D. They consist of quartz sand and comminuted shell debris with concretionary nodules and layers of shingle and gravel, and contain mammalian remains and marine and rare freshwater shells. The marine shells present a mixed assemblage of forms indicative of both cold and warm climate, as well as species common to sandy, muddy, and rocky habitats. Mammalian remains include bones and teeth of elephant, rhinoceros, horse, auroch Bos primigenius Bojanus, red, fallow and roe deer and hyaena- Crocuta crocuta (Erxleben). Of the freshwater shells the most interesting is Corbicula fluminalis (Müller), a species now extinct in Britain, but still living in warmer climates. The Burtle Beds apparently overlie or fringe 'reefs' or 'islands' of Lower Liassic and Keuper rocks.

Details

The main outcrop, about 1 mile wide, lies around Burtle Hill some 5 miles south-west of Wedmore, where the sands were formerly dug. There are now no good exposures, though the buff and yellow sands with shell debris are still visible in ditches etc. A boring [ST 403 431] at Catcott Burtle (on Sheet 296; Richardson 1928, p. 55) proved 1 ft of peat on 4 ft of sand overlying blue clay (Lower Lias). Bulleid and Jackson (1938, pp. 178, 184–6) recorded 5 species of marine or estuarine shells from the spoil of a well [ST 393 436] 1350 yd south-south-east of River House, where the sand was reported to be 12 ft deep. A small, hitherto undescribed, area of sand, probably Burtle Beds, occurs about 2.5 miles south of Wedmore, where a ditch exposure [ST 440 433] on the west side of the Wedmore road (B. 151) shows 1 to 2 ft of peat overlying 4 ft of buff sandy and pebbly gravel consisting of locally derived Liassic and Triassic debris, whilst nearby, a variable series of sand, sandy clay and gravel was proved by augering. The Lias 'basement' was not seen but its presence, below peat, has been confirmed in shallow borings in the adjacent area to the south-east (Godwin 1955, pp. 164–9).

Head

The deposits mapped under this category are poorly sorted, or completely unsorted, stony or gravelly loam and clay and they are believed to be mainly the products of solifluction under cold conditions. They grade into scree and rainwash material on the hill slopes, and into waterlaid material where they are confined within narrow valleys. The latter are seen flooring dry valleys on the Mendip plateau and in the valley of the Axe near Wookey.

The Head, usually 3 to 6 ft in thickness, varies between the extremes of 1 and 17 ft. The stony debris consists mainly of chert and limestone derived from the Carboniferous Limestone and Dolomitic Conglomerate, except where a source of Old Red Sandstone is available and here the Head includes sandstone and some quartz pebbles as maybe seen in the deposits west and north of Black Down, or south of Beacon Hill. Another type of Head is seen in the relatively thin and mainly loamy deposits in the Chew valley, which are derived from Mesozoic rocks.

A fairly typical example of the larger spreads of Head is that which extends between Burrington and Churchill. This has a flat upper surface which rises gently towards the foot of the Mendips and tongues into Burrington Combe and the valley south-east of Churchill. The Head ranges in thickness from 1.5 to over 15 ft, and in composition from rather dirty gravel to brown loam with scattered stones. The stone content is largely of Old Red Sandstone derivation, though with an admixture of Carboniferous Limestone and chert. This preponderance of Old Red Sandstone debris is of considerable interest as this material can only have come from the high ground forming Black Down via the long limestone gorges of Burrington Combe and the Churchill valley. As the volume of Head material occupying the gorge is small compared with the large spreads beyond the gorge mouths it is clear that the gorges mainly acted as feeders of Head from the uplands, where disintegration of the rocks was most active, to the open ground where it fanned out at the foot of the hills. It is envisaged that the high ground was covered by snow and that the Head material moved down the combes as a sludge saturated with melt water. The valley floors in the normally permeable Carboniferous Limestone were probably deeply frozen, and hence impermeable. The copious melt water accomplished some degree of sorting to form local gravel deposits, while much of the reddish loamy matrix was incorporated from the underlying Keuper Marl.

When they are traced away from the foot of the Mendips, the Head trains tend to split into two or three sheets or terraces separated by steeper slopes of Keuper Marl. In most cases the highest sheet is the only one that extends up slope into the Mendip coombes. The differences in level of the spreads may be due to original differences in topography in the ground upon which the debris moved. The relative absence of drift on the steeper inter-terrace slopes is explained by a smaller initial accumulation of drift due to the increased gradient and its subsequent removal through normal subaerial erosion. While, however, the Head deposits in the area may belong to one main period of formation there are instances in which Head formation took place earlier or later than the main accumulation. Thus in the Churchill–Banwell–Winscombe area, the lowest spreads of Head appear to have been formed by the degradation of the main deposit. On the other hand, e.g. near West Harptree, older Head deposits cap small isolated hills standing above the level of the main spreads.

Little or no lowering of the Mendip coombes has occurred since the Head trains flowed down them except in those coombes which are still occupied by active streams. The coombe with the largest stream is also the one in which the maximum downcutting has occurred—namely at Cheddar where the stream has cut a maximum of 40 ft below the sloping surface of Keuper Marl on which the Head rests. In many places along the margins of the Somerset moors the Head deposits are overlapped by both fluviatile and estuarine alluvium.

Details

North of the Mendips

Churchill–Burrington

The general features of the Head in the Churchill–Burrington area have already been described above. The small patch of Head, which outcrops on a flat-topped hill 1 mile north of Churchill, lies about 30 ft above the main spread to the east and is probably a remnant of an older deposit. In the main spread, excavations 0.25 mile south-south-east of Ladymead Farm, exposed 6 in of loamy soil overlying about 2.5 ft of stony loam resting on Keuper Marl. This may be an exceptionally thin development as Mr. D. C. Findlay (in litt.), of the Soil Survey of England and Wales, observed more than 15 ft in excavations [ST 437 600] south of Churchill church; and about 0.5 mile to the east-south-east, behind the clock tower [ST 444 597] in the village centre, he noted the following section: Reddish-brown sandy loam with few stones, about 1.5 ft, overlying sandstone gravel in dense red-brown sandy clay loam matrix with contorted patches of light-brown stoneless loam, 6.5 ft seen. The shallow valley which runs northwards from Upper Langford is floored by loamy drift, over 5 ft thick in places, overlying gravel. This drift is probably partly fluviatile in origin and deposited by the stream (Langford Brook) which rises to the south of the village. It merges northwards and westwards into the typical Old Red Sandstone Head. Farther east, lane sections [ST 483 598] north of Burrington station show at least 5 to 6 ft of loamy gravel with Old Red Sandstone debris overlying Keuper Marl.

Chew Valley^‡30 

Head deposits border the banks of the Chew and its tributaries for considerable distances and in places underlie the alluvium. North of Compton Martin and West Harptree two tongues of head extend northwards and join in the Heron's Green area. They consist of reddish loam, commonly 5 ft thick, with a very variable, but usually small, gravel content of Triassic conglomerate and chert. Farther east, a third tongue extends up to the mouth of the gorge (Harptree Combe) mile south-east of West Harptree church. A trench section [ST 571 568] at the pumping station, 700 yd west-south-west of Shrowl Bridge showed 7 to 7.5 ft of unsorted angular and subangular gravel in a matrix of red loam and clayey loam passing down into 1 to 1.5 ft of medium to fine, waterworn gravel resting on Keuper Marl. The gravel consisted of Triassic conglomerate and chert, with a little barytes.

Two low hills about 0.5 mile north-west and north-east of West Harptree church are capped by loamy gravel 4 ft or more in thickness. These deposits lie about 50 ft above the level of the Head tongues described above and probably represent the remnants of an earlier deposit, though the stone content indicates a common source. North-east of Heron's Green a small patch of Head consisting of stony loam 2.5 to 4 ft in thickness is of different derivation, as is shown by the abundance of Liassic limestone debris.

Mendip Hills

Sandy loam with angular sandstone debris occurs extensively over the Old Red Sandstone outcrops but since it is indistinguishable from material weathered in situ it cannot be mapped separately. In many places the drift extends on to the Lower Limestone Shale outcrop.

Ill-defined patches of chestnut-coloured loam are of widespread, but very irregular, occurrence on the Mendip plateau, particularly in hollows in the Palaeozoic rocks, but no attempt was made to map them. They may be classified as a type of Head. Between the Castle of Comfort Inn and Beacon Hill these deposits include angular Jurassic chert debris and grade into the Harptree Beds outcrops shown on the map. G.W.G.

Winscombe

Much of the Winscombe valley is floored by Head with Old Red Sandstone debris, similar to that seen in the Churchill–Burrington area.

Priddy

A long tongue of Head extends from the Hunters Lodge Inn through Priddy to a point about a mile to the north-west. The deposit, which varies from stony loam to poorly sorted gravel, is over 5 ft thick in places. A similar type of deposit, formed mainly of Old Red Sandstone and Harptree Beds debris, floors a dry valley south-east of Green Ore. F.B.A.W.

Maesbury Castle–Beacon Hill

An almost continuous belt of Head, largely of Old Red Sandstone origin, fringes the south side of the Old Red Sandstone outcrop, and consists of clayey sand and sandy loam with sandstone and siltstone debris together with occasional chert and barytes fragments. Near the Silurian inlier north of Beacon Farm, tuff and andesite fragments were noted in the drift, and around the Lower Lias Clay outcrops south of Warren Farm the Head becomes more clayey with fewer stones. The best exposure of Head is near Thrupe in a small stream [ST 605 462] 900 yd W.N.W. of Burnthouse Farm where up to 10 ft is seen overlying disturbed Lower Limestone Shale. In boreholes [ST 610 467] at Thrupemarsh Farm, and at the railway halt [ST 6050 4725], west of Maesbury Castle, sandstone rubble was recorded to depths of ? 22 ft and 8 ft respectively; the former figure probably includes some Lower Lias Clay.

Head of similar thickness and composition can be seen in ditches and swallets near Oakhill and Withybrook. Although not shown on the map, sandstone Head also overlies the Old Red Sandstone outcrop, over 6 ft being recorded in a pit [ST 622 464] 600 yd south-east of Warren Farm; some of this debris, however, may be due to disintegration in place.

Head of an unusual composition was mapped south-east of Warren Farm, where ditch sections and augering revealed up to 4 ft of stony and sandy loam overlying up to 4 ft of light grey to brown silty clay. Mineralogical study of the grey clay by Mr. R. W. Elliot revealed a quartz-illite composition. It is possible the clay may be the much disturbed remnant of a former outlier of Mesozoic rocks. D.R.A.P.

South of the Mendips

Axbridge–Rodney Stoke

Much of the lower ground in this area is covered by brown loam with abundant Carboniferous Limestone, and to a lesser extent, Dolomitic Conglomerate debris. In the railway cuttings at Axbridge the Head is 5 to 6 ft thick, whilst two boreholes between Axbridge and Cheddar proved 4.5 ft and 17 ft of stony drift respectively, overlying Keuper Marl (see pp. 173, 201).

The puddle trench of the Bristol Waterworks Reservoir, west of Cheddar, revealed gravelly drift overlying Keuper Marl along the northern and eastern sides, but no gravel was reported along the western side (inform. per Mr. J. A. Picken). On the north the drift was 8 to 11 ft in thickness, and this decreases to about 3 to 5 ft along the eastern side.

Of the various small exposures in Cheddar the best is that in the yard [ST 4550 5313] 100 yd south-east of the railway station, where dirty limestone gravel is seen to a depth of more than 5 ft. The railway cutting [ST 450 536] about 0.375 mile to the north-west exposes over 6 ft of dirty reddish sand with abundant angular and subangular limestone pebbles up to 6 in. in diameter.

South of Lyde Farm, a wide spread of red loam with abundant Triassic debris extends as far as Draycott. Just north of Draycott, a road cutting [ST 474 514] exposes up to 4 ft of angular and subangular dirty gravel with limestone fragments up to 4 in across. About 0.33 mile to the south-west, a railway cutting [ST 470 511] shows 6 ft of sand and gravel, with cobbles at the base, overlying Keuper Marl. G.W.G., F.B.A.W.

Wookey

A long tongue of red pebbly loam and loamy gravel extends from Wookey Hole along the course of the River Axe and passes under the alluvium east of Knowle Hill. In addition to the usual limestone pebbles, sandstone pebbles, possibly Old Red Sandstone material derived from the Dolomitic Conglomerate, are not uncommon. The shape of the pebbles varies from round to subangular and angular. Near the River Axe the deposit may locally pass into river gravel. Woodward (1876, p. 158) noted in the cutting at Wookey Station gravel of very irregular thickness but attaining 8 to 10 ft in places. The long axes of the pebbles were arranged vertically. A borehole [ST 5286 4645], 180 yd W.N.W. of the station proved 8 ft of brown clay overlying 6 ft of brown sand beneath which was 2 ft of gravel overlying Keuper rocks. Farther downstream there are good exposures in the river-banks where the thickness of drift varies from 1.5 to 5 ft.

Wells

South of Prior's Hill, a small patch of sandy drift with Old Red Sandstone debris probably exceeds 5 ft in thickness.

Between Wells and Coxley, to the east of the main road, a small patch of stony red loam more than 3 ft thick caps a low hill. G.W.G.

Cave deposits

The Mendip Hills have long been noted for their caves some of which have yielded human remains and artefacts and considerable Pleistocene faunas. A bibliography of the extensive literature on the subject has been compiled by Donovan (1954). All the faunas are probably referable either to the Last Glaciation or the immediately succeeding period. The most complete succession, which is at Gough's Cave, Cheddar, has been recently redescribed by Donovan (1955); in this account the late Pleistocene history of the Mendips is discussed and a suggested correlation of the cave deposits proposed. This correlation has been later extended to cover the whole of the local Pleistocene succession by Apsimon, Donovan, and Taylor (1961). The following account is based on these two papers.

Two local Palaeolithic flint cultures have been recognized in the cave deposits, the early has been termed 'proto-Solutrian' and the later, formerly identified as Cresswellian^‡31 , is now distinguished as Cheddarian (ibid. 1961, p. 100). No undoubted Mesolithic occupation levels have been discovered. Throughout the period represented by the cave deposits the climate was cold. The start of the Cheddarian was probably accompanied by a climatic change involving a decrease of precipitation, while there was a marked faunal change involving the disappearance from the area of mammoth, rhinoceros, cave-lion, hyaena and possibly bison. The remainder of the fauna persisted, including reindeer, the Irish giant deer, red and roe deer, horse, brown bear, arctic fox and ordinary fox, the variable hare, lemmings (Dicrostonyx and Lemmus)and other small mammals. Birds include ptarmigan. Cave pika is confined to the Cheddarian levels.

The best documented cave-deposits are the following. The famous Hyaena den at Wookey Hole (Boyd Dawkins 1862, 1863) is referable to the 'proto-Solutrian' though human occupation was infrequent. The lack of Cheddarian remains in the cave is probably due to it having become choked with sediment. The deposits filling Bridgend Pot in Ebbor Gorge are also 'proto-Solutrian'. Soldier's Hole, Cheddar Gorge contained a lower, 'proto-Solutrian', level and an upper Cheddarian level. Gough's Cave, Cheddar and Aveline's Hole, Burrington Combe contained deposits referable to the Cheddarian; in the 'proto-Solutrian' period these caves were probably too wet to be habitable. The lower levels of Sun Hole, Cheddar Gorge are also referable to the Cheddarian (Tratman, 1955a).

The cave sediments range from water-laid sand, silt and gravel, to frost breccias and brown and reddish-brown cave-earths with a variable content of frost-shattered limestone fragments. The latest Pleistocene deposits are usually capped by tufaceous stalagmite deposits which may have been formed in Mesolithic times. At Chelm's Combe, near Cheddar, a shallow rock shelter showed a Neolithic occupation and burial, underlain by about 16 ft of scree deposits containing only slight traces of human occupation, and a fauna including reindeer, horse, lemmings, and cave pika. Donovan (1955) suggested that the small thickness of deposit separating the reindeer bones from the Neolithic levels coupled with the absence of the Cheddarian culture, lends some support to the view that the reindeer may have died out before the final disappearance of the Pleistocene fauna.

Post-Mesolithic remains have been found in many Mendip caves and the finds are summarized by Tratman (1955b). G.W.G.

River terraces

River terrace deposits are poorly represented in the district. South of the Mendips, they are confined to the valley of the Sheppey and north of the Mendips to the Litton–Hollow Marsh and Chew Valleys.

Details

Chew valley

Before flooding by the Chew Valley Reservoir, the present day alluvium of the River Chew was seen to be bordered by discontinuous gently sloping belts of greyish silty clay representing the First Terrace (Loam). The level of the top of the terrace was only slightly above that of the alluvium, while part of the terrace underlay the alluvium.

About 800 yd north of Herriot's Bridge excavations [ST 570 589] showed 4 to 5 ft of grey clay with plant debris, the lower foot gravelly, overlying Keuper Marl. A continuation of the deposit extends into the unflooded part of a tributary stream at Heron's Green. Excavations across this valley on the site of the present road embankment showed the following succession of drift deposits:

Bed 1 represents the present day alluvium and was thickest near the stream, Bed 2, the First Terrace (Loam) and Bed 3, Head. Bed 1 expands at the expense of Bed 2, so the combined thickness remains at 5 to 6 ft. G.W.G.

A complete section through the Drift deposits in part of the Chew valley was exposed during the construction of the cut-off trench for the Chew Stoke Reservoir. The maximum thickness of drift, 20 ft, was to be seen at about 300 yd east of the centre of the Bywash which now takes the overflow water of the Reservoir. The basal layer of the drift consists of gravel up to 3.5 ft in thickness but somewhat irregular in its distribution owing to the uneven nature of the underlying floor. Most of the material consists of angular, subangular or rounded limestone with some sandstone, locally derived from Carboniferous and Mesozoic rocks. The gravel is itself covered and overlapped by up to 6 ft of brown silty clay in which there is a thin streak of blue clay developed locally at the base. This clay is identical with the grey clay noted at Heron's Green and elsewhere (see above). The blue and brown silty clay is overlain by about 10 to 12 ft of recent alluvium consisting of red-brown silt with occasional scattered stones of local Liassic and Triassic rocks. The surface of the recent alluvium lies at about 150 ft above O.D.

Details of exploratory boreholes and trial holes, proving the succession of the drift deposits in the neighbourhood of the cut-off trench, are given on pp. 145–6, and in (Figure 15). G.A.K.

Litton–Hollow Marsh

Around Litton the First Terrace is represented by a small dissected gravel spread, probably deposited along a former line of drainage of the River Chew which led north-eastwards through Hollow Marsh to Clutton. The River Chew has cut deeply through this deposit and now flows in a north-westerly direction. Hollow Marsh is floored by a gently sloping flat of dark blue to brown clay less than 5 ft thick which represents the First Terrace (Loam). At its outer margins this deposit passes into loamy clay or brickearth which may represent an earlier terrace; this brickearth is not differentiated from the First Terrace (Loam) on the one-inch map. Around Litton, the brickearth apparently overlies the terrace gravel. F.B.A.W.

Sheppey valley

Downstream from Dinder for about 2 miles to near Woodford there are, on either side of the River Sheppey, the dissected remnants of a gravel terrace. The terrace top rises from river level at Dinder to 15 ft above river level near Woodford. The gravel consists of Carboniferous Limestone debris mixed with Old Red Sandstone, Liassic and Triassic rocks, the fragments commonly being of large size. The matrix may be loamy thus approximating to the head gravel in the Axe valley.

At Coxley a thin spread of rather loamy gravel on the east side of the River Sheppey extends from a height of 15 ft above the alluvium down to alluvium level. In the lane [ST 5305 4369] 700 yd north-east of Coxley church over 4 ft can be seen, and in the churchyard, graves expose Keuper Marl under gravel. G.W.G., D.R.A.P.

Alluvial fan

Much of Wells is underlain by a thick flat-topped spread of gravel. A temporary excavation [ST 546 453] in Priory Road, 900 yd south-west of the Cathedral, showed 12 ft of gravel without the base being reached. The gravel consisted of a mixture of mainly subangular, locally derived Palaeozoic and Mesozoic rocks in a loamy sand matrix. Two boreholes in the City [ST 546 451], [ST 550 454] proved gravel 12 and 18 ft thick respectively. Up to 6 ft of gravel can be seen in the stream banks [ST 546 449] on the south side of Wells at Keward. The gravel is also reported to have been dug north-east of the Bishop's Palace. On the south-western side of Wells it is overlain by alluvium. Remains of mammoth, woolly rhinoceros, and wild boar are reported from the gravel (Balch, 1948, table facing p. 142). G.W.G.

Higher Estuarine Alluvium

The drift designated on the map as 'Higher Estuarine Alluvium' consists of thin spreads either of blue-grey or grey-brown clay and loam overlying locally derived subangular gravel or, most commonly, a mixture of the two. These deposits mantle low ground (up to 35 to 60 ft above O.D.) adjacent to and sloping down to the estuarine alluvium. They have been recognized at intervals around the margins of the moors over wide areas both north and south of the Mendips. They appear to post-date the principal head deposits and pre-date the youngest (Romano-British) estuarine clay. When they were originally surveyed their distribution and lithology was taken to indicate an estuarine origin. Owing to the great tidal ranges in the Bristol Channel area this is clearly a possibility that has to be considered. More recently it has been possible to compare these deposits with others mapped as Head in the Severn valley between Avonmouth (Bristol Sheet 264) and Sharpness (Malmesbury Sheet 251) and it is now considered probable that these, like the 'Higher Estuarine Alluvium', were formed in low-lying areas of impeded drainage. G.W.G.

Details

Wrington–Brinsea

The pebbly type of 'Higher Estuarine Alluvium' can be seen between Poplar Farm and Wrington Station. At about 900 yd east-south-east of Poplar Farm, just north of the road to Langford, the blue-grey alluvial clay ends against a small but definite feature cut in the Keuper Marl at 50 ft O.D. A ditch [ST 447 625] 600 yd east-north-east of Poplar Farm exposed 3 ft of blue-grey clay on 9 in of gravel overlying Keuper Marl.

Lox Yeo River

Bordering the estuarine alluvium there is an extensive spread of dark peaty clay with Old Red Sandstone and Carboniferous chert pebbles extending from a point north-west of Barton for 1.5 miles in an easterly direction. North of Winscombe station a narrow strip of blue-grey clay occurs near Sloughpit Farm.

Cheddar–Rodney Stoke

On the west and south-west sides of the Bristol Waterworks Reservoir west of Cheddar blue-brown pebbly clay occurs above the level of the moor, and mantles the low ridge of Keuper Marl extending from near Hythe north-eastwards towards Cheddar.

West of Draycott and south of Rodney Stoke, dark brown stony clays adjoin the estuarine alluvium and extend up to 40 to 45 ft O.D. A pipe-trench, 0.75 mile south-south-west of Rodney Stoke church, showed 1 to 2 ft of red-brown clay paler at the base resting on Keuper Marl with occasional pockets of gravel at the junction of the two formations. F.B.A.W.

Knowle Moor

A low mound of Keuper Marl, about 0.25 mile south-west of Knowle Hill, is fringed with a thin deposit of pebbly grey clay which has been mapped as 'Higher Estuarine Alluvium'.

Callow Hill–Isle of Wedmore

A mixed deposit of stony loam and grey-brown clay has been mapped at intervals on the rising ground bordering the north sides of Tadham, Westhay and Godney moors. The upper limit of this drift lies at about 35 ft above O.D. G.W.G.

Along the north-eastern margin of the Isle of Wedmore, between Wedmore and Weare, a narrow shelf adjoining the estuarine alluvium and extending up to 35 to 45 ft O.D. is mantled with blue-grey clay near Weare, and reddish-grey and grey- brown loamy clay elsewhere. F.B.A.W.

Estuarine Alluvium

The extensive moors south and north-west of the Mendips are formed of thick estuarine deposits which, on the seaward side, extend down to 100 ft below sea level. Borings prove a succession of sands and clays with occasional thin peat beds extending up to about Ordnance Datum; the highest beds consist of thick blue clay. These are succeeded by a thick peat bed, which is overlain by a layer of blue-grey estuarine clay up to 14 ft in thickness with a flat upper surface at 18 to 20 ft above O.D. Godwin (references p. 126) has shown that the thick peat was deposited between about 5000 B.C. and Romano-British times and that the upper clay (Roman Clay) was deposited during a marine transgression around about A.D. 250^‡32 . The Roman Clay never covered the peat moors south of the Isle of Wedmore though a tongue extended into Godney Moor through the gap east of Panborough. The upper surface of the Roman Clay is traversed by a branching system of channels (see (Figure 8)) filled with silt and very fine sand which probably represents an estuarine creek system comparable to the 'Kleiplaats' of the Isle of Walcheren (Edelman 1950, p. 104). On the landward side of the moors the Roman Clay is partly overlain by thin peat and fluviatile alluvium. G.W.G.

Details

Banwell

An extensive area of estuarine clay occupies the north-western part of the Wells Sheet. A complete succession measuring 59 ft in thickness was proved in a borehole [ST 399 609] 1.125 miles north of Banwell church (see Appendix, p. 198). The surface level of the borehole is 18 ft above O.D. and the 9.5 ft of blue clay at the top is the Roman Clay. A series of trial bores between West Wick and Rolston proved a similar succession to a level just below the main peat bed. F.B.A.W.

River Axe levels

Most of the surface of the moors is composed of the Roman Clay, on the upper surface of which a sandy-silt creek channel system is developed. The silts, when they have topographical expression show up as faintly defined levees, but normally their limits can be determined only by detailed angering; this was only carried out over part of the area. The silt channels increase in depth (5 ft or more) towards the present course of the River Axe, and are best developed on Monk Moor and in the 'elbow' of the Axe north of Theale (Figure 8).

A boring at Rooksbridge [ST 363 525] (on Sheet 279), 4.5 miles west-south-west of Axbridge, starting at 20 ft above O.D. proved the following succession (Richardson 1928, p. 45) in descending sequence: dark clay 14 ft, peat 4 ft, blue clay 24 ft, grey sand and rock 25 ft, dark blue and brownish clay 49 ft, fine sand 4 ft (on Lias). The total drift thickness was 120 ft, and the dark clay at the top clearly represents the Roman Clay. A trial boring [ST 4443 5189] 2050 yd south-west of Cheddar church, starting at 18 ft O.D., proved 14 ft of grey clay overlying peat. On the moors south of Draycott and east of the River Axe the Roman Clay is overlain by thin peat.

Moors south of the Isle of Wedmore

A lobe of the Roman Clay extends southwards into Godney Moor through the well-marked gap between Bleadney and Panborough. The western limit of the clay is hidden beneath peat and lies east of a line drawn southwards from Panborough village. This line was determined partly by augering and partly by plotting the distribution of a curious network pattern of depressions which is developed in the peat when the clay is present underneath (Godwin 1955, pp. 164–5). A boring [ST 473 449] at Drake's Drove, approximately 750 yd south-south-east of the Inn (New Inn) at Panborough, made by Godwin (1955, pp. 181–3; Appendix 2 by Macfadyen pp. 188–90) provides a detailed section near the western edge of the Roman Clay. A generalized summary of the boring, which started at 22.7 ft above O.D., is as follows: peat and detritus mud 4 ft, blue clay with brackish water microfauna 4.5 ft, detritus mud and peat 12.5 ft resting on soft blue-grey clay. Some 4 miles to the west on Monk Moor the Roman Clay, up to 14 ft in thickness, reappears above the main peat bed which covers the intervening ground. A boring at Shapwick Station [ST 424 414] (on Sheet 279), 4 miles south of Wedmore (see Richardson 1928, p. 63 and references), proved a total of 61.5 ft of drift resting on Lias; details of the drift are as follows: peat 16 ft, blue clay with sand towards base 15 ft, sand 15 ft, beds of blue clay, sand and gravel with 2 peat horizons (3.5 to 4 in thick) 15.5 ft. Similar sequences showing between 90 to 110 ft of drift, but without the thin peat beds near the base, have been proved in borings at Bason Bridge (on Weston super Mare Sheet 279) about 2 miles south-west of Mark (Ussher 1914, pl. II). In addition to these deep borings, the succession to a level a little below the main peat bed is known in detail from a number of shallow sections and trial borings (Godwin 1943, p. 221; 1955, pp. 165–180).

Peat

Beds of peat are associated with the alluvial clays of the Somerset moors both south and north-west of the Mendips. Borings, as at Banwell (p. 198) and Shapwick Station (see above) have proved thin peat beds in the lower and middle parts of the alluvial succession but the main bed lies near the top. This bed, which has been recognized throughout the area, rests on a flat surface of estuarine clay at about O.D. and is exposed in the moors south of the Isle of Wedmore, elsewhere it is overlain by some 8 to 14 ft of estuarine clay (Roman Clay). On the landward side of the moors the Roman Clay itself may be overlain by thin peat and clayey peat. G.W.G.

Details

Banwell

Borings in the moors (p. 121) proved peat horizons below the Roman Clay. The small patch of peat, rarely more than 3 ft in thickness, which occurs 0.75 mile north of Sandford, rests on the Roman Clay.

Axe basin

Patches of peat and peaty clay overlie the Roman Clay on many of the moors. Between Cross and Axbridge the peat is locally over 5 ft thick. Farther south, the peat on the moors between Draycott and Bleadney forms one continuous sheet, though the outcrop is subdivided by the thin spreads of fluviatile alluvium which rest upon it. The peat on Draycott Moor averages 3.5 ft in thickness, on Stoke Moor 2 ft, on Westbury Moor 1 to 1.5 ft, and on Knowle Moor 4 ft, with up to 7 ft in the lobe of the moor on the west side of Knowle Hill. The main peat bed probably occurs throughout the Axe basin underlying the estuarine alluvium (Roman Clay). G.W.G., F.B.A.W.

North Moor

Peat and peaty clay, commonly over 4 ft thick, underlies the extensive fluviatile alluvium, and rests upon the Roman Clay and main peat bed. The eastern limits, beneath the fluviatile alluvium, of the two latter deposits are unknown.

Godney Moor to Tealham Moor

These peat moors are broadly divided into an eastern and western part by a line drawn southwards from Panborough. The eastern part, with a surface level of about 22 to 25 ft O.D., is characterized by the presence of 3 to 5 ft of peat and clayey peat overlying in turn the Roman Clay and the main peat bed (see p. 121 above). In the western part, with a surface level of 10 to 13 ft O.D., only the main peat bed, measuring 10 to 16 ft in thickness, is present. Here, the peat has been extensively cut for fuel and is still being exploited, though now mainly for horticultural purposes. The position of the water table in the summer limits the depth of working to about 4 to 5 ft. Different histories of cutting are reflected in the many different levels both in adjacent fields and sometimes even in parts of the same field; probably little virgin peat remains. The low surface level of the peat is due largely to shrinkage consequent on the drying-out of the peat, so that an 'inversion' of the topography has occurred. Originally the Roman Clay must have been banked against the peat instead of, as now, occupying higher ground. Detailed accounts of the peat stratigraphy in this and adjoining areas have been published by Godwin and others (see references p. 126).

Mark Moor

Here the main peat bed is overlain by the Roman Clay which has a surface level of about 20 ft O.D.—some 5 to 10 ft higher than that of the peat moors to the east. G.W.G.

Fluviatile alluvium

Fluviatile alluvium, consisting mainly of red loam, forms relatively narrow strips along the principal rivers and streams. Good sections in the alluvium of the River Chew were seen at the time of the construction of the Chew Valley Reservoir (see pp. 117–8 above). On the moors south of the Mendips the freshwater alluvium is confined to levees bordering the main rivers and drainage channels. Near the Mendips these levees are composed of red loam adjacent to the stream banks, but both laterally and longitudinally (i.e. downstream) they pass into thin spreads of grey-brown gleyed clays which cannot be separately mapped from the underlying grey estuarine clays.

Details

Cheddar

South of Cheddar the place at which the Yeo River originally flowed into Cheddar Moor is marked by a delta fan or broad levee of red loam. The present course of the river is apparently the result of an artificial diversion. G.W.G.

Rodney Stoke

The stream which rises north-east of the village has deposited a levee up to 5 to 6 ft high of red-brown clay extending downstream for the initial three miles of its course through the moors (for a section see under 'calcareous tufa' p. 124). F.B.A.W.

Westbury Moor–Easton

The peat of Westbury and Knowle moors is overlain by a wide belt of thin levee material laid down by the two streams draining westwards from Windmill and Knowle hills. These streams now run in artificially straightened channels with banks heightened by recent dredgings. The broad alluvial tract between Easton and Knowle Hill exhibits typical levees of red loam 2 to 5 ft thick passing outwards into flats of brown and greyish clay 1 to 1.5 ft thick.

North Moor

This area is covered by the alluvium of the River Sheppey and its tributaries. Red loam levees 2 to 8 ft in thickness pass outwards into wide spreads of brown and greyish clay 1 to 2 ft in thickness. The present course of the Sheppey downstream of Hurn Farm is probably artificial and its former course, marked by a well-developed levee of red loam, leads northwards into the Axe basin through the gap east of Panborough.

Westhay Moor

The small area of alluvium on the edge of the map south of Decoy Pool Farm represents the northern edge of the alluvial and lacustrine deposits of the former Meare Pool—the site of the well-known Iron Age Meare lake villages. Detailed sections in these deposits have been published by Godwin (1955, pp. 161–90). Farther west, the River Brue is bordered by a levee of grey-brown clay 1 to 3 ft thick, which, in the area of River House, cannot be differentiated from the underlying estuarine clay. The boundary shown on the map has been inserted for cartographic convenience. G.W.G.

Calcareous tufa

Deposits of tufa are commonly associated with streams and springs throughout the area but are mostly too small to be shown on the one-inch map. At Rodney Stoke, the brook west of the church has laid down an extensive spread of grey-white calcareous tufa. The present stream is cutting through the deposit which is seen to be 8 ft thick in the stream banks west of the church. The following section [ST 476 497] in a rhine-bank 700 yd W. 14° S. of the church was noted (in descending order): red-brown clay (levee) 2.5 ft, tufa 2.5 to 2.75 ft, peat 0.75 to 1 ft, and grey estuarine clay seen to 2.5 ft. F.B.A.W.

Mendip gorges, dry valleys, and drainage

The limestone plateau of the Mendips is dissected by many shallow dry valleys. These valleys are of mature aspect with gentle gradients and sides, while their floors are in a number of cases covered by flat-topped spreads of gravelly drift (seep. 115) above. The valleys are arranged in dendritic drainage patterns and in their lower reaches coalesce to form a relatively small number of steep-sided combes or gorges which descend steeply to the low ground at the foot of the Mendips. In marked contrast to the upper reaches, the gorges and combes bear the marks of youth—narrowness, steep sides with cliffs and scree slopes, and steep uneven valley-floor gradients. These characteristics are seen on the most magnificient scale in the case of the Cheddar Gorge.

Two main theories have been put forward to account for the origin of Cheddar Gorge—and by implication the other Mendip gorges. On the first theory the Gorge is considered to be due mainly to normal subaerial river action; on the second theory it is considered to be the result of the unroofing of a series of caverns which had been excavated by underground streams (Dawkins 1862, p. 121). The role of chemical solution of the limestone by rainwater is given more prominence in the second than in the first explanation. These rival theories have recently been developed in some detail—the cavern collapse theory by Stride and Stride (1947) and the subaerial view by Stanton (1964). While it must be admitted that the collapse theory is plausible when applied to the lower reaches of Cheddar Gorge it breaks down when extended to include the other Mendip gorges or even the upper reaches of Cheddar Gorge itself. In these cases the meandering valleys with their interlocking valley spurs point unmistakably to river erosion while the extensive scree slopes can only have been produced by subaerial weathering. On the other hand the associated features that can definitely be ascribed to roof collapse, e.g. swallets and open caverns, are of very minor proportions. On the subaerial erosion hypothesis the unique scale of Cheddar Gorge is explained by the relatively very large size of the river which excavated it—a deduction which follows from the fact that the catchment area of the valley system that leads into Cheddar Gorge is many times greater than that which leads into the other gorges.

On geomorphogical grounds it would appear that the gorges are younger than the upper reaches of the Mendip valleys which themselves incise the Mendip plateau. The date of formation of the Mendip plateau—a remarkably uniform surface about 800 to 1000 ft above O.D.—is unknown but unlikely to have been much earlier than late Tertiary times. The excavation of the gorges was complete by late Pleistocene times as can be seen from the evidence of the head trains that lead from their outlets (see pp. 112–3 above) and the presence of caves with late Pleistocene faunas in the gorge bottoms e.g. Gough's Cave, Cheddar and Aveline's Hole, Burrington. The evidence therefore accords with the views of Reynolds (1927, p. 173), who suggested that the gorges were cut during 'the Glacial Period' i.e. Pleistocene times when owing to the ground being frozen the surface water was unable to escape underground as it does at the present day.

The Mendip limestone plateau is now drained underground by a complex cave and fissure system in the Carboniferous Limestone and the Dolomitic Conglomerate—the water table being hundreds of feet below the plateau surface. Accounts of the cave systems are given by Balch (1948) and in numerous papers in the Proceedings of the University of Bristol Spelaeological Society, and the journals of other caving societies. Tratman (1963) has recently discussed the development of the drainage and caves of the Burrington area and given details, elucidated by using fluorescein as a tracer dye, of the present-day underground drainage. A recent discussion on the formation of swallets is given by Coleman and Balchin (1959). The waters which pass through the underground drainage system of the Mendips issue as large springs or 'risings' at the foot of the hills, for example St. Andrew's Spring rising in the Palace Garden near Wells Cathedral. In many places these risings are situated at or near the outlets of the Mendip gorges and it is probable that they represent rivers which formerly flowed down the gorges. Thus the spectacular risings at Cheddar close to Gough's Cave and of the River Axe in the Wookey Hole caverns are situated in the lower parts of the Cheddar Gorge and the Ebbor Valley, respectively (see also, p. 172). G.W.G.

References

APSIMON, A. M., DONOVAN, D. T. and TAYLOR, H. 1961. The Stratigraphy and Archaeology of the Late-Glacial and Post-Glacial Deposits at Brean Down, Somerset. Proc. Univ. Bristol Spel. Soc., 9, 67–136.

AVERY, B. W. 1955. The Soils of the Glastonbury District of Somerset. Mem. Soil Surv. Gt. Brit.

BALCH, H. E. 1948. Mendip—its Swallet Caves and Rock Shelters. 2nd edit. Bristol.

BUCKLAND, W. and CONYBEARE, W. D. 1824. Observations on the South-western coal district of England. Trans. Geol. Soc. (2), 1, 210–316.

BULLEID, A. and JACKSON, J. W. 1938. The Burtle Sand Beds of Somerset. Proc. Somerset Arch. Nat. Hist. Soc., 83 for 1937,171–95.

BULLEID, A. and JACKSON, J. W. 1942. Further notes on the Burtle Sand Beds of Somerset. Proc. Somerset Arch. Nat. Hist. Soc., 87 for 1941,111–6.

CLAPHAM, A. R. and GODWIN, H. 1948. Studies of the post-glacial history of British Vegetation. VIII. Swamping surfaces in peats of the Somerset Levels. IX. Prehistoric trackways in the Somerset Levels. Phil. Trans. Roy. Soc. (B), 233,233–73.

COLEMAN, ALICE M. and BALCHIN, W. G. V. 1959. The Origin and Development of Surface Depressions in the Mendip Hills. Proc. Geol. Assoc., 70, 291–309.

DAWKINS, W. BOYD, 1862. On a Hyaena-den at Wookey Hole, near Wells. Quart. J. Geol. Soc., 18,115–25.

DAWKINS, W. BOYD, 1863. On a Hyaena-den at Wookey Hole, near Wells, No. II. Quart. J. Geol. Soc., 19, 260–74.

DONOVAN, D. T. 1954. A Bibliography of the Palaeolithic and Pleistocene Sites of the Mendip, Bath and Bristol Area. Proc. Univ. Bristol Spel. Soc., 7, 23–24.

DONOVAN, D. T. 1955. The Pleistocene Deposits at Gough's Cave, Cheddar, including an Account of Recent Excavations. Proc. Univ. Bristol Spel. Soc., 7, 76–104.

EDELMAN, C. H. 1950. Soils of the Netherlands. Amsterdam.

GODWIN, H. 1943. Coastal Peat Beds of the British Isles and North Sea. J. Ecology, 31, 199–247.

GODWIN, H. 1948. see under CLAPHAM, A. R.

GODWIN, H. 1955. Studies of the post-glacial history of British Vegetation. XIII. The Meare Pool region of the Somerset Level. Phil. Trans. Roy. Soc. (B), 239, 161–190.

MCBURNLEY, C. B. M. 1959. Report on the First Season's Fieldwork on British Upper Palaeolithic Cave Deposits. Proc. Prehistoric Soc., 25, 260–9.

REYNOLDS, S. H. 1927. The Mendips. Geography, 13, 169–76.

RICHARDSON, L. 1928. Wells and Springs of Somerset. Mem. Geol. Surv.

STANTON, W. I. 1964. The Origin of the Mendip Dry Valleys and Gorges (in the press).

STRIDE, R. D. and STRIDE, A. H. 1947. The formation of the Mendip Caves. Part I. Brit. Caver, 16, 63–7.

TRATMAN, E. K. 1955a. Second Report on the Excavations at Sun Hole, Cheddar. The Pleistocene Levels. Proc. Univ. Bristol Spel. Soc., 7, 61–70.

TRATMAN, E. K. 1955b. The Prehistoric Archaeology of the Bristol Region, pp. 147–63 of Bristol and its Adjoining Counties edited by C. M. Maclnnes and W. F. Whittard. Bristol.

TRATMAN, E. K. 1963. The Hydrology of the Burrington Area, Somerset. Proc. Univ. Bristol Spel. Soc., 10, 22–57.

USSHER, W. A. E. 1914. A Geological sketch of Brean Down, and its environs. Somerset Arch. Nat. Hist. Soc., 60, 17–40.

WOODWARD, H. B. 1876. Geology of East Somerset and Bristol Coal-Fields. Mem. Geol. Surv.

WOODWARD, 1908. Appendix VI-Notes on the Indexes of colours issued by the Geological Survey of Great Britain, and on the dates of the earlier published maps of Cornwall, Devon and West Somerset. Sum. Prog. Geol. Surv. for 1907, 158–63.

ZEUNER, F. E. 1959. The Pleistocene Period. 2nd edit. London.

Chapter 9 Structure

Introduction

The rocks in the area of the Wells Sheet have been subjected to earth movements during more than one geological period, as was long ago recognized by Buckland and Conybeare (1824) in a classic paper. These authors not only delineated the major Mendip periclines (pl. 38) but illustrated the intense folding of the Coal Measures in the south part of the Radstock Basin (pl. 37). The original Survey maps (1845) covered the same area in considerably greater detail. Our knowledge of the detailed structure of the Radstock Coal Basin is, however, in great part due to McMurtrie (1869, 1877, 1901) who first described the major faults and understood the form of the overfolding at the southern end of the Basin. The Royal Coal Commission, appointed in 1866, added greatly to the factual information available (Prestwich 1871) and provided the basis of an important work by Anstie (1873). These results were incorporated in the Geological Survey map (Old Series Sheet 19; 1889 Edition) and horizontal sections (103–5, 111–2; 1874–5). Prestwich (1871, pp. 146–66) was the first to appreciate fully the structural and stratigraphical relationship between the Palaeozoic and Mesozoic rocks of the Somerset and Franco-Belgian coalfields. Progress in the elucidation of the Armorican structures of the Mendips did not keep pace with the discoveries in the Coal Basin, where active mining provided a constant supply of fresh geological evidence, and it was not until 1906 that Sibly first recognized major overthrusting in the Mendips at Ebbor Rocks. It was left to F. B. A. Welch to establish in two important papers (1929, 1933) the nature and extent of the Armorican structures by detailed six-inch mapping of the Palaeozoic rocks. Later writers (Moore 1938; Moore and Trueman 1939; Moore in Trueman 1954) have dealt in more general terms with the structure and stratigraphy of the Somerset Coalfield.

The structure in the Mesozoic rocks in this and adjoining areas has been treated in general terms by O. T. Jones (1931, 1950) and by Kellaway and Welch (1948).

The earth movements are now considered in chronological order.

Pre-Upper Old Red Sandstone (Caledonian)

Evidence for these movements is provided by the unconformable junction of the Upper Old Red Sandstone on the Silurian (Upper Llandovery) volcanic rocks in the Beacon Hill area north-east of Shepton Mallet. In that part of the Beacon Hill inlier which falls within the Wells Sheet the rocks are poorly exposed, but where they are seen in adjacent areas to the east there is evidence of angular discordance between the two formations of some 15° to 20°.

Intra-Carboniferous

In the areas bordering the Bristol Channel to the north-east of Sheet 280 the cumulative effects of episodes of folding and erosion during post-Visean and pre-Upper Coal Measures times resulted locally in the production of a major unconformity between the Pennant Sandstone and the Old Red Sandstone (Kellaway and Welch 1948, fig. 8). It is probable that the effects of these movements extended into the north-eastern part of Sheet 280 where the Palaeozoic rocks are largely concealed by younger deposits. In this area the Pennant Series may rest unconformably on older formations including the Carboniferous Limestone Series (Figure 1).

Permo-Carboniferous (Armorican)

The major structures in the Palaeozoic rocks are referable to the Armorican movements (Plate 5). The largest visible structures in the Mendips consist of four periclines with exposed cores of pre-Carboniferous rocks. To the northeast of the Mendips, the Radstock and Pensford basins form two major structural units separated from each other by the Farmborough Fault Belt which probably extends westwards separating the Mendips from Broadfield Down. On the southern side of the Mendips the Cheddar–Wells Thrust Belt separates the Mendip periclines from other periclines, including the Dulcote Pericline, farther to the south. In the Mendips, the long axes of the four periclines have an east-west orientation and comprise from north to south, the Blackdown, North Hill, Pen Hill, and Beacon Hill periclines. These folds are asymmetrical with steeply dipping or overturned and thrust-faulted northern limbs, and more gently inclined southern limbs. Where the Lower Coal Series is present on the northern flanks of the folds, the limestone of the northern limb may be recumbent. The isolated Carboniferous Limestone inliers at Churchill and East Harptree, lying north of the Blackdown and North Hill periclines respectively, are probably remnants of thrust-masses (klippen)resting on Coal Measures (see also pp. 131–3). Major faulting is closely related to the folding, the faults consisting of southerly dipping thrusts and lags (normal faults) aligned roughly parallel to the folds, and cross faults which cut the structures at a high angle and are probably tear faults. In a number of cases the thrusts pass laterally into tear faults. The form of the folds in the Pennant and Upper Coal Series of the Radstock and Pensford basins differ markedly from that seen in the pre-Coal Measure rocks in the Mendip periclinal areas due to the strongly incompetent nature of the intervening Lower Coal Series (Plate 5). In default of any detailed evidence the degree of structural continuity between the two areas must remain a matter of conjecture. It is probable, however, that the major lines of disturbance which bound the periclines on their northern sides are represented in some form or other in the Coal Basin (cf. Welch in Bullerwell 1954, fig. 2). Thus, the major overfold of Pennant and Lower Coal Series bounded by the Southern Overthrust, which forms the southern edge of the Radstock Basin, is represented in the Mendips by the Pen and Beacon Hill periclines. The latter may be regarded as two parts of one major fold separated by the relatively shallow Binegar Syncline and bounded to the north by the Emborough Thrust. Although the Southern Overthrust and the Emborough Thrust have been represented on (Plate 5) as one continuous thrust fault, it is probable that the line of fracturing is represented in the Lower Coal Series as a belt of complex disharmonic disturbance rather than a simple fracture. Farther north, the line of disturbance represented by the Winterfield Fault may correspond to the complex overfolded northern limb of the eastern half of the North Hill Pericline in the Mendips. Farther north again, the Farmborough Fault Belt (Kellaway and Welch 1948, p. 39) probably continues westwards and separates the Blackdown Pericline from Broadfield Down. In the Clutton area this fault belt displaces the top of the Pennant some 2 miles to the east with a total upthrow from the south of more than 3000 ft. Elsewhere in the Coal Basin adjustment to the Armorican pressures has resulted in low-angle thrusts, such as the Radstock Slide, and lags, and steeply inclined normal faults such as the Temple Cloud (east-west) and the Clutton (north-south) faults.

South-west of the Mendip periclinal areas the structure, though much obscured by Triassic rocks, is clearly of a relatively complex nature. The main visible structural units consist of the Cheddar–Wells Thrust Belt and the Dulcote Pericline while small inliers of Carboniferous Limestone west of the main Palaeozoic outcrop indicate the presence of at least one other pericline. The Cheddar–Wells Thrust Belt, which divides the Mendip periclines from those to the south-west, comprises in its exposed parts: the South-Western Overthrust and associated folds, including the Stoke Woods Syncline; the Ebbor Thrust and associated folds; the Ebbor Fault; and the Lyatt Anticline and Fault. The Thrust Belt is most strongly developed in the Cheddar area, where the total upthrow from the south may attain some 3000 ft, but diminishes in importance eastwards from Wells and finally merges into the southern limb of the Beacon Hill Pericline. Between Cheddar and Wells the individual structures are aligned in a general north-west to south-east direction over the greater part of their length.

Analysis of Mendip Armorican tectonics

Kellaway and Welch (1948, p. 8) have pointed out that in the Bristol and Gloucester district structural lines of weakness, or axes, have reacted several times to deep-seated earth-movements, and of these the Mendip Axis (1948, fig. 3) is one of the most important. The term 'Axis' is here used in general terms and does not refer to a single continuous fold line, or the plane of a fold, but to a broad ridge-like area of uplift, sometimes associated with faulting.

The most striking feature of the Mendip Axis is its sickle-shape, and this, Jukes-Browne (1911, p. 180) considered to be related to the structure of the basement rocks. He suggested that a southward prolongation of the Malvern Axis would cross the Mendip Axis at its point of maximum concavity to the north and where "the oldest rocks [Silurian] in the Mendip arch come to the surface, as they would be likely to do if the Armorican fold was here overriding an older Malvernian fold". Welch (1933, pp. 46–9) in a detailed analysis of Mendip tectonics suggested that the resistance offered by a Coal Measures syncline of Malvernoid trend the forerunner of the Radstock Basin—to the northwardly directed Armorican earth-movements, gave rise to periclines arranged in échelon on two sides of the basin fold. Moore and Trueman (1939, p. 63) suggested that this older syncline may have been intra-Westphalian in age.

A third possible explanation for the sickle-shape of the Mendips is that this is a manifestation of an original sedimentary feature. The thickness variations in the Carboniferous Limestone Series (see pp. 15, 21, and (Figure 2)) show that the growth of an embryonic Radstock Basin and Mendip Axis was in progress during Lower Carboniferous times. It is probable, though the evidence is not conclusive, that the growth of these features continued in Millstone Grit and Coal Measures times. In effect, the shape of the Radstock Basin reflects the original form of the deeper part of the sedimentary basin in which the Carboniferous rocks were laid down. Subsequent earth-movements—possibly late Carboniferous in age—may have accentuated the form of the Radstock Basin which was folded along an axis of Malvernoid (North–South) trend (cf. Welch 1933; Moore and Trueman 1937). Finally the Armorican movements superimposed over the whole area the strong east-west structural pattern which characterizes these movements throughout Southern Britain.

These latter movements resulted at first in the development of the main Mendip periclines along the line of the Mendip Axis concurrently with the closure on its southern side of the Radstock Basin. Continued movement resulted in the steepening of the northern limbs of the folds and culminated in overfolding and thrusting. The fact that structures in the periclinal areas associated with this later phase of compression are in continuity with structures in the Cheddar–Wells Thrust Belt indicates that both belong to the same phase of movement. Thus, the Emborough and the Thrupe faults of the periclinal areas are continued westwards in the Thrust Belt as the South-Western Overthrust and the Ebbor Fault, respectively. Further, it is noteworthy how the east-west trending minor folds and associated thrusts in the western part of the North Hill Pericline bend to a north-westerly direction near the South-Western Overthrust see (Plate 5). The apparently anomalous north-westerly trend of the structures in the Thrust Belt may be related to the shape of the Mendip axis. Thus, where the latter takes on an east-west orientation, south-west of Wells, the Thrust Belt assumes the same direction. Within the Thrust Belt, the Ebbor Thrust post-dates the formation of the Stoke Woods Syncline, though the time interval was probably very short.

Normal faults associated with the Armorican folds and thrusts, are now seen to be of considerable importance in the Mendips just as they are in the Radstock Coal Basin. The two most important examples are here named the Ebbor and Rodney Stoke faults, the former is relatively well exposed while the latter is mainly inferred, beneath the Triassic cover, from borehole evidence. These faults bound the north side of fault blocks which have a downward and southward movement relative and apparently complementary to, the northward and upward movement of thrust passes in front of them (see (Figure 10)) i.e. they act as lag faults. Other normal (dip) faults are situated in the central areas of the main periclinal folds and may be due to tension generated in the crestal regions of these folds e.g. Lamb Leer, Maesbury and Priddy faults. Evidence for adjustment during the bending of a periclinal axis is best seen at the western end of the Beacon Hill Pericline where faulting apparently represents a response to compression on the north side and tension on the south side (see p. 138).

Formational competency in Armorican tectonics

The form of the structures in the Mendips and adjacent areas was strongly influenced by the degrees of competence of the formations involved in them. The Portishead Beds (Upper Old Red Sandstone), and the Pennant Sandstone appear to have been strong competent formations, whereas the Lower Limestone Shale and the Lower Coal Series were markedly incompetent. The Carboniferous Limestone on the whole behaved as a competent formation, though it is highly folded and sheared in areas where the strata are recumbent. Isolated thrust masses, or klippen, of inverted Carboniferous Limestone resting on Coal Measures to the north of the Beacon Hill Pericline have long been known at Vobster and Luckington (Frome Sheet 281; see Welch 1933, and references). They represent the remnants, isolated by erosion, of the recumbent northern limb of the Beacon Hill Pericline. The inlier of Carboniferous Limestone to the north of the Blackdown Pericline at Churchill (Welch 1932, p. 395) and that north of the North Hill Pericline at East Harptree are probably of a similar nature see (Figure 9) and (Figure 11). Welch (1933, pp. 45–7 and pl. IV) suggested that the degree of inversion attained by the northern limb of the Beacon Hill Pericline was in part due to large-scale detachment of the Carboniferous Limestone from the underlying Old Red Sandstone by sliding on the soft Lower Limestone Shale. In the case of the Blackdown and North Hill periclines there is little evidence that extensive movement of the Carboniferous Limestone over the Lower Limestone Shale has taken place and it is probable, therefore, that the present position of the klippen to the north of these periclines was attained by the shearing and thinning of the Carboniferous Limestone in the northern limbs and crests of the folds (see (Figure 9)).

The incompetence of the Lower Coal Series not only explains the relative independence of structures in the Carboniferous Limestone and Pennant Sandstone (see pp. 128–30) but has also strongly influenced the form of the structures in the Carboniferous Limestone. Thus, the extent of the overfolding in the Carboniferous Limestone depends on the amount of the Lower Coal Series available for displacement and where this is small, thrusting may take the place of overfolding. This is clearly brought out by a comparison of attitude of the limestone in the northern limbs of the North and Pen Hill periclines ((Figure 9); (Plate 5)). In both cases the northern limbs are inverted but where erosion has led to the removal of the Lower Coal Series in the complementary synclines to the north the amount of overturning in the limestone is small (10° to 20°). On the other hand, where appreciable Coal Measures is still preserved in the deeper part of the adjacent synclinal folds the amount of overturning in the limestone of the northern limbs of the periclines is much greater (40° or more). In the later stages of the development of the North Hill Overfold there is evidence to suggest (see p. 137) that its western side eventually met with resistance from the Carboniferous Limestone of the Blackdown Pericline and the northwards movement of this part of the pericline was then continued by thrusting in the Old Red Sandstone (Miner's Arms Thrust). The recumbent northern limb of the Blackdown Pericline at Churchill is analogous to that of the North Hill Pericline (compare (Figure 9); inset, and (Figure 11)). It is probable that once appreciable inversion of the Carboniferous Limestone had been attained gravitational gliding on the Lower Coal Series played an important role in the development of these structures.

The Lower Limestone Shale due to its incompetent nature has also influenced the development of some Armorican structures in the Mendips; its effect has, however, been less marked than that due to the Lower Coal Series. This is probably because of its smaller thickness and the presence of numerous limestone bands in parts of the succession. The conformability of the structure in the Carboniferous Limestone with that in the underlying Old Red Sandstone over the greater part of the Mendip periclinal areas indicates that the Lower Limestone Shale has here behaved as a relatively competent bed. In certain cases in these areas, however, variations in its thickness, e.g. near the Rookham Fault (p. 206), probably indicate incompetence although in other cases rupture by thrusting has occurred e.g. along parts of the inverted limbs of the North and Pen Hill periclines. Within the Cheddar–Wells Thrust Belt, however, partial decollement of the Carboniferous Limestone seems to have taken place (Figure 10) and large thrust masses have moved on the Lower Limestone Shale. In general it appears that the thrusting in which the Lower Limestone Shale exerted a controlling influence is part of a later more intense phase of tectonic adjustment than that represented by the doming of the major Mendip periclines.

Sub-Mesozoic Armorican structure south of the Mendips

Many workers (e.g. Prestwich 1871; Woodward 1876, pp. 46–9; McMurtrie 1911, fig. 2) have suggested that the Radstock Coal Basin has a structural homologue to the south of the Mendips. This suggestion is open to serious doubt as not only is it based on the assumption that the Mendips form a relatively simple anticlinal axis, but also that the low-lying nature of the ground south of the Mendips was because of its having 'been excavated in strata more yielding than those forming the Mendip Hills' (Woodward 1876, p. 49). Consideration of the Rhaetic Structure Contour Map (Figure 14), and the Triassic Base Contour Map (Plate 3), clearly shows that both the present elevation of the Mendips and the depression of the Central Somerset Basin are due largely to Triassic and post-Triassic movements. The most westerly and southerly outcrops of Carboniferous Limestone suggest the presence, under the moors to the west, of at least two more periclines similar to those of the Mendips (see also pp. 141–3); a suggestion which is in accord with the geophysical evidence (pp. 152–4), while farther to the west, in the deeper part of the Central Somerset Basin, the geophysical evidence appears to preclude the presence of a thick Coal Measures sequence, similar to that of the Radstock Basin (see p. 155). G.W.G.

Details

Blackdown Pericline

Along the northern edge of the pericline the dips average about 60° to 70° to the north, but decrease to 45° in the Old Red Sandstone near the centre of the fold. On the southern limb the dips are about 45° in the axial region of the fold, and fall to about 20° along the southern edge of the pericline. In plan the periclinal axis is curved and convex to the north with the culmination of the fold situated in the middle of the arc. Here at the centre the Carboniferous Limestone is strongly inverted and the isolated mass of recumbent limestone at Windmill Hill, to the north of the Mendips, is thought to rest on Coal Measures. It is probable that the exceptionally wide outcrop of the Burrington Oolite is due to faulting as shown in (Figure 11) (cf. the Clifton Down Limestone in the North Hill Overfold, (Figure 9); inset).

The western nose of the pericline is buckled into two anticlines separated by the east-west Bleadon Thrust Fault (see p. 23). The fault seen north-west of Compton Bishop may be a south-easterly prolongation of the Bleadon Thrust. In the Old Red Sandstone the eastern nose of the pericline is a simple anticline pitching eastwards at 13° to 14°, but farther east, in the Carboniferous Limestone, this is replaced by a series of fairly gentle folds pitching at about 7° to 10°. At the eastern edge of the limestone outcrop the pitch of the folds increases to 25° or more. The nose area is cut by two large mineralized faults, the more northerly, the Lamb Leer Fault, has an east-west trend and is a normal fault with a steep dip to the south. It is exposed in Lamb Leer Cavern where it downthrows 200 ft to the south. At its western end, the Lamb Leer Fault divides into three branches, the best exposure being the northern branch along a line of old workings [ST 5384 5520] 1100 yd east-south-east of Fernhill Farm. Here the fault is associated with a belt of brecciated and dolomitized limestone some 20 yd in width. The mineralized fault which runs southeastwards from Nordrach towards the Castle of Comfort Inn appears to be a continuation of the Stock Hill Fault. Exposures at intervals along the line of the fault show extensive shattering and re-crystallization of the strata.

North Hill Pericline

The pericline is divided into three parts by the northerly trending Stock Hill and Biddle faults, each part having a distinctive structural pattern. In the western part, the Carboniferous Limestone shows two well-marked subsidiary folds trending east-west over much of their length but becoming aligned with the South-Western Overthrust (p. 131) at the western extremity of the pericline. The Priddy Fault cuts through the area in a general east-west and north-east to south-west direction with a downthrow to the south. If the Priddy Fault, seen 100 yd south of Priddy church, is connected with the faulted ground present at depth in Swildon's Cavern (Stanton 1960) it follows that the Priddy Fault is a normal fault with a high dip to the south. The virtual absence of any lateral displacement of the axis of the Stoke Woods Syncline is evidence that the displacement along the fault is mainly vertical. G.W.G.

About 1.25 miles to the south-west of Priddy church the line of the Priddy Fault is marked by wide veins of calcite and on its western side the Hotwells Limestone, which normally has a north-west to south-east strike, swings to a direction parallel with the fault. On its eastern side, the Priddy Fault downthrows about 500 ft to the south-east, and displaces the South-Western Overthrust. F.B. A. W.

In the northern limb of the pericline, for about a mile westwards from the Stock Hill Fault, all the beds above the Black Rock Limestone are vertical or inverted with dips of 62° to 80° to the south and south-south-west. It is probable that this inversion is related to a thrust (Kingdown Thrust of Welch, 1932, p. 389) hidden beneath Triassic rocks to the north, and running westwards into the belt of disturbed ground about 150 yd south of King Down Farm. The clearest exposure of this disturbed ground is in crags to the west of the road [ST 507 539], where strongly jointed and sheared calcite-mudstones in the Clifton Down Limestone are folded into a syncline in which the northern limb dips at about 20° to the south and the southern limb is vertical.

Between the Stock Hill and Biddle faults the southern limb of the pericline, showing southerly dips of about 20° to 30°, is bounded on the south by the Emborough Thrust. The northern limb, by contrast, is for the greater part strongly inverted between the Stock Hill and Eaker faults. This overfold is best exposed at Niver Hill between 1 and 2 miles to the east of the Stock Hill Fault, where, with increasing degree of inversion the southerly dips steadily decrease northwards from 80° to 30°. The inlier of northerly striking Clifton Down Limestone lying 0.5 mile to the north of the main limestone outcrop is probably a klippe resting on Coal Measures ((Figure 9), and inset, see also p. 133). The northerly strike of the rocks in this inlier may be quite local and produced by gravitational gliding. In the main limestone mass, the outcrop of the upper calcite-mudstone beds of the Clifton Down Limestone is twice its expected width apparently because of strike faulting. Adjacent to the Stock Hill Fault, the Carboniferous Limestone outcrop south of the Castle of Comfort Inn and the evidence from a borehole (Castle of Comfort Borehole, (Plate 5)) at the Inn itself (see p. 203) show that while the strike and dips remain the same as those of the limestone of the Niver Hill area farther east, the outcrops of the individual formations lie some 300 yd to the south, while an important fault, the Miners' Arms Thrust, separates the Black Rock Limestone from the Old Red Sandstone. This thrust which has a stratigraphical downthrow to the north of 750 ft or more, is not represented west of the Stock Hill Fault. An explanation of these differences in the structure of the North Hill Overfold in the Castle of Comfort and Niver Hill areas, may be as follows. At Niver Hill the northwards movement of the overfold over and into the northerly flanking Coal Measures was unrestricted throughout its development. In the Castle of Comfort area the initial development of the overfold was similar to that at Niver Hill to the east but eventually the overfolded Carboniferous Limestone met with resistance from the limestone of the Blackdown Pericline to the north and the resultant stresses were relieved when movement took place along the line of the Miners' Arms Thrust. It is probable that lateral adjustment between the eastern and western parts of the overfold was achieved by tear faulting (cf. the Eaker Fault) along a line now represented by the deep Triassic-filled valley 0.5 mile east of the Castle of Comfort Inn.

The eastern part of the North Hill Pericline is defined by the Eaker and Biddle faults, and is a simple anticlinal structure with an easterly pitch bounded on the south side by the Emborough Thrust. The part of the Biddle Fault lying north of its junction with the Eaker Fault is post-Liassic in age with a downthrow to the east of 100 ft (p. 147); the south part is largely Armorican with a stratigraphical down-throw to the east of some 400 to 500 ft. The marked difference in structure on either side of the Eaker Fault suggests that it is a tear fault. G.W.G.

Pen Hill Pericline

This pericline has a vertical or overturned northern limb with southerly dips ranging from 55° to 85°, and is bounded on the north by the Em-borough Thrust. For about a mile eastwards from Hill Grove the Lower Limestone Shale of the northern limb has been cut out by a thrust fault running sub-parallel to the Emborough Thrust. North of Pen Hill, the Emborough Thrust appears to bifurcate westwards giving rise to the South-Western Overthrust and the Rookham Fault. The structure between these faults is described in a later section (p. 138). Westwards from its junction with the Emborough Thrust the Rookham Fault bends from an east-west to a north-south direction. Presumably the east-west part of the fault is like the Emborough Thrust, a low angle thrust, while the north-south part has a high dip to the west as proved by a borehole at Rookham (see pp. 205–7) and is therefore probably a tear fault. The fault can be traced for some 250 yd south of the Rookham Borehole beyond which it is concealed by Mesozoic rocks. It would appear however that the Rookham Fault does not cut the Ebbor Fault which is partly exposed to the south and much of the movement of the Rookham Fault may have been taken up in the Lower Limestone Shale of the southern limb of the pericline.

The change from the inversion of the northern limb of the pericline to the normal order of strata seen in the southern limb is very abrupt and can be seen in the Emborough quarries (p. 42) where it is accompanied by minor tear faulting. Between these quarries and the northern limb of the Beacon Hill Pericline, 1.5 miles to the south, the Carboniferous Limestone is folded in the broad sweep of the Binegar Syncline. In the synclinal area between the Pen Hill and Beacon Hill periclines the north-south trending Slab House Fault, appears to be a tear fault for the structures on either side are dissimilar; thus an east-west fault, on the east side of the Slab House Fault, which cuts out most of the Lower Limestone Shale outcrop is not represented on the west side. The position and alignment of the latter fault near the axis of the Pen Hill Pericline suggests that it may be a normal fault analogous to the Maesbury Fault etc. (see p. 131). West of the Slab House Fault, the southern limb of the pericline shows gently corrugated strata with general southerly dips ranging from 10° to 30°.

The Biddle Fault has a stratigraphical throw eastwards of about 400 ft and is seen as a zone of calcite veining 6 ft wide in the bottom of the lane (Pease Close Lane) [ST 5697 4738] 900 yd north-north-west of the Wells Mental Hospital chapel. Southwards of this point it is covered by Liassic limestones and it may end against the Ebbor Fault since there is no evidence for its prolongation into the Lyatt Hill area. G.W.G., D.R.A.P., F.B.A.W.

Beacon Hill Pericline

This structure extends eastwards into the Frome (281) Sheet and has a total length of at least 12 miles of which only the western third lies on the Wells (280) Sheet. The anticlinal axis follows the east-west strike of the Silurian rocks in the core of the fold and then turns to a more northerly direction west of the Silurian outcrop. The dips, which in the western nose area of the pericline range from about 15° to 25°, increase to about 50° near the eastern margin of the Wells Sheet in the northern limb and about 40° in the southern limb. The western nose of the pericline, corrugated by minor folds with a general east-west trend, is cut by the Maesbury and Thrupe faults with a similar trend, and with southerly downthrows of about 250 ft and 400 ft respectively. The former, together with an associated branch fault, is exposed in the approach cutting to a disused quarry [ST 597 475], see also p. 45. Here the fault planes dip southwards at 65° and 68°, respectively, and the drag features show these are both normal faults (Plate 4B). About 0.5 mile to the west, the structure becomes complex and near the intersection of the Maesbury and Slab House faults, small inliers of Burrington Oolite have been introduced.

On the northern limb of the pericline, north of Warren Farm, two north-east to south-west trending faults with a downthrow to the west cut the Old Red Sandstone but apparently die out in the Lower Limestone Shale. Evidence of minor folding in the Lower Limestone Shale is provided by the variable width of the outcrop of this formation adjacent to the faults. Farther east, the Oakhill Fault has the same trend but extends northwards into the Coal Measures. It will be noted that in the western part of the pericline the faults have a radial arrangement and probably represent planes of adjustment accompanying the northwards bending of the periclinal axis. D.R.A.P.

South side of the Mendips

The exposed Armorican structures south of the Mendip Hills lie mainly within the Cheddar–Wells Thrust Belt (Figure 1). The larger and more completely exposed structures are described individually below, while the remainder, largely concealed by the Triassic, are treated as a group.

Stoke Woods Syncline

The Stoke Woods Syncline pitches to the south-east and consists of a series of shallow folds lying between and aligned parallel with the South-Western Overthrust and the Rodney Stoke Fault (see p. 141). At the north-western end of the syncline, the isolated inlier of Carboniferous Limestone north of Bradley Cross forms part of the southern limb of the syncline, though its position indicates that it is faulted (see (Plate 5)). The interpretation of the structure at this end of the syncline is complicated by the presence of an inlier of reddish sub-quartzitic sandstone 0.5 mile east of Lyde Farm (see also, p. 12). This was shown on the Old Series 1-inch map (Sheet 19) as Old Red Sandstone but Lloyd Morgan (1890, pp. 178–9) was 'on the whole . inclined to regard this as Millstone Grit'. Welch (1929, p. 63) explained the outcrop as a 'window' in the sole of the South-Western Overthrust and its presence at the surface as being due to the latter having a rolling keel, the softer limestones riding over the relatively harder 'Millstone Grit'. The petrological evidence however, suggests that these rocks are of Old Red Sandstone age and this view is in accord with the structural setting and with the proximity of basal Black Rock Limestone 120 yd to the north-east (see (Plate 5), and (Figure 10)A). East of the Priddy Fault, due to the extensive covering of the Triassic, only the northern limb of the Stoke Woods Syncline is seen, while a mile to the south-east the southern limb is cut off by the Ebbor Thrust. It is possible that the Stoke Woods Syncline, east of the Ebbor Thrust, is represented by a syncline (see p. 140) north of the Lyatt Anticline.

South-Western Overthrust

This thrust branches off the Emborough Thrust north of Pen Hill, and after following a west-south-west direction for about a mile continues along a general north-westerly line before passing under Dolomite Conglomerate 1 mile east of Cheddar church. North-westwards of here the outcrops of Hotwells Limestone to the north show that the thrust must swing towards the south-west (Plate 5); the change in direction probably being comparable with that at the western end of the Rookham Fault (see p. 137). The stratigraphical downthrow of the thrust is to the north and increases from about 300 to 400 ft at its eastern end, to some 2000 ft or so, north-east of Draycott, and at its north-western end, where it splits into several branches, to a total of about 3000 ft. The Overthrust is displaced by both the Priddy and Stock Hill faults.

Ebbor Thrust

The maximum development of this thrust is at Ebbor Rocks where Black Rock Limestone is thrust against Lower Coal Series giving a stratigraphical downthrow to the north-east of about 2000 ft (Figure 5). The position of the thrust plane appears to have been determined by the Lower Limestone Shale ((Figure 10)c and p. 134 above). The south-easterly continuation of the thrust is seen in the Lower Milton area (Figure 12) where it veers to a southerly direction before finally disappearing beneath the Trias about 0.33 mile south-west of Milton Lodge. The north-westerly continuation from Ebbor Rocks of the Ebbor Thrust is covered by Dolomitic Conglomerate; but it may be a 'mirror image' of the eastern side (see (Plate 5)).

Ebbor Fault

The western end of this fault is seen in the Ebbor Rocks area whence it passes north-westwards into a steep monoclinal fold (Figure 5). It is a normal fault with its maximum downthrow of about 1000 ft to the south developed between Lower Milton and Knapp Hill, about 1 mile east-north-east of Wells. It may be accompanied by well-marked terminal drag. At Lower Milton, the structure between the northernmost branch of the Ebbor Fault and the Ebbor Thrust is complex and consists of several fault-slices (Figure 12). G.W.G.

At Knapp Hill, the Black Rock Limestone is bent down sharply against the fault and abuts against gently folded Clifton Down Limestone. Eastwards the fault passes under the Triassic-filled depression, south of the Mental Hospital, emerging south of Chilcote Manor where Clifton Down Limestone is thrown down against disturbed Burrington Oolite, the vertical displacement being small. Farther east, the Thrupe Fault probably represents the eastward continuation of the Ebbor Fault. D.R.A.P.

North and north-east of Wells, between the Ebbor Thrust and Fault the Carboniferous rocks are folded into a syncline all of which, except the peripheral parts (Stoberry Park, and north of the Lyatt Fault), is covered by Triassic rocks. The central parts of the syncline may include Millstone Grit and Coal Measures (Plate 5). G.W.G.

Lyatt Anticline

This anticline, which is aligned in a general east-west direction, appears from beneath the Triassic east of Wells and extends eastwards for about 3 miles before merging into the southern limb of the Beacon Hill Pericline. To the north the anticline is bounded by the Lyatt Fault against which its northern limb is vertical or overturned while the southern limb has dips of about 30° to 40°. The inversion of the strata south of the Lyatt Fault suggests thrusting from the south, yet the apparent vertical downthrow of some 100 ft to the south is in a direction contrary to that normally expected. This anomaly may be due to horizontal movement along the fault plane. In Dinder Wood a north-north-easterly trending fault separates the overfolded western part of the anticline from the more gently folded eastern part in which dips are of the order of 20° to 25°. The fault is clearly a dextral tear fault showing a negligible vertical displacement and a horizontal displacement of about 400 ft. East of this fault the continuation of the Lyatt Fault is seen 0.5 mile north of Staceys. D.R.A.P.

The highest beds of the southern limb of the Lyatt Anticline are hidden beneath the Triassic but may include Millstone Grit or Coal Measures (Figure 13).

South-west of the Mendips

Because of the fragmentary nature of the available evidence any conclusions regarding the nature of the Armorican structures south of the main Palaeozoic outcrop must be regarded as tentative.

Two boreholes (Richardson 1928, pp. 119–21) drilled through the Triassic into the underlying Carboniferous rocks at Rodney Stoke are of considerable importance.

In the Rodney Stoke Borehole [ST 4874 5012], 630 yd E. 28° N. of the church (see (Plate 5)) a steeply inclined and disturbed sequence of shales and limestones was proved to a depth of 400 ft, and interpreted as Millstone Grit or Coal Measures faulted against 'Carboniferous Limestone [Series of] Tournaisian [age]' (presumably Lower Limestone Shale). In the Main Spring Boring 200 yd to the north-north-west shattered limestone queried as of 'C₂ age' was recorded. The latter record suggests Burrington Oolite though Welch (1929, p. 64) implied an 'upper Z age' for these rocks i.e. Black Rock Limestone. In default of more detailed supporting evidence the identification of these Carboniferous Limestone horizons must remain open to doubt. Clearly, however, a major fault with a southerly downthrow of some 2000 ft —the Rodney Stoke Fault (p. 131) must intervene between the Rodney Stoke Borehole and the nearest limestone outcrop to the north which is Black Rock Limestone forming the southern limb of the Stoke Woods Syncline ((Figure 10)B). If the Main Spring Boring is in the Black Rock Limestone the latter probably forms part of the Stoke Woods Syncline and therefore lies north of the Rodney Stoke Fault (see (Plate 5), and footnote on p. 12).

The small inliers of Carboniferous Limestone ranging from Black Rock Limestone to Clifton Down Limestone west of Rodney Stoke and Westbury-sub-Mendip appear to form the northern limb of a large anticlinal structure. The two inliers 0.75 mile south of Easton probably represent a low horizon in the Black Rock Limestone; in the northernmost the beds are bent in a gentle anticline, whilst the southernmost shows a southerly dip. It is possible that the two series of inliers form part of one large structure (see (Plate 5)). This structure bears no obvious relationship to those on the southern side of the Mendips from which it is separated by a deep Triassic-filled depression (see (Plate 3)). Triassic erosion appear to have selected a line of weakness caused by the presence of a strong thrust (or thrusts) and by a belt of soft Coal Measures, as seen in the Rodney Stoke Borehole.

In the area south of Dinder there are several inliers which form part of a large buried structure described as the Dulcote Pericline. The form of this pericline is clearly shown by the results of an aero-magnetic survey, which strongly suggest the presence of buried Silurian volcanic rocks in its core (see pp. 158–9, and (Figure 18)). From south to north, the inliers expose the southern limb of the Dulcote Pericline, the inverted thrust-faulted northern limb of the pericline, and finally a thrust mass lying to the north of the pericline (Figure 9) and (Figure 13), (Plate 5)). The southernmost thrust mass, consisting of Black Rock Limestone, is seen in two inliers, respectively mile south and 1.25 mile south-east of Dinder, which extend onto Sheet 296. In these the strata are gently folded with southerly dips of about 10^° to 25°. The Triassic-filled depression to the north of these inlier marks the site of a strong east-west trending thrust-fault with an estimated throw of at least 1000 ft (Figure 13). To the north of this thrust, the Dulcote Hill–Church Hill inlier shows an inverted succession ranging from the Burrington Oolite up to the Millstone Grit dipping at 35 to 80° to the south. The strata are cut both by small strike (thrusts) and dip (tear) faults, well seen in numerous quarries. The east-west structures of this inlier and of that to the south are displaced by a large north-north-easterly trending fault the—Corrington Fault—which is probably a sinistral tear fault with a maximum horizontal displacement of 600 ft. The inlier (Cliff Wood Thrust Slice) in the small wood (Cliff Wood) about 0.25 mile north of Church Hill may represent part of the succession of the southern limb of the Lyatt Anticline, repeated by thrust faulting (Figure 13). The exposed rocks, consisting of the upper half of the Clifton Down Limestone, strike east-west, with dips ranging from 48° to the south to vertical. The inlier is bounded on the north by a steep-sided Triassic-filled valley which is presumed to be on the site of an east-west thrust-fault. This thrust may have originated in the Lower Limestone Shale in a manner analagous to the Ebbor Thrust and the South-Western Overthrust (see (Figure 9) and (Figure 10). The eastward continuation of all these structures is hidden by Mesozoic rocks except for an inlier of Clifton Down Limestone and Hotwells Limestone 0.5 mile south-east of Croscombe which shows an inverted succession with southerly dips of 40° to 55°. It is possible that this may form part of the inverted limb of the Dulcote Pericline which has been separated by tear faulting (cp. the Corrington Fault) from the part seen at Church Hill, 0.25 mile to the south-west. G.W.G.

North of the Mendip Hills

The pre-Triassic rocks of this area consist entirely of Coal Measures and form part of the Radstock and Pensford Basins. A detailed account of their structure will be given in a forthcoming memoir dealing with the Bristol and Somerset Coalfields.

Post-Armorican

Since Permo-Carboniferous times the district has been affected by a series of earth movements. The combined effects of intra-Triassic and intra-Jurassic warping movements are considerable, and are shown by wide variations in thickness, unconformities and non-sequences in the Keuper, Lias and Inferior Oolite. Much of the movement occurred in post-Bajocian times as many of the structures seen in the Lias extend also into the Inferior Oolite (and Fuller's Earth). By analogy with the Yeovil–Bridport area to the south (Wilson and others 1958, pp. 181–3) post-Middle Jurassic movements may have taken place in two main episodes, namely intra-Cretaceous (pre-Albian) and post-Cretaceous (probably Miocene).

The combined effects of post-Keuper movements are graphically summarized in the form of a structure contour map of the base of the Rhaetic (see (Figure 14)) the main features being the Mendip Axial Fold and the north-eastern part of the Central Somerset Basin. In comparison with dips in the Palaeozoic rocks those in the Mesozoic formations are usually gentle; nevertheless it will be seen that the base of the Rhaetic in the Central Somerset Basin has been depressed some 2500 ft below its base level in the Mendips. Faulting with a dominant east-west (Armoricanoid) trend is widespread but there are other important faults with an approximately north-south trend. In general, the latter end against the east-west faults but this is not invariably the case, and the close inter-relation of the faults to one another suggests that they may belong to one major fracture system. In a number of instances in the Mendips and in the mining areas of the Radstock Basin it can be shown that the post-Triassic faults have been located along Armorican lines of weakness. Other post-Armorican faults and folds, however, are apparently independent of the older structures. In the case of large structural units, such as the Mendip Axis, the Armorican and post-Armorican features appear to show only a very general relationship. The post-Armorican faults appear to be normal faults and most of them have throws of less than 100 ft. However, the Mudgley Fault in the south-western corner of the Sheet has a downthrow of some 200 to 350 ft to the south.

It will be noted that most of the post-Triassic faults shown on the map appear to be concentrated where Tea Green Marl and later rocks are present. This may be due to the difficulty of identifying faults in the Keuper Marl. Thus the large excavations made in Keuper Marl for the Chew Valley and Axbridge reservoirs exposed a number of small faults which might otherwise have remained undetected owing to the lack of marker bands.

Details

North of the Mendips

In the Banwell area, the Mesozoic rocks are folded into a shallow elongated basin (Stonebridge Syncline) with an east-west axis which may extend eastwards into the syncline which runs along the Yeo valley. The grey marls (Butcombe Sandstone horizon) between Honey Hall and Bean Bridge dip southwards forming part of the northern limb of this syncline. G.W.G.

In the area between Chew Stoke and Chelwood it is locally impossible to say whether mapped contacts between the basal Triassic sandstone and the Coal Measures are erosional fault scarps of Permo-Triassic age or post-Triassic features. The Bromley Horst (see (Plate 5)) is one such complex structure which has been accentuated by strong post-Liassic faulting, particularly along its southern side.

Many post-Liassic faults occur north of the Mendips, the most important being the east-west Gravel Hill Fault which has a southerly downthrow of about 130 ft at the north-eastern end of Gravel Hill. The Woodford Hill Fault runs parallel to, and about 0.25 mile south of the eastern part of the Gravel Hill Fault, and attains a maximum southerly displacement of about 50 ft at Woodford Hill. South of Woodford Hill, another east-west trending fault appears to continue eastwards to form the southern flank of Knowle Hill where it probably passes into a monoclinal fold before dying out. The precise relationship between these faults and the many disturbances noted in and around the cut-off trench of the Chew Stoke Reservoir is obscure. The probable disposition of the main faults in the vicinity of the cut-off trench and the sites of the principal exploratory bore and trial holes is shown on (Figure 15). Of the trial holes only No. 2 proved relatively undisturbed Keuper Marl, all the remaining pits proved disturbed strata. The principal features of the trial pits and boreholes are as follows. Trial Hole No. 1 [ST 5710 6156], height above O.D. 149 ft, Soil and alluvium to 12.5 ft, gravel to 14 ft 2 in, Keuper Marl and Sandstone to 36 ft, Coal Measures to 52 ft (dip at 49 ft, about 17° to north-west). Trial Hole No. 2, height above O.D. 178 ft, Keuper Marl to 63.5 ft. Trial Hole No. 3, height above O.D. 150 ft, Soil and alluvium to 9 ft, highly disturbed Trias and Coal Measures to 20.25 ft. Trial Hole No. 4, height above O.D. 180 ft, Soil to 1 ft, Keuper Marl to 57.5 ft (dip about 50° southerly in lowest 14 ft). Borehole 'A', height above O.D. 148 ft, Soil and alluvium to 14 ft, Keuper Marl with basal sandstone to 46 ft, Coal Measures to 106 ft (for details of the latter, see Moore 1940). Borehole height above O.D. 149 ft, Soil and alluvium to 10 ft, sand and gravel to 15 ft, Keuper Marl with sandstone beds to 44 ft, Coal Measures to 54 ft. Borehole 'C', height above O.D. 182 ft, Keuper Marl with beds of sandstone to 104 ft. Borehole 'E', height above O.D. 149 ft, Soil to 1 ft, alluvium to 14 ft, Keuper Marl to 54 ft. Borehole 'D' [ST 5698 6111] lies 470 yd S. 12° W. of Borehole 'E', height above O.D. 150 ft, Soil and alluvium to 6 ft, sand and gravel to 9 ft, Keuper Marl with beds of sandstone to 154 ft. It will be noted that the Coal Measures in Trial Hole No. 3 was encountered at a markedly higher level than in the other holes. This anomaly may be explicable in terms of diapiric movement of the Coal Measures between convergent fault planes. G.A.K.

In the Clutton area, the Marksbury Plain Fault together with the parallel fault some 200 yd to the south appear to lie along the general line of the Farmborough Fault Belt of Armorican age. The Clutton Fault, which bounds the area of coal-workings on the western side may indicate post-Triassic movement along a line of Armorican disturbance (Bullerwell 1954, p. 51).

In the Temple Cloud–Paulton–Farrington Gurney district three easterly trending faults in the Mesozoic rocks probably represent post-Triassic movements along the lines of earlier faults proved in colliery workings. These faults are: Timsbury Fault with southwards throw of about 400 ft in the Coal Measures (the '70 Fathom' or Temple Cloud Fault), and 90 ft in the Mesozoic rocks; Paulton Fault with southwards throw of 100 ft in the Coal Measures, and 40 ft in the Mesozoic strata; and the Winterfield Fault throwing northwards 70 ft in the Coal Measures, and 25 ft in the Mesozoic beds.

Between Chewton Mendip and Chilcompton the most noticeable structure shown on the map is the Chewton Fault. In a ridge, known as Chew Down, extending eastwards from Chewton Mendip for about 0.25mile, the Lower Lias limestone is folded into a narrow anticline some 120 yd wide and with a height of some 70 ft; the dips of the south limb in places reach as much as 45°. The northern limb, in which the dips are of the order of 20°, is truncated along its northern edge by the Chewton Fault which throws downwards to the north. A borehole [ST 6075 5313] 1230 yd east of Chewton Mendip church, proves that the base of the Blue Lias has been thrown down to 110 ft below the surface. To the north and north-west of Chewton Mendip, a southerly-throwing fault, extending westwards to near Hook's Hill, is developed along the southern limb of the anticline. At a distance of 1.25 miles east-south-east of Chewton Mendip church a second smaller anticlinal fold with Rhaetic clay exposed in the core is developed on the south side of the Chewton Fault. North of Old Down the Chewton Fault bifurcates, one branch turning eastwards through Chilcompton, the other maintaining a south-easterly trend through Old Down to Blacker's Hill. The Chilcompton branch lies parallel to, and a little north of, a large east-west disturbance which limits the northward workings of the New Rock Colliery; still farther east (on one-inch Sheet 281) this post-Triassic fault runs south of and parallel to the Armorican structure termed the 'Southern Overthrust' (see p. 54). The Old Down branch of the Chewton Fault trends in a direction parallel to the strike of the Coal Measures but no effects have been observed in the underlying colliery workings. In the railway cutting at Old Down the fault introduces the remarkable folded strip of Rhaetic and Lias beds figured by Woodward (1876, pl. iv, p. 79). F.B.A.W.

Mendips

Owing to the removal of the Rhaetic and later rocks over much of the Mendips the pattern of post-Rhaetic folding is only imperfectly known—it may follow that seen in the Castle of Comfort–Chewton Mendip area where the Mesozoic rocks are flexed into gentle north-west to south-east trending folds with the Rhaetic base lying between the limits of 850 and 1000 ft O.D. (Green 1958, fig. 2, p. 76). In order to complete the Rhaetic Structure Contour Map (Figure 14) over the Mendip area where the Jurassic rocks have overstepped the Rhaetic onto the Palaeozoic rocks the base of the younger rocks along the plane of the unconformity has been plotted in continuation of the Rhaetic base contours. This involves some degree of error but not sufficient to invalidate the general structural picture. The Biddle Fault shows post-Liassic movement of about 100 ft downthrow to the east in its northern part, but this diminishes to 50 ft or less, south-east of Red Quarr and disappears entirely farther south (see also p. 137). G.W.G.

The Downside Fault shows evidence of both Armorican and post-Liassic movement. The post-Liassic downthrow increases from about 20 ft to the south-east in the railway cutting at Windsor Hill to 50 ft or more near Darshill. East of the railway cutting evidence of more severe (pre-Liassic) disturbance can be seen in a large quarry (Downside Quarry) in Black Rock Limestone, where the rocks are sharply folded locally into south-west pitching flexures. The Armorican fault downthrows to the south-east. North-eastwards the post-Liassic throw apparently dies out but it will be noted that the fault is in line with a prominent north-east to south-west feature and col in the Old Red Sandstone ridge west of Beacon Hill.

In the Shepton Mallet–Doulting area the Bodden Fault is a prominent east–west line of fracture downthrowing the Lias some 40 to 50 ft to the south. In a roadside section, 450 yd N. of Shepton Mallet church, originally pictured by De la Beche (1846, p. 278), the fault plane was shown as nearly vertical with a second fault, a little to the south, downthrowing in the same direction. The faults are no longer exposed. The combined effects of faulting and folding has depressed the Lias into a basin elongated in an east-west direction in this northern part of Shepton Mallet. As the Inferior Oolite at Doulting shows no sign of synclinal folding the basin must either die out eastwards or it must be assumed that it continues eastwards as a pre-Inferior Oolite feature. D.R.A.P.

South of the Mendips

Structurally this area is very approximately divided in two by a composite fault line consisting of the western half of the Weare Fault, the Wedmore Fault and the eastern half of the Mudgeley Fault. To the south and west of this line lies the deeper part of the Central Somerset Basin, to the north and east lies a broad area of folds showing a general alignment with the edge of the Mendips. Between Doulting and Dinder this faulting appears to be related to the underlying east-west Armorican structures.

The Isle of Wedmore is terminated to the north by the east-west trending Weare Fault, which has a variable southward throw, due to folding on either side, of from 50 to 100 ft. It probably continues under the estuarine alluvium for some distance both to the east and west.

The Wedmore Fault joins the Weare Fault in the north to the Mudgley Fault in the south of the Isle of Wedmore. Its downthrow, which is to the east, increases from 25 ft in the north to 75 ft in the south.

The Mudgley Fault, which forms the southern margin of the Isle of Wedmore is a composite structure consisting of 3 faults arranged en echelon with a combined maximum throw to the south of about 300 ft, or more, between Mudgeley Hill and Bagley. East of Panborough, the Mudgley Fault bends from a general east-west to an east-south-east to south-easterly direction, and the former east-west line of faulting is prolonged by a series of branch faults associated with folding. The main fault line extends along the south side of Callow Hill, throwing Belemnite Marls down against the Tea Green Marl and later rocks to the north-east. From here it continues through Fenny Castle Hill and finally bends back to its original east-west trend at a point (on Glastonbury Sheet 296) about 0.25 mile south of Melsbury Farm. Thence, it continues just south of the Sheet margin into the faulted ground southeast and east of Coxley, where it throws Lower Lias Clay against Keuper Marl.

Between Coxley and Stump Cross about 4 miles to the east, the Mesozoic rocks are folded in a gentle elongated dome centred on Twinhills Wood (on the boundary of Sheets 280 and 296) about 0.5 mile south of Wellesley Farm. This structure is cut by a number of east-west trending faults with throws of from 25 ft to 75 ft which extend onto Sheet 296. A little to the north of this area and immediately south of Dulcote and Church hills an east-west fault showing a post-Liassic downthrow of 30 to 50 ft to the south is probably on the line of a strong Armorican thrust (Figure 13). G.W.G.

References

ANSTIE, J. 1873. The Coal Fields of Gloucestershire and Somersetshire, and their resources. London.

BUCKLAND, W. and CONYBEARE, W. D. 1824. Observations on the south-western coal district of England. Trans. Geol. Soc. (2), 1, 210–316.

BULLERWELL, W. 1954. A Gravimeter Survey of the Ston Easton–Harptree District, East Somerset. Bull Geol. Surv. Gt. Brit., No. 6,36–56.

DE LA BECHE, H. T. 1846. On the Formation of the Rocks of South Wales and South-western England. Mem. Geol. Surv. 1.

GREEN, G. W. 1958. The Central Mendip Lead-Zinc Orefield. Bull. Geol. Surv. Gt. Brit., No. 14,70–90.

JONES, O. T. 1931. Some Episodes in the Geological History of the Bristol Channel Region. Rep. Brit. Assoc. for 1930,57–82.

JONES, O. T. 1950. The Structural History of England and Wales. Rep. 18th Int. Geol. Congress, pt. 1,216–29.

JUKES-BROWNE, A. J. 1911. The Building of the British Isles, 3rd. edit. London.

KELLAWAY, G. A. and WELCH, F. B. A. 1948. Bristol and Gloucester District, 2nd edit. British Regional Geology, Geol. Surv.

MCMURTRIE, J. 1869. On the Faults and Contortions of the Somersetshire Coal Field. Proc. Bath Nat. Hist. Soc. and Antiq. Field Club, 1. 127–47.

MCMURTRIE, J. 1877. On certain isolated Areas of Mountain Limestone at Luckington and Vobster in the County of Somerset. Proc. Bath Nat. Hist. Soc. and Antiq. Field Club, 3, 287–300.

MCMURTRIE, J. 1901. The Geological Features of the Somerset and Bristol Coalfield with special reference to the Geology of the Somerset Basin. Trans. Inst. Mining Eng., 20, 306–35.

MCMURTRIE, J. 1911. On a Boring at Puriton, near Bridgewater, in search of Coal south of the Mendip Hills. Proc. Somerset Arch. Nat. Hist. Soc., 57, 25–53.

MOORE, L. R. 1938. The sequence and structure of the Radstock Basin. Proc. Bristol Nat. Soc. (4), 8, 267–305.

MOORE, L. R. 1940. Recent boring reaching the Coal Measures at Chew Stoke, near Bristol. Proc. Bristol Nat. Soc. (4), 9, 66–8.

MOORE, L. R. and TRUEMAN, A. E. 1937. The Coal Measures of Bristol and Somer- set. Quart. J. Geol. Soc., 93, 195–240.

MOORE, L. R. and TRUEMAN, A. E. 1939. The structure of the Bristol and Somerset Coalfields. Proc. Geol. Assoc., 50, 46–67.

MORGAN, C. LLOYD, 1890. Mendip Notes. Proc. Bristol Nat. Soc. (3), 6, 169–82.

PRESTWICH, J. 1871. In Report of the Royal Coal Commission, 1, London.

RICHARDSON, L. 1928. Wells and Springs of Somerset. Mem. Geol. Surv.

SIBLY, T. F. 1906. On the Carboniferous Limestone (Avonian) of the Mendip Area (Somerset), with especial reference to the Palaeontological Sequence. Quart. J. Geol. Soc., 62, 324–80.

STANTON, W. I. 1960. Plan of Swildon's Hole Cavern, Priddy, Mendip. (For private circulation only.)

TRUEMAN, A. E. (Editor). 1954. The Coalfields of Great Britain. London.

WELCH, F. B. A. 1929. The Geological Structure of the Central Mendips. Quart. J. Geol. Soc., 85, 45–76.

WELCH, F. B. A. 1932. The Geological Structure of the Blackdown Pericline, Proc. Bristol Nat. Soc. (4), 7, 388–96.

WELCH, F. B. A. 1933. The Geological Structure of the Eastern Mendips. Quart. J. Geol. Soc., 89, 14–52.

WILSON, V., WELCH, F. B. A., ROBBIE, J. A. and GREEN, G. W. 1958. Geology of the Country around Bridport and Yeovil. Mem. Geol. Surv.

WOODWARD, H. B. 1876. Geology of East Somerset and Bristol Coal-Fields. Mem. Geol. Surv.

Chapter 10 Geophysical investigations

This account presents results from geophysical investigations in the Wells district and surrounding areas. Ground geophysical work carried out in the region over a number of years, comprises regional gravity and magnetic surveys, detailed gravity and magnetic traverses in areas of special interest, and some reconnaissance shallow seismic investigations and electrical resistivity measurements. Geophysical measurements have also been made in the air, during an aeromagnetic survey of Southern England sponsored by the Geological Survey and flown in 1957.

Results from these geophysical surveys have been of considerable help in elucidating the geological structure. In particular, they give information about periclinal structures, largely concealed by Mesozoic strata, south of the Mendip Hills.

Gravity surveys

The first gravity survey in the general region of the present work was commenced in 1945 by the Anglo-Iranian Oil Company (Falcon and Tarrant 1950). This survey, which later covered all south-central England, extended to the north and east margins of the district under review. In 1949, a regional survey of the Bristol and Somerset Coalfields (Cook and Thirlaway 1952) extended over the Mendip Hills and five gravity traverses were established across the Hills. Also in 1949, the Geological Survey carried out a gravity survey in a part of the Radstock Coal Basin in the vicinity of Ston Easton and Harptree designed to elucidate structures within the Coal Measures (Bullerwell 1954). Gravity survey of the south-eastern part of the Wells Sheet, up to the line of the Mendip Hills, was completed by the Geological Survey in 1959 as part of a regional survey in Somerset and Dorset.

To complete the Geological Survey's regional gravity coverage of the area around Wells those parts not included in earlier work were completed during 1962 with an average of two gravity meter readings per square mile. Nearly 1000 gravity stations in the area under review are arranged areally with an average of three per square mile. In addition sixteen traverses comprising about 1000 stations were surveyed in localities of particular interest. A uniform station spacing of 100 yards was used on all traverses.

Reduction of data

Readings of gravity have been referred to a value of —108.70 milligals at a station in Shepton Mallet first occupied by the Anglo-Iranian Oil Company. Connexions to stations in Radstock and Glastonbury, used during the 1949 and 1959 surveys respectively, facilitated a co-ordination of all the gravity data for the area.

Formation densities have been taken into account in the reduction of the gravity data to sea-level and the Bouguer anomalies represent departures from theoretical gravity at sea-level, given by the International Gravity Formula of 1930. Terrain corrections have been applied out to Zone M (Hammer 1939) or about 13.5 miles; they are important only in the vicinity of the Mendip Hills, where the highest corrections exceed three milligals.

Rock densities used in the reduction of the gravity data, together with those used by Bullerwell (1954) and Cook and Thirlaway (1952), are given in the following table:

Reynolds (1907) reported a density of 2.72 g/cm³ obtained on a specimen of Silurian andesite.

Results and interpretation

The Bouguer anomaly map (Figure 16) is a compilation of Geological Survey work mentioned above. On this map, the Mendip Hills and a strip of country to the south correspond with a zone of relatively high gravity anomaly, with peak values at about 4 milligals, lying usually between two and four miles to the south of the exposed Mendip fold axes. The gravity field falls gently northwards across the Mendip periclines, though each pericline produces a secondary disturbance on this gradient.

The Bouguer anomaly map and traverses of Cook and Thirlaway reveal a similar displacement of maximum gravity values from Mendip fold axes, though the gravity field south of the Mendip Hills is evaluated by them only in the broadest terms. Cook and Thirlaway (1952, p. 270) attribute the continued rise of the anomaly to the south of the Hills to a southward prolongation of Mendip structures.

To the south-west, values fall gently to about –2 milligals at the south-west corner of the Sheet. Towards the north-east, values fall more rapidly, reaching —8 milligals at the north-east corner of the Sheet, near to the centre of the Radstock Coal Basin.

Gravity gradients reach a maximum value of about four milligals per mile along the entire northern front of the Mendip Hills, and locally south of Wells.

The positive gravity anomaly associated with the Mendip Hills continues eastward into the area of the Frome (281) Sheet included in the Anglo-Iranian Oil Company's survey of south-central England. On the Bouguer anomaly map given by Falcon and Tarrant (1950, p1. ix) the gravity anomaly turns north-eastwards, along the direction of a related magnetic anomaly (op. cit. p1. x), to pass near to Westbury and die out near Devizes.

Gravity anomalies in the Mendip Hills

There is some evidence in the regional data to suggest that variations in density between the Silurian andesite, Old Red Sandstone and Carboniferous Limestone bring about important, though small, gravity anomalies over the Mendip Hills. To confirm this, two of the four exposed periclines were traversed and Bouguer anomaly profiles were produced assuming a uniform density of 2.65 g/cm³ down to sea-level. This type of reduction leaves in the Bouguer anomaly the gravity effects of density changes above sea level, except in the special case where a number of structural effects counterbalance. The profile over North Hill reveals positive anomalies with peak values over the Black Rock Limestone and an axial low anomaly of over one milligal clearly associated with the exposed core of Old Red Sandstone, the Lower Limestone Shale having an effect intermediate between that of the massive Carboniferous Limestone and the Old Red Sandstone. The profile across Beacon Hill ((Figure 17), traverse 9) traverses the fold at a position where no Silurian rocks crop out and there is a bilaterally symmetrical anomaly pattern similar to that obtained over North Hill. Here, however, in the better-developed central low anomaly, there is a very small axial positive anomaly of less than 0.25 milligal which may be a reflexion of Silurian rocks at no great depth. A short traverse ((Figure 17)a; 10) over the outcropping Silurian succession south of Stoke St. Michael, less than 0.75 mile east of the above traverse, demonstrates an associated positive anomaly of one milligal. The anomalies on these three traverses show clearly that the individual Palaeozoic formations must be considered separately when interpreting the local detail of the gravity field.

Gravity anomalies south of the Mendip Hills

Nine gravity traverses ((Figure 17)a; 1–9) were surveyed south of the Mendip Hills in an attempt to detect and trace the position of anticlinal axes shown on (Plate 5). These axes are inferred from surface mapping to pass through Clewer and Wookey and south of Dulcote Hill, though there is some positional uncertainty as the Palaeozoic rocks are largely concealed by Mesozoic strata. A typical example of the anomaly pattern encountered is shown in (Figure 17)b (traverse 5). The anomaly profiles, computed using a uniform density of 2.65 g/cm³, are generally complex, though almost all show a striking bilateral symmetry similar to that observed over the exposed Mendip periclines. If it be assumed that the general anomaly pattern is due to an anticline comprising strata of different densities, it is possible to ascribe differences in form of the anomaly between profiles to differences in the structural level of the fold. On this basis, the centre of symmetry of the anomaly will represent approximately (or if the fold is symmetrical, precisely) the position of the fold axis. The coufse of this symmetry axis is shown on (Figure 17)a. West of Wells the positions of the fold axis deduced from mapping and the anomaly symmetry axis lie close together. South of Dulcote Hill the symmetry axis lies about 0.25 mile south of the inferred axis of the Dulcote Pericline. Gravity traverses farther to the south-east have indicated the continuation of a similar anomaly pattern and hence the probable continuation in that direction of anticlinal folding.

An analysis of the anomaly profiles was carried out in an attempt to support the above conclusions and to obtain more detailed information about the structure of the area. Three types of anomaly pattern obtained along traverse lines south of the Mendip Hills were: (1) a simple gravity peak, (2) a gravity trough flanked on each side by a gravity peak, (3) a central gravity peak succeeded outwards in each direction by a trough followed by a peak. A chart devised by Jung (Heiland 1940, p. 154) was used to calculate the gravity effect at various levels of a symmetrical fold with 45° slopes involving Palaeozoic rocks under a thin Mesozoic cover, considered down to a plane at the estimated depth of the base of the Silurian andesites. Because of the imposed depth limitation, such an approximate structural model will not produce the complete anomaly at any level but will indicate plausible variations in the shape of the anomaly as the erosional level of the fold changes.

Using an Old Red Sandstone thickness of 1300 ft and density of 2.60 g/cm³ and a Silurian thickness of 1000 ft and density of 2.70 g/cm³ it can be shown that an anticline with a crest of thick unbreached Carboniferous Limestone of density 2.65 g/cm³ will produce a single gravity peak whose shape is influenced very little by the effect of the predicted density contrasts within the Palaeozoic at depth. As the Carboniferous Limestone thins the first significant effect on the gravity field is a reduction in the gravity peak due to the superimposition of a negative secondary anomaly representing the effect of the Old Red Sandstone predominating over the effect of the deeper Silurian. This effect will reach its maximum as Old Red Sandstone reaches the top of the fold, when there will be a well-marked gravity trough associated with it. The gravity traverses over North Hill and Beacon Hill mentioned above confirm this interpretation. Thereafter, as the axial Old Red Sandstone thins, the positive effect of the higher-density Silurian andesite becomes increasingly important, so that the axial negative anomaly due to the Old Red Sandstone is modified and eventually contains a positive anomaly due to the andesite. The short traverse south of Stoke St. Michael mentioned above confirms the presence of this positive anomaly.

The precise form of the anomaly will depend upon the thickness of Mesozoic cover, the density contrasts involved, and the stratigraphical detail of the Carboniferous Limestone, Old Red Sandstone and Silurian succession. The effect of pre-Silurian strata also is significant. Since these factors cannot all be assessed accurately, the shape of the anomaly cannot be used to establish the nature of strata immediately below the Mesozoic cover or to estimate their depth or thickness with any certainty. Nevertheless the above analysis shows that changes in form of the anomaly between profiles can be used to determine qualitatively the nature of the structural changes which they reflect. Further, the close relationship between gravity anomaly features and individual Palaeozoic units means that the strike of the gravity anomalies will also represent approximately the strike of the Palaeozoic strata.

Correlations of anomaly features on adjacent traverses are indicated in (Figure 17)a. The central symmetry axis is interpreted above (p. 152) as the approximate representation of the fold axis. In three localities gravity peaks and troughs, when followed from west to east, first diverge from and then converge back on to the symmetry axis, and this anomaly pattern is taken to indicate the existence of three periclines south of the Mendip Hills centred: (i) 1.5 miles south of Shepton Mallet; (ii) 1.25 miles south-south-east of Wells; (iii) 1.25 miles south-south-west of Westbury-sub-Mendip. Each of the pericline anomaly patterns has an axial positive anomaly indicative of a high density core and the two easterly, near Shepton Mallet and Wells, probably involve Silurian andesite for they are associated with magnetic anomalies attributed to the presence of these rocks (pp. 158–9). The anomaly pattern south of Westbury-sub-Mendip has no associated magnetic anomaly and the axial gravity anomaly probably reflects the presence of high density non-magnetic Silurian or pre-Silurian strata in the centre of the pericline. The anomaly pattern near Wells indicates a fold axis position 0.25 mile south of the axis of the Dulcote Pericline inferred from surface outcrops. This displacement is explicable by disturbance in depth due to thrust faulting in the northern limb of the fold (Figure 9). Geological evidence indicates that the gravity low in the anomaly of this pericline is probably due in part to Lower Limestone Shale. The degree of agreement between gravity and magnetic indications of position is discussed below (p. 158) after a description of the magnetic survey results.

Certain of the gravity anomalies encountered on traverses 1–9 are due to other causes listed below. Erosional features across the postulated folds such as Triassic-filled valleys (see (Plate 3)) produce gravity effects of the same order of magnitude as those ascribed to Palaeozoic structure. A traverse from Shortwood [ST 513 468] to Lodge Hill [ST 491 482] across one of these valleys showed a negative anomaly of about one milligal attributable to variation in Triassic thickness. If the anomaly be due solely to the Triassic resting unconformably on Carboniferous Limestone the maximum thickness of the Triassic is calculated to be slightly more than 300 feet. In addition a negative gravity anomaly axis with an amplitude of about one milligal, running east-west through Dinder, is interpreted as due to a similar thickness of Triassic filling a structural depression between Dulcote Hill and Lyatt. Alongthe mainly NE–SW direction of the gravity traverses the effect of Triassic-filled valleys with a similar direction (Plate 3) is probably only to depress the level of the anomaly. If there is any differential erosion of the concealed Palaeozoic surface filled in by Mesozoic or later sediments it is likely to produce gravity effects which reinforce the effects due solely to structure.

Immediately south of the southern margin of the Wells Sheet, east of National Grid line 50E, the gravity gradient southwards from the Mendip Axis steepens locally to four milligals per mile. If this results from structure involving the Palaeozoic surface, it represents a southward down-faulting or sharp down-folding of 1400 feet; it may, in part, be due to a Triassic erosional feature. On the Wells Sheet, the gravity fall southwards from the Mendip Axis is less sharp, rarely exceeding one milligal per mile. In the south-west corner of the Sheet, the lowest gravity values occur over Mark Moor, about three miles south-west of Wedmore, where the level of the gravity field is five milligals below that over the Mendip Axis. The thickness of the Mesozoic rocks inferred from the contour map of the base of the Triassic (Plate 3) would be expected to produce a larger negative anomaly than that observed, and high-density Palaeozoic strata are required to reconcile the Mesozoic thickness and the observed anomaly. The absence of a large negative anomaly similar to that over the Radstock Basin, in spite of the much thicker Mesozoic succession, precludes the existence of a similar thick Coal Measures succession under the area south-west of Wedmore.

Gravity anomalies north of the Mendip Hills

The steepest gravity gradients, of about four milligals per mile, follow the northern front of the Mendip Hills and are a reflexion of the steeply dipping northern limestone limbs of the Mendip periclines. Bullerwell (1954, p. 46) suggested that a northerly thickening of the Triassic also contributes to the gravity gradient in this area. The asymmetry of gravity profiles over the northern front of the Mendip Hills, suggestive of an overturning of the northern limbs of the folds, is attributed by Cook and Thirlaway (1952, p. 270) to structures in the Coal Measures immediately north of the Hills.

No further gravity work has been carried out in the north-east part of the Wells Sheet, previously covered by a survey of the Ston Easton and Harptree districts (Bullerwell 1954). This survey helped to elucidate structure within the Radstock Coal Basin.

Magnetic surveys

A vertical force magnetic survey of south-central England by the Anglo-Iranian Oil Company (Falcon and Tarrant 1950), carried out at the same time as their gravity survey mentioned above, extended to the eastern margin of the area under review.

In 1957 the Geological Survey commissioned and supervised an aeromagnetic survey of southern England which covered the Mendip Hills area. Flight lines were in a north-south direction at 2 kilometre intervals along selected National Grid lines, and east-west tie lines were flown every ten kilometres. The survey was flown at a constant barometric height of 1800 ft above sea-level.

Aeromagnetic total force anomalies shown in (Figure 18) represent deviations from a geomagnetic field for Great Britain calculated from observations at an open network of magnetic stations throughout the country.

During 1962 the Geological Survey carried out a proton magnetometer survey on the ground in the Wells Sheet and surrounding areas. Nearly 1400 total magnetic force readings were made, of which about 600 were arranged on an areal basis of two readings per square mile and the remainder were along eighteen traverses in areas of particular interest. Along the traverses the station spacing was reduced occasionally to five yards in an endeavour to detect the detail of the anomalies under investigation.

Results and interpretation

Over most of the area, north of National Grid line 50N and west of line 50E, the magnetic field is entirely negative with respect to the accepted geomagnetic field datum. On the aeromagnetic map the most clearly defined feature within this area is a broad negative anomaly of about 40 gammas, ten miles wide and trending north-east to south-west across the Wells (280) Sheet. To the north-west, the field rises by about 40 gammas to a culmination crossing the western margin of the Sheet, near Loxton. On the ground this anomaly is less clearly defined, due to local disturbances. There is no surface evidence for the cause of these broad regional anomalies.

Magnetic traverses across Beacon Hill Pericline, traversing outcropping volcanic rocks, confirm that the large magnetic anomalies in the south-east part of the survey area reflect the presence of Silurian andesitic lavas and tuffs. The largest anomaly in this area is associated with Beacon Hill where the amplitude of the ground anomaly exceeds 200 gammas. Falcon and Tarrant (1950, pl. X) show the vertical component anomaly associated with Beacon Hill to extend across the Frome (281) Sheet with an east-north-easterly, becoming north-easterly, trend and to die out near Devizes.

Magnetic anomalies near Launcherley and Prestleigh

South and south-west of Beacon Hill are two other anomalies, of smaller amplitude but similar areal extent (see (Figure 18)), centred on Launcherley and Prestleigh. Against the normal background field level of about —40 to —50 gammas the aeromagnetic anomalies are entirely positive, though the ground magnetic results indicate that each has a small negative anomaly on its northern margin.

On the ground the Launcherley anomaly has an amplitude of 110 gammas, the Prestleigh anomaly 80 gammas; in the air these reduce to about 80 and 60 gammas respectively. Differences in shape and areal extent between air and ground results are small and there is a very small displacement of the positions of the peak values. The magnetic anomalies centred on Launcherley and Prestleigh together have a length of 15 miles.

The indications provided by these anomalies, of structure involving the up-folding of the Silurian andesite, lend support to the gravity interpretation (pp. 154–5) of the local existence of concealed and partially concealed periclines. The Launcherley anomaly, covering the area of the Dulcote Pericline, is assumed to be due to andesite in the core of this fold. Magnetic peak values do not coincide precisely with the pericline centres indicated by the gravity interpretation. However, aeromagnetic profiles over Beacon Hill and a ground magnetic profile east of the Downhead Fault show that the magnetic peak is displaced to the south of the structural axis when the survey level is remote from the andesite. Aeromagnetic profiles show the magnetic peak over Beacon Hill offset up to 0.25 mile south of the andesite outcrop.

Since the peak value of the Launcherley magnetic anomaly lies 0.25 mile south of the symmetry axis of the gravity anomaly SSE of Wells and the minor axes of the gravity and magnetic anomalies coincide, the magnetic evidence closely supports the gravity interpretation of the position of this pericline. Another reason for southward displacement from the axis deduced from outcrops is the thrusting which is known to affect the northern limb of the pericline and which is assumed to be responsible for the offset of the gravity axis.

The peak value of the Prestleigh anomaly, on the other hand, lies on the major axis of the gravity anomaly south of Shepton Mallet, offset one mile to the east of the probable centre. This disagreement may be due to complexity of structure within the pericline, or to an eastward thickening of the volcanic succession.

Two ground magnetic traverses were established over the Launcherley anomaly. One, across the centre of the anomaly, shows it to be broad, generally smooth and to have an amplitude of +90 gammas and a width of about three miles. Irregularities with an amplitude of 20 gammas interrupt the anomaly near to the peak values. The other profile, one mile farther east, is important in showing a negative anomaly of 50 gammas, 600 yards across, immediately to the north of the peak values. A very similar anomaly occurs in a profile across the Prestleigh anomaly. These two negative anomalies are not represented in the aeromagnetic contour map.

Irregularities of between 100 and 200 gammas superimposed on the large anomaly observed over the andesites of Beacon Hill may represent reversed magnetization in part of the volcanic succession. Of three remanent magnetization measurements on andesite samples from Moons Hill (Sum. Prog. Geol. Surv. for 1955, p. 68), one indicated reversed magnetization. It is suggested tentatively that the negative anomalies near Launcherley and Prestleigh may be due to the same cause. Using a formula derived by Smith (1961, p. 400; equation 2.5), the maximum depth to the top of the magnetic source is calculated to be slightly more than 500 ft for the Launcherley anomaly and about 400 ft for the Prestleigh anomaly. Aeromagnetic anomaly profiles running north-south through the area were analysed by the process of downward continuation, using a modification of a method of Henderson (1960) which enabled it to be used on profiles. The field was computed at a level 300 feet above the local ground level to determine whether the broad single peak observed on both profiles at 1800 ft O.D. would resolve into two components. In each case the anomaly remained as a single peak so that the aeromagnetic profiles do not reveal the negative anomaly observed at ground level.

Considerable irregularity observed in the ground profile near Prestleigh and less marked irregularities in the two profiles near Launcherley may be due to andesite at shallow depth. Support for this view is given by the increase in irregularity towards the positions of peak magnetic values, in the vicinity of the fold axes. On this criterion alone, the Prestleigh anomaly, because of its greater irregularity, should have the shallower source and its lower peak values and amplitude would have to be accounted for by a thinner andesite succession.

Magnetic anomalies over Beacon Hill

Detailed traverses over Beacon Hill were designed to interpret Silurian structure in the core of the pericline. Shortage of data on the magnetic properties of the andesites and tuffs, the probability of a significant remanence vector in strongly folded rocks superimposed on the induced magnetization, together with the irregular shape of the magnetic body, preclude quantitative interpretation. Nevertheless it was considered that the correlation of magnetic features between adjacent traverses may give information on the strike and stratigraphy of the Silurian strata. Though, in fact, no worthwhile structural information was gained, the profiles do reveal the existence of magnetic 'marker' horizons and indicate that local structural problems might be elucidated by intensive magnetic surveying with very close traverse spacing.

A ground magnetic traverse over Pen Hill and Prior's Hill, across the Pen Hill Pericline, shows a positive anomaly of 30 gammas, displaced 0.25 mile to the south-east of the mapped axis, which may be due to the presence of a thin volcanic succession. The abrupt diminution of the aeromagnetic anomalies to the north-east suggests a rapid thinning of the volcanic sequence in that direction. A fault running through Wells towards Ashwick, with a downthrow to the north, could produce the same effect though a large fault is considered, on geological grounds, to be unlikely.

Seismic measurements

Some reconnaissance work with hammer seismic equipments was designed to determine the utility of the method in solving local structural problems and also to obtain representative seismic velocities.

An FS-2 equipment, demonstrated by Huntings Surveys Ltd., successfully detected the Carboniferous Limestone platform, immediately below the Bajocian unconformity, at a depth of 57 ft in a field north of Vallis Vale, near Hapsford on the Frome (281) Sheet. In addition the equipment was used to measure the following seismic velocities in the Wells and Frome Sheets.

Seismic velocities of Jurassic limestones, though very variable, are rarely as low as the value quoted above for the Inferior Oolite. The value is probably that of a weathered or fragmented limestone and is not likely to be more than locally significant. The velocities given indicate the plausibility of locating, by seismic methods, buried Palaeozoic structure but show that difficulty would be experienced in distinguishing between Carboniferous Limestone and Silurian andesite.

Electrical measurements

An electrical resistivity survey near the bottom of Burrington Combe was carried out by Palmer (1958) to investigate the effects of inclined boundary planes. Results of expanding probes (Palmer 1958, pp. 6–7) show a clear distinction between resistivity values for wet and dry Carboniferous Limestone and Dolomitic Conglomerate, though no values are given.

The following average resistivity values have been compiled for local rock groups. Information is partly from electrical logs of Ston Easton No. 1 Borehole, obtained for the National Coal Board by the Société de Prospection Électrique.

Conclusions

Results of the gravity and magnetic surveys contribute to a knowledge of the detailed structure of the Mendip Hills, giving information in particular on the nature of folding south of the main fold axes. Precise information on depths and structural details cannot be obtained without the additional control of borehole information or seismic investigations.

A complete analysis of the negative gravity anomaly crossing the south-west corner of the Wells Sheet, associated with a thick Mesozoic succession, cannot be undertaken without more regional gravity data south of the area under review and a consideration of the regional setting of the anomaly. M.B.

References

BULLERWELL, W. 1954. A Gravimeter Survey of the Ston Easton–Harptree District, East Somerset. Bull. Geol. Surv. Gt. Brit., No. 6, pp. 36–56.

COOK, A. H. and THIRLAWAY, H. I. S. 1952. A gravimeter survey in the Bristol and Somerset coalfields. Quart. J. Geol. Soc.,107 [for 1951], 255–86.

FALCON, N. L. and TARRANT, L. H. 1950. The gravitational and magnetic exploration of parts of the Mesozoic-covered areas of south-central England. Quart. J. Geol. Soc., 106, 141–70.

HAMMER, S. 1939. Terrain corrections for gravity stations. Geophysics, 4, 184–94.

HEILAND, C. A. 1940. Geophysical Exploration. 1st edit. New York.

HENDERSON, R. G. 1960. A comprehensive system of automatic computation in magnetic and gravity interpretation. Geophysics, 25, 569–86.

PALMER, L. S. 1958. Examples of geolectric surveys. Pap. Instn. Elect. Eng., No. 2791.

REYNOLDS, S. H. 1907. A Silurian inlier in the Eastern Mendips. Quart. J. Geol. Soc., 63, 217–40.

SMITH, R. A. 1961. Some theorems concerning local magnetic anomalies. Geophysical Prospecting, 9, 399–410.

Chapter 11 Economic geology

Introduction

A general survey of mining and quarrying in this and adjoining areas has been made by Loupekine (1956), while the Quarry Managers' Journal compiles up-to-date lists of quarries. Coal, though an important economic product on the Wells Sheet, is to be considered in detail in a forthcoming memoir dealing with the Bristol and Somerset Coalfields as a whole.

Roadstone and aggregates

The quarrying of roadstone is one of the most important industries in the district under discussion. In times past the range of roadmaking materials was derived from a wide variety of geological formations and supplies from numerous small local sources. Now production is mainly concentrated in a small number of large quarries in the Carboniferous Limestone Series, all the main limestone formations of which are used. The limestone is sold as hardcore, roadstone and concrete aggregates, tarmacadam and agricultural dust. North-east of Shepton Mallet, Silurian andesitic lavas are also an important source of roadstone and concrete aggregates, but the present exploited areas lie beyond the eastern margin of the Wells (280) Sheet.

Mesozoic limestones are used on a small scale for hardcore or crushed for the manufacture of artificial blocks. In the former category are quarries at Clapton (Blue and White Lias) and Stowey, near Bishop Sutton (Blue Lias). In the latter are the Beacon Farm Quarry (Bristol Stone and Concrete Co.) 1.25 miles north-east of Shepton Mallet in Lower Lias limestones (Downside Stone) and a quarry at Doulting 900 yd north-east of the church in Inferior Oolite (Doulting Stone).

Lime

Lime burning on a small scale was formerly widely practised using the limestones in the Lower Lias and the Carboniferous Limestone but now the bulk of the lime production comes from the single large quarry of the Callow Rock Lime Co. situated 1.5 miles south of Shipham. Here gas combustion methods are used to produce high grade lime from the Burlington Oolite (Carboniferous Limestone). Smaller limeworks are situated at Tor Hill, Wells (Carboniferous Limestone) and near Bishop Sutton (Blue Lias, see below).

Building stone

In former days building stones were quarried and used locally from many different formations. These included the Carboniferous Limestone, Pennant Sandstone, Dolomitic Conglomerate (including Draycott Marble), Keuper sandstones, Rhaetic limestone (Wedmore Stone), littoral Lower Lias limestones (Chilcote and Downside stones), White and Blue Lias, and Upper Inferior Oolite (Doulting Stone). Of these only the Doulting Stone can be described as a good freestone though the Blue Lias has also been used for interior carving, in Wells Cathedral and elsewhere.

Pennant Sandstone has been much wrought in the past at Temple Cloud and workings still persist.

At the Doulting Quarries of the Bath and Portland Stone Firms Ltd., situated 400 yd north of Doulting church, the Upper Inferior Oolite freestones are worked for high grade building and repair work, though on a much reduced scale to former times. The commercial stone-bed is about 20 ft thick; its weathering properties have been described by Arkell (1947, pp. 102–5).

Blue Lias is worked for paving and wallstone in a quarry at Clapton near Easton, and at Stowey Quarry [ST 597 586] about 1 mile south-east of Bishop Sutton, where the rock is also used as a source of burnt lime.

Lead and zinc ores

The working of lead ores in the Mendips dates from Roman times, the hey-day of the industry was probably in the 17th century but activity continued on a reduced scale until the early years of this century. A conservative estimate of the tonnage produced of lead concentrates suggests a figure of the order of a quarter of a million tons. Serious exploitation of the zinc ores occurred between the beginning of the 17th century and the middle of the 19th century and the amount produced must have been considerable, possibly even approaching that of lead but no statistics are known. Mining apparently ceased with the exhaustion of the readily accessible ore. The history of mining in the Mendips has been written by Gough (1930) while a recent paper (Green, 1958) deals with the lead-zinc mineralization from a geological and economic standpoint. A note on the paragenesis of the ores based on a preliminary study of about 25 specimens has been given by Dearnley (1960, p. 49) while a recent study by Moorbath (1962) of lead isotope abundances in British galenas included specimens from the Mendips.

The main ore of lead was galena and that of zinc was smithsonite (calamine ZnCO₃). These occurred as veins or fissure fillings in the Carboniferous Limestone and Dolomitic Conglomerate in association with a gangue of calcite and, to a less extent, baryte and, rarely, barytocelestine. The number of veins worked was considerable and only a small proportion of those mapped on the 6-in scale are shown on the 1-in map. Minor amounts of galena and sphalerite (ZnS) have been noted in the Harptree Beds (see p. 95 above), and the Lower Lias. Mineral occurrences lying outside the area specifically dealt with by Green (1958) are given below.

At the western end of the Carboniferous Limestone outcrop there are many old workings south and south-west of Banwell, and north-west of Christon. The veins are aligned along the strike (east-west) of the steeply dipping limestone and contained smithsonite as well as galena (see also Green 1958, pp. 83–4). G.W.G.

To the south of the main orefield, the Carboniferous Limestone between West Horrington and the Haydon Farms is pitted by many shallow shafts with spoil heaps containing calcite, baryte and some galena. About a mile to the east, an oxidized lead-zinc vein can be seen in the approach cutting to an abandoned quarry in Carboniferous Limestone 1 mile west of Maesbury Castle (see pp. 45, 106, above). The vein trends east-south-east and is vertical locally. Elsewhere the dip is southerly, the inclination being nowhere less than 65°. About 1 ft thick at the bottom of the cutting it thins upwards, is displaced southwards by three small bedding-plane faults with a total throw of about 4 ft, and then dies out some 20 ft above the cutting floor. The wall rock (Black Rock Limestone) of the vein is much altered and locally rotten. The relatively undisturbed nature of the vein shows that it post-dates the main period of Armorican (Permo-Carboniferous) earth movements which markedly affect the Carboniferous Limestone in its immediate vicinity. D.R.A.P.

Mr. R. W. Elliot reports that the vein rock (E26772) is composed largely of granular smithsonite with irregular aggregates and veinlets of limonite. Small irregular crystals of galena contain carbonate in the cleavage cracks. Locally baryte forms sheaf-like bladed aggregates or small crystals. A chemical analysis (Lab. No. 1892) by Mr. G. A. Sergeant of a channel sample taken across the vein near the cutting floor showed percentages as follows: Fe₂O₃ 2.1; CaO 0.2; MnO 0.02; CO₂ 22.4; Sulphide S 0.005; PbO 3.9; ZnO 40.2; SiO₂, silicates etc. 1.5; BaSO₄ 28.1. Dr. R. Dearnley has recalculated this into the following approximate percentage mineral composition: Smithsonite 62.2; Baryte 28.1; ?Cerussite 4.7; Limonite 2.5; Quartz 1.5; Calcite 0 .4; Galena 0.04; MnCO₃ 0.04. There is a deficiency of about 0.6 per cent of CO₂ for allocation to calcite, MnCO₃ and possibly cerussite. Dr. Deamley comments 'that no cerussite has definitely been determined in this section or crush but since the sulphide figure of the chemical analysis is so low, presumably lead is in the form of PbCO₃.

A crush of representative portions of the vein gave the following mineral percentages: Smithsonite (?possibly also some cerussite) 75.3; Iron Oxides 2.0; and, by difference, Baryte (also including small amounts of galena and quartz) 22.7. Dr. Dearnley further comments that 'although the above figures derived by chemical analysis and by mineral separation do not agree closely—they give the same order of percentages of the minerals present. In particular the proportion of galena seems very low in comparison with the general appearance of the specimen but the sulphide determination is low and very little galena was found in the mineral crushes'.

Note on the age of the Mendip lead–zinc mineralization

The latest rocks to be affected by large-scale lead-zinc mineralization occur in the upper part of the Keuper succession although there are minor amounts of galena and sphalerite in the Lower Lias, and (on Frome Sheet 281) in the Upper Inferior Oolite. Moorbath (1962, pp. 318–9, 335, 345) has determined the lead isotope abundance of three specimens of galena from the Mendips taken from occurrences in the Carboniferous Limestone, the Dolomitic Conglomerate and the Lower Lias (Downside Stone), respectively. The model ages of the specimens were in good agreement, and ranged from 220 ± 60 million years to 240 ± 60 million years giving a mean figure of 230 ± 30 million years i.e. late Permian–late Triassic. Moorbath (p. 319) concluded that this 'average model age . is significantly lower than the value obtained from the Cornwall–Devon mineralization and supports the suggestion of Triassic mineralization for the Mendips'. He further commented that the specimen from the Lower Lias limestone 'shows no significant isotopic difference from the other two specimens but it cannot, without evidence, be regarded as implying a post-Liassic mineralization for the major Mendip occurrences. The minor occurrences in the Lower Jurassic might be derived or regenerated from older deposits. In any case the absolute time differences involved are too small for any definite conclusions regarding this point without sufficient further measurements for statistical comparison'.

If the metasomatic alteration which has given rise to the Harptree Beds in the Wells District is regarded as hydrothermal in origin, it is a possibility that the baryte and the associated small amounts of galena and sphalerite that occur in these beds belong to a further, and much later period of mineralization—possibly post-Albian in age (see p. 95 above). G.W.G.

Iron ore

Traces of former iron workings are widespread in the Mendips but it is unlikely that annual production ever exceeded hundreds of tons. Gough (1930, p. 247) considers that the peak of mining activity may have been reached in the 19th century. The ores are more or less hydrated ferric oxides and different varieties were wrought for smelting and as yellow and red ochre for use as pigments in the manufacture of paint (Cantrill, Sherlock and Dewey 1919).

The ores occur in pipes or lining the sides of fissures in the Carboniferous Limestone and the Dolomitic Conglomerate, or, more commonly, as vaguely defined beds or patches of metasomatic replacement in the latter. At Temple Cloud the ore occurs in fissures and joints in the Pennant Sandstone. The iron commonly contains vughs of calcite and quartz but products attributable to the lead-zinc mineralization are very rare. It is probable, however, that the emplacement of the iron predates the lead-zinc mineralization because in a few cases it is still possible to see haematitic material lining the walls of worked-out veins and in one place (see below) baryte was noted veining the iron ore. Lloyd Morgan (1890, p. 178) reported the presence of calcite-galena veins in the iron ore at Higher Pitts Farm, near Priddy.

The iron ores appear to be similar to those well known in the Forest of Dean and elsewhere and therefore, by analogy, are probably due to descending iron-rich solutions in Triassic times (Sibly and Lloyd 1927; Trotter 1942). These authors are exponents of two opposing views which have long been held concerning the origin of the iron in the Forest of Dean. Sibly (1927, pp. 86–9) considered that the iron was mainly derived from Triassic rocks, whereas Trotter (1942, pp. 75–7) argued in favour of derivation mainly from the denudation of the Coal Measures.

Details

Small workings for yellow ochre are confined to scattered occurrences in the Carboniferous Limestone and Dolomitic Conglomerate in the area of the Blackdown Pericline, the most recent being during the 1930s in the area 0.75 mile south-west of Tyning's Farm, Charterhouse. Here the ochre occupied fissures and pipes in the Carboniferous Limestone and was mostly worked by means of shafts less than 30 ft in depth. Another group of shafts are present about 500 yd farther west from which spoil containing abundant calcite, and yellow powdery limonite with nests of calcite and goethite was noted.

Small workings for red haematitic ores are relatively numerous in the vivid red Dolomitic Conglomerate which rings round the eastern end of the Blackdown Pericline between Compton Martin and King Down Farm (2 miles north-west of Priddy). All workings have long been abandoned with the exception of a small mine at Cliff Quarry, Compton Martin (see p. 74 above). In the Carboniferous Limestone at the eastern end of the Blackdown Pericline traces of haematitic material, commonly more or less siliceous, can be seen in places adhering to the joint-faces and along the sides of worked-out lead veins. Dumps of iron ore with numerous vughs of quartz, seen about 1 mile south-south-east of Compton Martin church, have probably been abandoned because of their silica content. Iron ore in situ in the Carboniferous Limestone (Hotwells Limestone) is seen in a fissure [ST 5459 5483] 1 mile north-north-east of the Castle of Comfort Inn. The upper part of the fissure, which measures 6 ft across and is filled with iron ore, rapidly decreases downwards to a width of 1 ft and this narrow part of the fissure is empty. The ore, which is mainly goethite, is cut by numerous veinlets of baryte and contains many irregular vughs infilled with idiomorphic quartz and baryte. When both these minerals are present in the same vugh the quartz has grown on the baryte. Vivid red Dolomitic Conglomerate now outcrops within 30 ft of this exposure and it is estimated that originally the Triassic base must have lain some 15 to 20 ft vertically above.

In the heavily mined area to the south-east of the Castle of Comfort Inn the mining waste from lead veins in the Dolomitic Conglomerate commonly includes small amounts of more or less siliceous iron ore.

Haematitic ore was mined on a small scale up till 1891 in the neighbourhood of Higher Pitts Farm, north of Wookey Hole. The ore occurred as 'pockets and veins' in the Dolomitic Conglomerate and included irregular patches of earthy manganese oxides ('wad'). This locality is well known due to the presence of rare secondary lead and copper minerals in small quantities in vughs in the ores (Kingsbury 1941, pp. 76–7, and references). G.W.G.

A quarry, [ST 5677 4641] 400 yd south-west of the chapel of the Wells Mental Hospital, in Carboniferous Limestone shows thin haematite–goethite–quartz joint infillings. About 400 yd south-east of Washingpool Farm, East Horrington a narrow belt of red ochreous clay probably represents a weathered haematitic vein or fissure deposit. A small trial pit 600 yd east of Croscombe church on the eastern side of Ham Woods lies in Dolomitic Conglomerate with haematite–quartz–chalcedony impregnations and much associated alteration to quartz–chalcedony. D.R.A.P.

Manganese ore

Small quantities of earthy black manganese oxides 'wad' have been worked in the Mendips for the manufacture of glass and of pottery glazes from the middle of the 17th century to the end of the 19th century. The ores, which occur both in the Dolomitic Conglomerate and the Carboniferous Limestone, are almost invariably associated with iron ores and commonly with small amounts of secondary lead and copper minerals (Gough 1930, pp. 233–9; Kingsbury 1941, pp. 67–80).

Baryte, barytocelestine

Baryte is a common gangue mineral in the lead-zinc veins of the Mendips and was known to the miners as 'cawk' or 'caulk'. It was mined for a short time in the mid-19th century on Banwell Hill together with lead and zinc ores (Green 1958, p. 83).

Debris from two bomb craters [ST 535 538] about 1000 yd east-south-east of Haydon Grange consisted of silicified and haematitized Dolomitic Conglomerate and abundant red and yellow-stained massive baryte. The latter is described by Mr. R. W. Elliot (E24618) as 'an iron-stained baryte rock consisting essentially of interlocking prismatic baryte (α = 1.636: γ= 1.649) in patches coarse-grained. Locally the interstices are impregnated by limonite …and haematite… . ' At this locality the Triassic overlies the pre-Triassic Stock Hill Fault which is strongly mineralized in the Carboniferous Limestone to the north-west.

The highest proportion of baryte in old mining spoil occurs in that derived from veins in the Dolomitic Conglomerate north of Stock Hill and Green Ore. Also in this area, about 1000 yd south-east of Red Quarr, spoil of vein material from a group of shafts in the Old Red Sandstone includes much baryte. Mr. E. A. Jobbins reports on a specimen of the latter as follows:='Radiating platy crystals of baryte with small amounts of celestine, many showing evidence of zoning (α = 1 .6348; (β = 1 .6470; γ = 1.636)'. At Emborough, 0.25 mile south-west of Old Down Inn, the basal Triassic strata overlying the Carboniferous Limestone were reported to be heavily mineralized by an association of baryte, calcite and quartz (Morgan and Reynolds 1899, pp. 109–17). G.W.G.

A somewhat similar association was noted in the area west of Beacon Hill and north of Downside, Shepton Mallet where abundant baryte, barytocelestine and quartz-chalcedony rock fragments were seen in the surface drift in a number of localities. D.R.A.P.

Celestine

Celestine (strontium sulphate SrSO₄) locally known as 'sugar stone' or 'strontia' is found in small quantities in sediments of Keuper age both north and south of the Mendips. Sherlock (in Sherlock and Hollingworth 1938, pp. 80–8) has discussed the geology and mode of occurrence of celestine in Somerset and Gloucestershire and although he recorded no occurrences on the Wells Sheet the mineral was, in fact, worked before the First World War (1914–18) on a small scale in a few localities south of the Mendips. Apart from these occurrences, which are apparently of sedimentary origin, celestine and barytocelestine are now known from Mendip vein material (see previous section).

North of the Mendips, celestine is found as nodules in the Keuper Marl at about the level of the Woodford Hill Sandstone although one case is also known from the Butcombe Sandstone (p. 70 above). Apart from the latter, the occurrences have all been reported from a relatively narrow belt of country extending from Litton north-westwards to the old celestine workings at Regilbury Court (Bristol Sheet 264), about 1 mile north-east of Butcombe.

South of the Mendips, celestine has been found at about the level of the Westclose Hill Conglomerate, and at about the middle of the Tea Green Marl. The area involved extends from Westbury to south of Croscombe and westwards as far as Bleadney.

Details. North of the Mendips, in addition to occurrences mentioned elsewhere (pp. 70, 73 above), Kingsbury (1941, p. 75) recorded celestine from an excavation 400 yd north of the mill at Coley, while 'sugar stone' was reported from a well 650 yd north-west of the Rectory, Nempnett Thrubwell.

South of the Mendips, celestine was dug from a small field [ST 5082 4824] 950 yd E. 28° S. of Westbury church at the base of the Dolomitic Conglomerate (Westclose Hill Conglomerate). Celestine nodules can still be seen in the red marls just below the conglomerate in the immediate vicinity.

It is probable that the same horizon is represented in several localities in the Wells–Dinder area where it is estimated to lie at about 80 to 100 ft below the base of the Rhaetic. Reynolds (1912, p. 124) stated that a 'considerable amount of impure celestine was found in the Trias' at Underwood Quarry [ST 539 468] 1000 yd north-east of Wookey Station. Excavations (1955) for a crusher [ST 5690 4405] at Dulcote Quarry 900 yd south-east of Dinder church yielded quartz-calcite nodules in red marlstone. Mr. E. A. Jobbins reports that the surface of a nodule 'is composed of tiny prismatic crystals of quartz, presumably forming pseudomorphs after some other mineral (baryte or celestine ?)'. Lastly, 'strontia' is reported to have been dug from two fields about 1200 yd S.S.E. of Dinder church on the south side of the railway. The bed was up to 4 ft thick and was worked out (information per H. E. Balch, Esq.).

South of the Mendips, celestine occurs not uncommonly as nodules, associated with calcite, in the middle of the Tea Green Marl but apparently the only area to be worked was situated on the plateau 1 mile south-west of Dinder. Here the surface rubble contains much calcite-celestine debris on which Mr. R. W. Elliot reports (E27567)–(E27568) as follows:—'A celestine-calcite rock composed of laths of celestine (α = 1 .621;γ= 1.631), varying in size from doubly terminated crystals commonly 0.1 mm long to large plates about 1 mm long with coarse plates of calcite (ω = 1.658) . the celestine apparently commenced to crystallize before the calcite but the crystallization of the two minerals overlapped . '. Other celestine occurrences (listed west to east) are as follows:—Surface rubble [ST 484 453] 350 yd south-west of the Inn at Bleadney, shallow excavations [ST 503 442] on Callow Hill, 850 yd E. 10° S. of Hum Farm, and surface rubble on the northern slopes of Ben Knowle Hill and on the southern slopes of Hay Hill. ^‡33 

Peat

The extensive deposits of peat south of the Isle of Wedmore have long been cut on a small scale, formerly for fuel and now mainly for horticultural purposes. The thickness of peat removed is determined by the level of the water table and varies from 4 to 7 ft. Differences of level as between adjacent fields reflect different histories of cutting and probably few virgin areas remain. G.W.G.

Gravel

On Beacon Hill, about 0.5 mile north-east of Beacon Farm, gravel was formerly won from disintegrated pebbly sandstone bands in the Old Red Sandstone. Shallow trenches up to 10 ft deep along the strike of the beds can still be seen. D. R. A .P .

Fuller's earth

At Emborough, 0.25 mile south-west of Old Down Inn, the Tea Green Marl and Westbury Beds (Rhaetic) were dug on a small scale as sources for 'light' and 'dark' fuller's earth respectively (Morgan and Reynolds 1899, pp. 109–17). Small workings, 0.25 mile west of Binegar church, in clay (Harptree Beds) were reported to be for fuller's earth which was ground in a mill nearby (see p. 109 above).

Moulding sand

On Blackberry Hill, north-east of Clutton, sand (Midford Sands) was dug about 1885 and used as a moulding sand at Paulton Foundry. F.B.A.W.

Agriculture

In the district covered by the Wells Sheet the chief connexion between agriculture and geology lies in the close relationship between the rocks and the soils.' ^‡34 Rainfall and temperature differences due to variations in topographical relief also have an important influence on the agriculture.

In the Old Red Sandstone areas, mainly restricted to high ground, the leaching effect of high rainfall and the sandstone bedrock give rise to shallow acid soils which often remain as uncultivated moorland. In some of these areas afforestation with coniferous trees has been carried out.

The Carboniferous Limestone of the Mendips usually bears fairly shallow light brown soils and much of the ground is only of moderate quality. Most of the land is given over to the pasturage of dairy and store cattle although the flatter areas may be ploughed and used for growing winter feed. Mining for lead and zinc in the Carboniferous Limestone and Dolomitic Conglomerate has left large tracts of derelict 'gruffy ground' pitted with old workings and covered with mining spoil. When lead working was active in the Mendips animals grazing on this ground were said to suffer from lead poisoning (Gough 1930, pp. 12–4). Some of these old working areas are being afforested whilst others, since the last war, have been levelled and are now used for food production.

The Coal Measures give rise to non-calcareous soils, those derived from the Lower Coal Series are mainly stiff clay, those on the Pennant mainly thin and sandy. The soils of the Upper Coal Series show greater diversity being partly sandy and partly clay or silt. The Triassic rocks weather to give deep fertile soils, typically red in colour. The marls give rise to heavy loam, the sandstones to rather sandy barns and the Dolomitic Conglomerate to shallow red soils akin to those developed on the Carboniferous Limestone.

Heavy clay land characterizes the outcrop of the Rhaetic and Liassic clay, while the White and Blue Lias usually give rise to clay mixed with abundant limestone debris or 'brash'. Light, dry soils are found on the Inferior Oolite and Downside Stone, the latter being not unlike Carboniferous Limestone in its soil-forming properties. The Harptree Beds are generally deeply weathered and form wide spreads of yellowish and chestnut coloured sandy loam with scattered cherty debris which extends for considerable distances beyond the mapped outcrops onto the adjoining Carboniferous Limestone outcrops. This land tends to be of rather poor quality with acid, hungry soils, and an uneven surface due to the presence of numerous swallets.

Loamy Head overlying Triassic rocks at the foot of the Mendips usually forms well-drained deep red-brown soils suitable for intensive cultivation. On the warm south-facing slopes at the foot of the Mendips between Axbridge and Wells this type of land is much used for horticulture the principal crops being early strawberries followed by anemones. In recent years the productivity of this land is being enhanced by the increasing use of irrigation. In marked contrast to the areas just considered the Head deposits overlying the Palaeozoic rocks on the Mendips usually contain much Old Red Sandstone debris. Owing to this situation in exposed areas of high rainfall their soils resemble those of the surrounding Palaeozoic areas.

Winter flooding limits the use of the wide alluvial areas of the Somerset moors to cattle grazing during the months of May to October. Heavy hay crops are produced on the moist land of the peat areas. G.W.G.

Water supply

In the district covered by the Wells Sheet local water requirements are mainly met by ground water derived from springs, boreholes and wells. The area also includes three large reservoirs which form an important source of water for Bristol. The important aquifers are considered below, other formations make little or no contribution to the water supplies of the area.

Old Red Sandstone

For a formation consisting largely of sandstone the Old Red Sandstone of the Mendips has proved to be a disappointing aquifer. This is partly because the pore-spaces between the individual sand-grains have been largely filled with secondary silica, and partly because the sandstones are massive with bedding-planes and joints exceedingly close-set. Adjacent to fault-zones the rocks may be highly shattered and jointed but even in these cases the yields may prove disappointing because the joints may have been infilled by minerals, particularly calcite. Springs thrown out from the Old Red Sandstone by the overlying impervious Lower Limestone Shale have been quoted as examples of 'overflow springs' (Richardson 1928, pp. 9–10) but due to the low permeability of the sandstones, the flows are small. Whittard (1949, p. 479) has summarized the water-bearing properties of the Old Red Sandstone and concludes 'that it is generally true to state that only under exceptional circumstances will supplies in excess of 5000 gallons an hour be won from the ground'.

The largest recorded yield from the Old Red Sandstone is from Biddle Combe, 2 miles north-north-east of Wells where two boreholes (p. 000) sited within a few feet of each other together yielded on test 5000 g.p.h. The boreholes are respectively 402 ft and 220 ft deep and the water is reported to have been struck at about 200 ft. When pumping exceeded 3000 g.p.h. red colloidal iron appeared in the water; continuous pumping for one month did not result in any reduction in iron. These boreholes are adjacent to the Biddle Fault. At Priddy, a borehole (p. 13) adjacent to the Priddy Fault gave disappointing results. The top 190 ft of this bore was in Lower Limestone Shale and the remaining 100 ft in fractured Old Red Sandstone. Although the rocks were strongly jointed, in nearly all cases the joints were filled with calcite. The yield was only 200 g.p.h. on test over 24 hours with a depression of water level from 14.5 ft to 201 ft below the surface. The deepest borehole in Old Red Sandstone in the Wells area actually went through a major fault. This was the Rookham Borehole (p. 206), 1.5 miles north of Wells, which entered Old Red Sandstone at a depth of 411 ft and continued in that formation to the bottom at 574 ft. The fault zone lay between 459 ft and 500 ft. Due to the massive quartzitic nature of the sandstone and the extensive clayey gouge in the fault breccia the yield from the Old Red Sandstone was estimated to be only about 1500 g.p. day.

The majority of boreholes in the Old Red Sandstone are from 50 to 250 ft in depth and have yielded up to a few hundred g.p.h. A borehole [ST 539 513] on North Hill, however, drilled to 86 ft in very broken sandstone, yielded on test 1000 g.p.h. over 8 hours with a depression of water level from 2 ft to 20 ft below the surface (Whittard 1949, p. 481).

In all the Mendip periclines small springs issue from the Old Red Sandstone especially near the junction of this formation with the overlying Lower Limestone Shale. Many of these springs are used for local water supplies. In an attempt to develop one of these springs a horizontal adit in the Western Twin Stream valley at Burrington Combe (p. 111) was driven through the Lower Limestone Shale and continued for 175 ft into the Old Red Sandstone. The resulting supply only gives minimum and maximum daily yields of 11 000 and 76 000 gallons (see also, Tratman 1963, p. 25).

Analyses of water from various springs issuing from the Old Red Sandstone gives figures of total hardness of between 55 and 96 p.p.m. (as calcium carbonate) and for total solids of between 87 and 150 p.p.m. To the south of Beacon Hill, however, where much of the Old Red Sandstone outcrop is covered by Lower Lias limestones, water issuing from the Old Red Sandstone is markedly harder due to water infiltrating from the limestones. Thus, at Beacon Hill itself the spring water has a total hardness of 85 p.p.m. but 600 yd to the south-west down the Windsor Hill valley, in springs near Millbrook Farm, the figure is 217 p.p.m., whilst farther to the south-west in the same valley in springs at the junction of the Old Red Sandstone and the Lower Limestone Shale the figure has risen to 285 p.p.m.

Lower Limestone Shale

The water-bearing properties of the Lower Limestone Shale are unexpectedly variable for a formation which consists largely of shales. Although seepages or small springs are commonly thrown out from the Old Red Sandstone by the overlying Lower Limestone Shale the quantities of water involved are small. An important group of springs near Lower Farm, Charter-house [ST 487 560], which yield some 400 000 to 500 000 g.p. day, mostly rise from about the middle of the Lower Limestone Shale sequence. A very similar daily yield has also been recorded from a well in the Lower Limestone Shale at Priddy (p. 35) where it has been recently shown by Bristol Waterworks Company that the water comes from fissured limestones within the top 20 ft of the well. Both the Charterhouse springs and the Priddy well are situated in the floors of valleys whose higher reaches extend onto the Old Red Sandstone outcrop. The high yields suggest that there may be hydraulic continuity between the Old Red Sandstone and the Lower Limestone Shale, possibly along joint systems developed under the valley floors. Massive limestone bands in the lower part of the Lower Limestone Shale in the Pen Hill Pericline may contain appreciable amounts of water in the relatively gently dipping southern limb of the pericline. A borehole at Rookham (see pp. 205–7) at the western end of the pericline struck an artesian flow at a depth of 318 ft in red 'Bryozoa-Bed type' limestones. Acidization of this borehole recently undertaken by Bristol Waterworks Company increased the yield from some 18 000 to 66 000 g.p. day. The nearby Holes Ash spring (Richardson 1928, p. 187) with a yield varying from 60 000 to nearly 200 000 g.p. day probably issues from these limestones, as may other springs about 1 mile to the east in Biddle Combe. Farther east, a 440-ft deep bore of 6-in diameter at Slab House [ST 591 482], which started near the base of the Black Rock Limestone and continued through the greater part of the underlying Lower Limestone Shale sequence overflowed at the rate of about 100 g.p.h. and on test produced about 900 g.p.h. by pumping. No details are available as to the depths at which water was struck.

Boreholes drilled in the Lower Limestone Shale in the area of the North Hill Pericline and along the vertical strata comprising the northern limb of the Pen Hill Pericline have produced yields of 300 g.p.h. or less, or have even been complete failures. At Burrington Combe a 1000-ft adit in the bottom of the Western Twin stream (see p. 177) which traversed the greater part of the Lower Limestone Shale sequence, including 'Bryozoa Bed' limestones, and extended some way into the Old Red Sandstone failed to produce more than 27 000 to 60 000 g.p. day.

The chemical constitution of water from the Lower Limestone Shale is intermediate in type between typical Old Red Sandstone and Carboniferous Limestone waters giving figures of—total hardness 165 to 200 p.p.m.; temporary hardness 130 to 170 p.p.m.; total solids 170 to 250 p.p.m.

Carboniferous Limestone

The Carboniferous Limestone is the most important aquifer in the area of the Mendips and the surrounding country, due to the combination of the relatively high rainfall over the Mendips and the permeability of the limestone itself which enables the rain-water to be absorbed into the ground without large losses due to surface run-off. The water percolates downwards through, and is stored in, a complex system of fissures, joints and caves within the limestone and the overflow escapes as large springs, known locally as risings, at the foot of the Mendips. Under the greater part of the Mendip plateau the water table in the Carboniferous Limestone lies many hundreds of feet below the surface—as demonstrated by cave exploration. At Swildon's Cavern, Priddy, for instance, the water table lies some 450 ft below the surface (about 800 ft O.D.) and at G.B. Cavern, near Charterhouse, at a depth of 480 ft below the surface (about 850 ft O.D.). In addition to rain falling on the limestone outcrop the water table is also fed by surface streams, draining the Old Red Sandstone and Lower Limestone Shale outcrops, which disappear underground down swallets or swallow holes when they reach the Carboniferous Limestone. The hydraulic connection between these surface streams and the risings at the foot of the hills has been proved by the pollution of the risings by refuse turned into the upland streams by lead miners during the 19th century and earlier (see Gough 1930, pp. 14–6, also p. 125 above). Although most of the risings break out in Triassic or even younger rocks the relatively limited local catchment areas of these rocks and the chemical composition of the water shows that the bulk is derived from the Carboniferous Limestone.

Large quantities of water are discharged from the risings, much of which is used for public water supplies. The largest water undertaking is the Bristol Waterworks Company whose sources include most of the water from the risings at Rickford (average daily flow 2.5 million g.), Langford (1 million g.), Cheddar (15 million g.), Sherborne Spring, Litton (1.5 million g.), springs north-west of Chewton Mendip (3.25 million g.) and between Draycott and Rodney Stoke. The average daily flow of the rising at Wookey Hole is equal to about 23 million g. (Gough 1930, p. 15). In contrast to the copious supplies obtainable from springs the yields of wells and boreholes sunk in the Carboniferous Limestone are relatively small due to the infrequency of large water-bearing fissures in this formation. At Gurneyslade Bottom, Binegar, a 36-ft well with associated headings, belonging to the Downside Abbey authorities, supplies over 250 000 g./day to various local authorities. Yields in excess of a few thousand g.p.h. are, however, exceptional.

The location of water supplies in the Carboniferous Limestone is hazardous due to the infrequency of fissures in this formation coupled with the difficulty, or impossibility, of predicting their course from surface observations. Groundwater movement is probably most active, and hence the presence of water-bearing fissures most likely, beneath the floors of present-day valleys and particularly in areas adjacent to springs. Even in such situations, however, success in boring is uncertain. Nearby boreholes in similar hydrogeological conditions may need to be drilled to widely differing depths below the water table before water-bearing fissures are encountered. Thus in the lower part of the Shipham Gorge valley between Cheddar and Axbridge two boreholes 400 yd apart [ST 449 550], [ST 446 548] and starting at 194 ft and 196 ft above O.D. proved 107 ft and 78 ft of Drift and Triassic, but drilling had to be continued in the underlying Carboniferous Limestone to total depths of 502 ft and 180 ft respectively before water was struck. In the former case the water rose to within 92 ft of the surface and test pumping yielded 2000 g.p.h. with a depression to 120 ft below ground level. In the latter case the water rose to 62 ft below surface level and pumping at 1000 g.p.h. failed to lower the water level appreciably.

The difficulty of locating water-bearing fissures in massive limestone is further illustrated by a shaft [ST 628 431] sunk in Shepton Mallet during 1935–6. Here some 70 ft of massive Lower Lias limestones (Downside Stone) with some thin shaly bands, overlie thin Rhaetic shales which rest directly upon the Carboniferous Limestone. The shaft was sunk to 178 ft and a borehole of 12-in diameter was drilled a further 20 ft in the bottom of the shaft without locating a water supply. At a depth of 64 ft from the surface a heading driven in a south-westerly direction for 94 ft in the Lower Lias limestones failed to strike water. A second heading was then driven in a southerly direction from shaft-bottom and at 30 ft intercepted a small fissure which yielded water at 250 g.p.h. A drill-hole (2-in diameter) was then driven upwards for 25 ft in the roof of the heading and located a fissure which yielded about 3000 g.p.h. Finally, a third heading was driven from a depth of 160 ft to intercept the fissure proved in the drill-hole and this heading together with additional small diameter holes drilled outwards from it increased the yield to over 5000 g.p.h. When pumping was stopped the water rose to within 46 ft of the surface, or about 20 ft above the top of the Carboniferous Limestone in the shaft, thus suggesting that the shaly beds in the Lias and Rhaetic are confining the water under pressure in the underlying Carboniferous Limestone. This contrasts with conditions in other parts of Shepton Mallet where there appears to be some degree of hydraulic continuity between the Lias and Carboniferous limestones, presumably through interconnecting fissures between the two formations. About 650 yd east of the church, a borehole [ST 626 436] in the Sheppey valley, sunk through 70 ft of Lias limestones resting on thin Rhaetic shales, was continued into the underlying Carboniferous Limestone to a total depth from the surface of 142 ft where an artesian flow of 1000 g.p.h. was encountered. Test pumping at 5125 g.p.h. for 7 days depressed the water level to 82 ft but on cessation of pumping the recovery time was only 37 minutes. The yield of a spring issuing from the Lower Lias limestones in the valley bottom about 200 yd up the valley was, however, more or less diminished in proportion to the pumping of the borehole. A second borehole situated only some 100 yd away from the successful bore and drilled to 200 ft was dry. Also in the Sheppey valley, but west of Shepton Mallet, springs at Darshill issue from the base of the Lower Lias limestones at the rate of some 250 000 g.p. day. The relatively large quantity of water in this case suggests that part of it is derived from the Carboniferous Limestone although two wells nearby which were sunk to depths of 98.5 ft and 120 ft and penetrated some 20 ft into the Carboniferous Limestone were unsuccessful (Richardson 1928, p. 173).

Chemical analysis of water from the Carboniferous Limestone gives figures for total hardness of between 235 and 320 p.p.m. of which temporary hardness, mainly due to calcium bicarbonate, accounts for some 215 to 280 p.p.m. Total solids range from about 350 to 400 p.p.m.

Keuper

The Dolomitic Conglomerate yields important supplies of water both in springs and from boreholes. Water moves freely from the Carboniferous Limestone and the Old Red Sandstone into the Dolomitic Conglomerate and the bulk of the water from the risings that issue from the Triassic at the foot of the Mendips is derived from the Carboniferous Limestone (see above). As with the latter formation drainage is mainly underground through a complex joint system and success in drilling for water depends on striking water-bearing fissures. Water movement in the Dolomitic Conglomerate may be restricted by bands of marl interbedded with the more pervious conglomerates. The marl bands, depending on their topographical position, may either keep down water in underlying beds of conglomerate or may throw out springs from overlying conglomerates. The marls of Keuper age, including the Tea Green Marl, are cut by innumerable little fissures in which small supplies of water collect. More important, however, as aquifers are bands of marlstone, sandstone and conglomerate which are interbedded with the marls and which under favourable circumstances may yield appreciable quantities of water.

On the Mendip plateau where the Triassic overlies the Carboniferous Limestone the ground water is rapidly lost to the latter formation. In the North Hill area, however, where the Triassic overlies the Old Red Sandstone, mining operations during the 17th century were apparently much hindered by water in the Dolomitic Conglomerate (Green 1958, p. 84). All around the foot of the Mendips numerous boreholes for water have been sunk in the Triassic marls and conglomerates, the depths usually being between about 50 ft and 200 ft and the yields varying from a few hundred g.p.h. to some 2000 g.p.h. and sometimes considerably more. The main aquifers are the conglomerates and hence the chances of large yields tend to decrease with increasing distance from the Mendips because of the lateral passage of the conglomerates into marls away from the Mendips. The largest recorded yields are from a series of four boreholes sunk in the area adjoining Cox's springs, about a mile north-east of Winscombe. Here massive conglomerates, underlying some 20 ft or so of red marl, contain an artesian supply which on test-pumping yielded from 8000 to 28 000 g.p.h. About a mile north of Banwell two adjacent trial boreholes (p. 198) of 6-in and 24-in diameter struck an artesian flow in conglomerates at the base of the Triassic at a depth of about 600 ft, which on test yielded 53 000 g.p. day and 184 000 g.p. day respectively. In the City of Wells the Triassic succession consists of red marls with bands of conglomerate and marlstone, the proportion of conglomerate increasing with depth (see pp. 204–5). The main supplies are derived from the hard bands. The deepest bore is 300 ft of which the lowest 120 ft is massive conglomerate with interbedded sandy marl. Water was struck in quantity below about 175 ft and on cessation of boring the water level stood at 18 ft below the surface. Over 11 days, test-pumping averaged about 100 000 g.p. day with a depression of the water level to 270 ft below the surface; the recovery time was 1 hour. Other boreholes in the City drilled to between 150 ft and 200 ft gave yields of from 2000 to 3000 g.p.h. At Easton a borehole (p. 78) sunk to 350 ft in red marl with some harder bands gave a steady yield of 5000 g.p.h. during a 14-day test with a depression of the water level from 15 ft to 130 ft and a recovery time of 24 hours. In this case, however, the water proved excessively hard for domestic purposes.

Both to the north and south of the Mendips where the Keuper succession consists of marls with some thin sandstone and marlstone bands yields of several hundred g.p.h. are recorded from numerous boreholes whose depths range from 35 ft to 170 ft. Exceptionally, rather higher yields are recorded. At Tudor's Farm, South Widcombe [ST 578 565] a 100-ft bore in marl with thin conglomeratic sandstone bands yielded over 1000 g.p.h. At Coxley, a 99-ft bore [ST 527 433] in Keuper Marl yielded 840 g.p.h. during an 8-hour test with virtually no depression of the water level. In this case the water came from a rock band near the bottom of the borehole but was excessively hard for domestic use.

Artesian flows from the Keuper at a long distance from the Mendips are unusual and therefore it is of interest to note that two boreholes at Blackford [ST 410 478], [ST 413 474] are recorded as having struck artesian flows at depths from the surface of about 60 and 75 ft respectively. The tops of the boreholes lie at about the junction of the White and Blue Lias and the supply appears to have come from hard bands within the uppermost 20 ft of the Tea Green Marl.

Except where appreciable dilution from Carboniferous Limestone waters has occurred, as in the Mendip risings, chemical analyses of water from the Keuper rocks show marked differences from those of water derived from the Carboniferous Limestone. Total hardness varies from 300 to 650 p.p.m. of which permanent hardness accounts for 130 to 540 p.p.m. Total solids vary from 430 to 800 p.p.m. Typically, the waters contain much less calcium salts and much greater amounts of magnesium salts than those derived from the Carboniferous Limestone. Water emanating from thick conglomerate bands contains a much lower percentage of dissolved solids than that originating from the predominantly marl successions.

Lower Lias limestones

The presence of shales in the White and Blue Lias severely limits the permeability of these formations which in consequence yield only very limited supplies of hard water. In the Shepton Mallet–Wells area the Lower Lias is chiefly represented by 50 to 100 ft of massive limestones (Downside Stone). Where these rocks directly overlie the Carboniferous Limestone the ground water passes directly into the underlying rocks but where Rhaetic shales or other impermeable rocks intervene between the two a perched water table may occur. Even where the Rhaetic is present beneath the Downside Stone, some degree of hydraulic continuity between the Lias and Carboniferous has however, in some instances, been proved (see above) and this may explain why, apart from the exception quoted below, boreholes have failed to locate appreciable supplies in the Downside Stone. At Darshill, in the Sheppey valley, (see above) strong springs issue from the base of the Downside Stone and a borehole (p. 204) drilled within 60 yd of these springs gave a yield of 2000 g.p.h. during a 10-hour test.

Inferior Oolite

In the Doulting area the drainage of the Inferior Oolite is underground through fissures, whilst the overflow escapes as springs at the top of the underlying Lias clays. The main springs occur around the flanks of Ingsdon Hill and include St. Aldhelm's Well, south-east of Doulting church, which is reported to have a daily flow of 40 000 gallons. G.W.G.

References

ARKELL, W. J. 1947. Oxford Stone. London.

CANTRILL, T. C., SHERLOCK, R. L. and DEWEY, H. 1919. Iron Ores: Sundry Unbedded Ores of Durham, East Cumberland, North Wales, Derbyshire, the Isle of Man, Bristol District and Somerset, Devon and Cornwall. Mem. Geol. Surv., Min. Resources. 9.

DEARNLEY, R. 1960. in Sum. Prog. Geol. Surv. for 1959,49.

DONOVAN, D. T. and REID, R. D. 1963. The Stone Insets of Somerset Churches. Proc. Somerset Arch. Nat Soc., 108, 60–71.

GOUGH, J. 1930. The Mines of Mendip. Oxford.

GREEN, G. W. 1958. The Central Mendip Lead-Zinc Orefield. Bull. Geol. Surv. Gt. Brit., No. 14,70–90.

KINGSBURY, A. W. G. 1941. Mineral Localities on the Mendip Hills, Somerset. Min. Mag., 26,67–80.

LOUPEKINE, I. S. 1956. Mining and Quarrying in the Bristol District, 1955. Proc. Bristol Nat. Soc., 29,155–61.

MOORBATH, S. 1962. Lead Isotope Abundance Studies on Mineral Occurrences in the British Isles and their geological significance. Phil. Trans. Roy. Soc. (A), 254,295–360.

MORGAN, C. LLOYD, 1890. Mendip Notes. Proc. Bristol Nat. Soc. (3), 6,169–82.

MORGAN, C. LLOYD, and REYNOLDS, S. H. 1899. Triassic Deposits at Emborough. Proc. Bristol Nat. Soc., 9,109–17.

REYNOLDS, S. H. 1912. A Geological Excursion Handbook for the Bristol District. Bristol.

RICHARDSON, L. 1928. Wells and Springs of Somerset. Mem. Geol. Surv.

SHERLOCK, R. L. and HOLLINGWORTH, S. E. 1938. Gypsum and Anhydrite & Celestine and Strontianite. Mem. Geol. Surv., Min. Resources, 3, 3rd edit.

SIBLY, T. F. and LLOYD, W. 1927. Iron Ores: Haematites of the Forest of Dean and South Wales. Mem. Geol. Surv. Min. Resources, 10, 2nd edit.

TRATMAN, E. K. 1963. The Hydrology of the Burrington Area, Somerset. Proc. Univ. Bristol Spel. Soc., 10, 22–57.

TROTTER, F. M. 1942. Geology of the Forest of Dean Coal and Iron-Ore Field Mem. Geol. Surv.

WHITTARD, W. F. 1949. Temporary Exposures and Borehole Records in the Bristol Area, IV Boreholes on Mendip. Proc. Bristol Nat. Soc., 27,479–82.

Appendix 1 The faunal succession in the Carboniferous Limestone of Burrington Combe by M. Mitchell And G. W. Green

The Burrington section was first described in detail by Sibly (1905) and more fully by Reynolds and Vaughan (1911). In both these works the stratigraphical ranges of macro-fossils deemed to be characteristic are described in terms of Avonian coral-brachiopod zones and subzones. Recently Cummings (1958) has published an account of the stratigraphical distribution of some of the foraminifera and has compared the 'microfaunal' succession at Burrington with that of other localities in Britain. In the present account the local ranges of some fossil species are defined and are related to the formations shown on the One-inch Geological Survey maps. Palaeontological details and zonal correlations have been supplied by M. Mitchell, the description of the Burrington section and a map illustrating principal landmarks and formational boundaries (Figure 19) have been provided by G. W. Green. In preparing the palaeontological evidence M. Mitchell has himself identified the corals and a part of the shelly fauna while most of the brachiopods and molluscs have been named by Dr. W. H. C. Ramsbottom. Some of the material was collected by G. W. Green, but the bulk of the fossils were obtained during a systematic programme undertaken by the Palaeontological Department and executed by Mr. S. W. Hester assisted by Mr. M. J. Hughes.

In the ensuing account, each formation is described lithologically, then faunally, and finally its relationship to the Avonian Zones proposed by Vaughan and later workers in the Mendips is discussed.

Lower Limestone Shale

This formation although only poorly exposed in the Burrington Combe section was formerly seen during the driving of an adit in the floor of the valley of the Western Twin Stream. A summary of the section recorded by Hepworth and Stride (1950, pp. 135–8) is given below:

Fauna

Hepworth and Stride (1950, p. 135) obtained 'most of the fossils characteristic of the K beds', but gave few details. The present account is based on fossils collected in the Western Twin Stream section at the following levels (given in feet above the base of the Lower Limestone Shale):-50 to 58 ft, 73 to 78 ft, 202 to 210 ft, 266 to 274 ft, 344 to 366 ft, 383 to 408 ft and 423 to 437 ft. In view of this uneven spread of exposures, the fossil ranges quoted must be regarded as tentative.

Brachiopods are the commonest fossils with Chonetids predominating. The stoddarti group of Plicochonetes is represented by three forms at successive levels. Chonetes (Plicochonetes) stoddarti Vaughan (sensu stricto)is restricted to the middle part of the Lower Limestone Shale (202–267 ft). The shells from the higher beds are larger and have a greater number of costae. One form is here named C. (P.) sp. nov.[aff. stoddarti] (359–388 ft) and the other C. (P.)aff. stoddarti (404–437 ft). The latter is probably the form from the K₂ Subzone at Westbury referred by Smith (1925, p. 88) to C. stoddarti. Chonetes failandensis S. Smith has a restricted range in the upper part of the formation (398–402 ft). A number of different forms of Chonetes (Rugosochonetes) sp.occur within the succession (50–437 ft). One of these, a large form with fine costae, has a restricted range (344–387 ft).

Productus (Avonia?) bassus Vaughan has been found only in the upper beds (344–391 ft), associated with P. (Dictyoclostus) vaughani Muir-Wood (359–397 ft). Of the remainder of the brachiopod fauna, Camarotoechia mitcheldeanensis Vaughan is common throughout, whilst Syringothyris spp. and Unispirifer tornacensis (de Koninck) are seen for the first time at 359 ft above the base of the formation and range up into the Black Rock Limestone. Eumetria carbonaria (Davidson) was found at a single horizon (73–74 ft) and Cleiothyridina royssii β (Vaughan) is restricted to a thin band in the upper part of the formation (390–397 ft). The only record of a new subspecies of C. royssii (Davidson) is that from near the base (51–58 ft). Indeterminate specimens of Orthotetids occur through the upper part of the succession (344–403 ft).

Polyzoan beds occur at several levels in the middle of the succession (202–267 ft), in addition to the 'Bryozoa Bed' (80–90 ft) and the 'Limestone with many Polyzoa' (125 ft) recorded by Reynolds and Vaughan (1911, fig. 2, p. 350) which are not exposed in the Western Twin Stream. Scattered gastropods and lamellibranchs occur near the base, and ostracods have been noted especially in the lower part of the formation (51–267 ft).

Black Rock Limestone

With the exception of relatively small gaps, the Black Rock Limestone is almost continuously exposed. For convenience of description the Black Rock Limestone is regarded as including both the black crinoidal limestone and the dolomite which overlies it.

The Main and Lower Chert beds (see p. 17) correspond to Beds 13 and 3, respectively in this section. The lithological succession is as follows. The bed-numbers correspond with the column-numbers on (Table 1).

Fauna

Three main faunal assemblages have been recognized and they are here referred to as the Lower, Middle and Upper Faunas respectively. The Lower Fauna (0–306 ft above the base of the formation; Beds 1–9) is dominated by brachiopods, the Middle Fauna (306–606 ft; Beds 10–15) is characterized by a rich coral fauna but includes many of the brachiopods seen in the beds below, and the Upper Fauna (681–833 ft; Beds 16–19) is indicated by a change in the characteristic Caniniid and also the appearance of pustulose Productids. The dolomites at the top of the formation are poorly fossiliferous.

(i) Lower Fauna (0–306 ft; Beds 1–9)

The fauna in the lowest beds (0–7 ft), despite the gap in the section below the base of the formation shows a clear link with the Lower Limestone Shale fauna below and contains C. mitcheldeanensis, C. (P.)aff. stoddarti and C. royssii β, but is distinguished by the presence of Zaphrentites delanouei (Edwards and Haime) and Chonetes (Rugosochonetes) vaughani Muir-Wood sp. (= C. cf. laguessiana de Koninck of Vaughan 1905 p. 294, pl. 26, fig. 1).

Z. delanouei and Fasciculophyllum omaliusi (Edwards and Haime) both occur sparsely within the range of the Lower Fauna, together with Michelinia favosa (Goldfuss) (100–278 ft; Beds 4–8) and Syringopora θVaughan (178–278 ft; Beds 6–8). Michelinia konincki Vaughan (278–290 ft; Bed 9) is recorded from near the top of beds containing the Lower Fauna together with 'Zaphrentis' konincki forma α Carruthers (278–306 ft; Bed 9).

The brachiopods restricted to the Lower Fauna include Cleiothyridina glabristria Phillips sp. (178–198 ft; Bed 6), C. cf. glabristria Vaughan non Phillips sp. (100–306 ft; Beds 4–9), C. (R.) vaughani (0–255 ft; Beds 1–7), Productus (Dictyoclostus) multispiniferus Muir-Wood (261–278 ft; Bed 8), Rhipidomella michelini Léveillé sp. (178–290 ft; Beds 6–9) and Syringothyris cuspidata cyrtorhyncha North (178–238 ft; Beds 6–7).

Of the long-ranging species, Leptaena analoga (Phillips) is common in the Lower Fauna, ranging from the base of the Black Rock Limestone to the top of the beds containing the Middle Fauna (0–606 ft; Beds 1–15), while U. tornacensis occurs in the top of the Lower Limestone Shale and throughout the Black Rock Limestone, although it is commonest in the beds with the Lower Fauna. Schellwienella aspis Smyth similarly is a long-ranging form common in the Lower Fauna (158–813 ft; Beds 5–18).

(ii) Middle Fauna (306–606 ft; Beds 10–15).

The incoming of the Middle Fauna is characterized by the first occurrence of the corals Caninophyllum patulum (Michelin), Cyathaxonia cornu Michelin and Fasciculophyllum densum (Carruthers) at 306 ft, and of Caninia cornucopiae Michelin at 326 ft above the base of the Black Rock Limestone here (Bed 10). All these corals continue through the beds containing the Middle Fauna, but F. densum also ranges to the top of the beds with the Upper Fauna. Two forms of C. patulum are present, the closely septate form (Vaughan 1911, pl. 30, figs. 6a, b) occurring between 306 and 401 ft (Beds 10–12), and the widely septate form (Vaughan 1911, pl. 30, figs. 5a, b) between 426 and 606 ft (Beds 13–15). Single specimens of Cravenia aff. tela Hudson found in Beds 10 (326–346 ft) and 14 (513–543 ft) are similar to the coral described by Vaughan (1911, p. 380, pl. 31, fig. 3) as an 'early Clisiophyllid'. Fasciculophyllum aff. ambiguum Carruthers sp. (426–606 ft; Beds 13–15) and Koninckophyllum tortuosum Michelin sp. (573–606 ft; Bed 15) are also recorded only from the Middle Fauna. Cyathoclisia tabernaculum Dingwall ( = Cyathophyllum θ Vaughan) also occurs in the Middle Fauna, but ranges up into the base of the Upper Fauna (490–741 ft; Beds 13–16).

'Z.' konincki forma typica Carruthers is characteristic of the Middle Fauna (306–543 ft; Beds 10–14), whereas forma α is recorded only from the top of the beds with the Lower Fauna at Burrington. It is noted, however, that forma α was collected at a level 25 ft below the Main Chert beds, 400 ft above the base of the Black Rock Limestone in the Charterhouse area (p. 28).

The brachiopods of the Middle Fauna are similar to those of the Lower Fauna. Tylothyris cf. laminosa (M 'Coy) however, has a restricted range at the base of the Middle Fauna (306–359 ft; Beds 10–11).

(iii) Upper Fauna (681–833 ft; Beds 16–19).

The nature of the junction between the beds containing the Middle and Upper Faunas is unknown because 75 ft of rock are unexposed. C. patulum is not seen above this gap, and Caninia cylindrica (Scouler) replaces it as the dominant large Caniniid (681–771 ft; Beds 16–17). Elsewhere in the Mendips however, the ranges of C. patulum and C. cylindrica show some degree of overlap (Bamber 1924; Bush 1925). A specimen of 'Z.' konincki Edwards and Haime from Bed 17 (741–771 ft) is identified as being between forma typica and the 'intermediate form' of Hudson and Mitchell (1937, p. 9 footnote).

The brachiopods of the Upper Fauna differ from those below, with Chonetids and pustulose Productids becoming increasingly important. The Chonetids include 'Chonetes aff. papilionacea, Phil.' of Vaughan (1905, pl. 26, fig. 3) (771–833 ft; Beds 18–19), C. (Plicochonetes) sp. nov.[aff. stoddarti](813–833 ft; Bed 19), and a finely costate form here recorded as C. sp. nov.(741–813 ft; Beds 17–18). A Chonetid close to Delepinea comoides (J. Sowerby) is also recorded from the uppermost bed (813–833 ft; Bed 19).

The pustulose Productids found in Beds 16–19 include Productus (Pustula)cf. nodosus (I. Thomas) and P. (P.)cf. pustuliformis Rotai (both from 813–833 ft; Bed 19). P. (P.) pyxidiformis de Koninck is also present (771–813 ft; Bed 18) together with closely related forms (741–833 ft; Beds 17–19).

Productus (Eomarginifera)aff. derbiensis Muir-Wood and Spirifer cf. konincki Dewalque are restricted to the Upper Fauna (681–771 ft; Beds 16–17) while Schuchertella cf. wexfordensis Smyth, S. aff. fascifera (Tornquist) and Syringothyris aff. elongata North are recorded only from the top beds (813–833 ft; Bed 19). S. cf. wexfordensis was collected by one of the writers, however, from Dolebury Warren, about a mile to the west of Burrington Combe at a horizon estimated to lie (in terms of Burrington) a little below the base of the beds containing the Upper Fauna.

The fauna of the dolomite at the top of the Black Rock Limestone is extremely poor due to dolomitization having destroyed many fossils. About 100 yd north of the Great Scarp, shallow trials for minerals in much dolomitized crinoidal limestone, estimated to lie about 850 ft above the base of the Black Rock Limestone, yielded Chonetes (Megachonetes) magnus Rotai (see p. 18). The remaining fossils collected from the dolomite give no indication of the age of these beds and include the following: indeterminate coral fragments; Balanoconcha ?, Chonetes (Megachonetes) sp., C. (Rugosochonetes) sp., Eumetria?; and Straparollus (Straparollus) sp.

Burrington Oolite

The formation is well exposed except for its junction with the overlying Clifton Down Limestone. This junction is, however, well seen in an old quarry [ST 498 582] about 1.25 miles east of the Combe (see p. 25). The exposed section in the Combe starts in the top of 'Quarry 2' of Reynolds and Vaughan (1911), while Beds 1 to 2 and part of 3 are included in the 'Dolomitized limestone, C₁(b)' of those authors (1911, p. 348). The bed-numbers correspond with the column-numbers on (Table 2).

Fauna

The fauna of the lowest beds of the Burrington Oolite has been largely obliterated by dolomitization. The lowest recognizable fauna (36–77 ft; Bed 3) is of Viséan appearance and includes 'Zaphrentis'konincki gr. kentensis Garwood Cyathophyllum ϕ Vaughan; small form convergent to Z. konincki mut. C₂ of Vaughan 1911, p. 368). This coral differs from 'Z.' konincki forma typica, the characteristic form in the Middle Fauna of the Black Rock Limestone, by having an increased number of septa and a greater diameter, but it is not as large as the typical kentensis. A similar form has been collected, in association with an early Visean fauna, from the Vallis Limestone in the Priddy–Binegar area (p. 35).

Palaeosmilia murchisoni Edwards and Haime (= Cyathophyllum ϕ Vaughan) and Lithostrotion martini Edwards and Haime are the commonest corals in the Burrington Oolite, the former in the lower part (156–480 ft; Beds 4–11) and the latter in the upper part (425–620 ft; Beds 9–15) and their occurrence together, with a short range of overlap, is of interest and may be compared with the much longer range of overlap (about 130 ft) in the Cheddar Gorge (see p. 29). Carcinophyllum spp. occur rarely, but one specimen at 544 ft (Bed 12) shows affinities with C. mendipense Sibly.

The Productids are represented by specimens of P. (Gigantoproductus)Vaughan. The specimens from 156–221 ft (Bed 4), here identified as P. (G. ?)aff. θmay refer to the forms which Vaughan (1911, p. 368) called 'P. corrugatus mut. C₂'. They differ from P. (G.)θchiefly because the shell of the ventral valve is thinner; they may belong to the stock from which Productus θarose. From 281–321 ft (Bed 6 part) the form is still not the typical thick-shelled type, while the specimens from 321–371 ft (Bed 6 part) are referred to as P. (G.)early θ. The typical form is abundant from 406–442 ft (Beds 7–9) but some of the specimens from 406–409 ft (Bed 7) simulate Productus humerosus J. Sowerby (= P. sublaevis de Koninck of Vaughan 1911, p. 369), and it is possible that some of them led Vaughan to think that P. humerosus is present at Burrington. However, the reputed occurrence of P. humerosus cannot be confirmed.

Chonetes (Megachonetes) sp. papilionaceus (Phillips) group and R. michelini both occur commonly in the lower part of the Burrington Oolite (36–281 ft; Beds 3–5). A specimen close to Delepinea carinata (Garwood) is of interest (156–221 ft; Bed 4) as is the first appearance of Davidsonina carbonaria (McCoy) in Bed 15, (612–620 ft)^‡37 .

Clifton Down Limestone

The top 120 ft of the formation are well exposed in 'Quarry 1' but the exposures of the remainder are rather poor. The lithological succession is as follows. The bed-numbers correspond with the column-numbers on (Table 3).

The limits of the lithological subdivisions recognized in the surrounding areas (see p. 19) are as follows: Beds 1–4, Banded oolite-mudstone beds; Beds 5–8, Lithostrotion beds; Beds 9–13, upper calcite-mudstone beds.

Fauna

L. martini is the commonest of the corals and ranges through the whole of the Clifton Down Limestone. Carcinophyllum vaughani Salee C. θ. Vaughan) also ranges through these beds, but is never abundant. A form of Lithostrotion slightly larger than L. martini is recorded from 154–262 ft (Beds 4–6) as L. aff. martini. A larger form close to L. sociale (Phillips) was found at two levels (173–215 ft; Bed 5 and 502–510 ft; Bed 13). Associated with L. cf. sociale in the lower of the two beds is a specimen here referred to as P. aff. murchisoni, which is similar to the specimen described by Vaughan (1911,p. 379, pl. 31, fig. 6).

As in the Burrington Oolite, the commonest of the Chonetids belong to the papilionaceus group, and these occur most abundantly in the Lithostrotion beds. C. (Plicochonetes)aff. buchianus Davidson non de Koninck is recorded at several horizons between 109 and 296 ft (Beds 3–7). A Chonetid from the base of the Clifton Down Limestone (60–79 ft; Bed 1), which is more convex than C. papilionaceus, is recorded as Delepinea aff. comoides (J. Sowerby). D. carbonaria first recorded near the top of the Burrington Oolite is found only in the basal part of the Clifton Down Limestone (up to 122 ft; Beds 1–3). This range is considerably more restricted than that recorded by Reynolds and Vaughan who stated (1911, p. 370) that it 'ranges through [S₂]' i.e. to the top of the Clifton Dowr Limestone.

Composita spp.,including C. ficoidea (Vaughan), range through the Clifton Down Limestone and occur abundantly in numerous bands together with Orthotetid fragments. Productids are common, but are usually poorly preserved.

Productus (Linoproductus) corrugatohemisphericus Garwood (= P. cora Milt. S₂ Vaughan) is recorded from 89–173 ft (Beds 2–4) and forms compared with this species occur above and below these limits.

Hotwells Limestone

Only the basal beds of the Hotwells Limestone are exposed at Burrington Combe of which the lowest 47 ft are seen in the 'Quarry 1' of Reynolds and Vaughan. The lithological succession is as follows:

Fauna

The base of the Hotwells Limestone is marked faunally by the first appearance of Dibunophyllum bourtonense Garwood and Goodyear and D. bourtonense ϕVaughan, and the reappearance of Palaeosmilia murchisoni Edwards and Haime. The Lithostrotionids are represented by Lithostrotion junceum (Fleming), L. martini Edwards and Haime and L. pauciradiale (McCoy). The brachiopods Productus (Gigantoproductus) maximus McCoy and P. (Linoproductus) hemisphaericus J. Sowerby are present, with specimens of an Orthotetid compared with Streptorhynchus senilis (Phillips).

This coral-brachiopod fauna shows a marked change from that in the underlying Clifton Down Limestone. The only species common to both, are cf. S. senilis which is first seen in the topmost beds of the Clifton Down Limestone and L. martini which ranges down into the Burrington Oolite. The specimens of L. martini in the Hotwells Limestone however, are of the D₁ form, having a thicker columella and longer, thicker major septa than the S₂ form (see Whittard and Smith 1943, pl. 15, fig. 3 (D₁ form) and fig 4 (S₂ form))

Relationship of the mapped formations to the Avonian zones

It is now recognized that most of the nominal genera and index fossils by which Vaughan originally defined the Avonian zones are of only local value and used in a wider context may be inappropriate or even misleading. Later workers have tended to retain the zonal symbols whilst redefining the zones (and subzones) in terms of wider faunal assemblages and commonly using different diagnostic fossils from those employed by Vaughan in the Avon Gorge (George 1958, pp. 231–4).

The Lower Limestone Shale is conterminous with Vaughan's Cleistopora (K) Zone both here and throughout the Bristol area. Vaughan (1905, pp. 188–90; 1906, p. 101) subdivided the K Zone of the Avon Gorge into K_m (= Modiola or M Zone), K₁ and K₂ subzones. These subdivisions have not been substantiated throughout the South West Province (George 1952, p. 36).

At Burrington, the mudstones near the base of the formation contain a Modiola–ostracod fauna comparable with Vaughan's K_m Subzone, but they are interbedded with limestones containing a typical K fauna and are without zonal significance. Vaughan's K₁ index fossil, Productus bassus Vaughan, has been found only in the upper part of the Lower Limestone Shale at Burrington, and his K₂ index fossil, Spiriferina octoplicata (J. de C. Sowerby), recorded as rare by both Sibly (1905, p. 19) and Reynolds and Vaughan (1911, p. 363), is not present in this collection. The index fossil of the K Zone–Cleistopora aff. geometrica Edwards and Haime sp. (= Vaughania vetus Smyth) is of restricted occurrence but has not been found at Burrington or elsewhere on the Wells Sheet during the present survey, although Hepworth and Stride (1950, p. 135) recorded it from the top 110 ft of the Western Twin Stream adit section.

The Black Rock Limestone in terms of Vaughan's zonal classification of Burrington Combe (Reynolds and Vaughan 1911, pp. 363–7) is equivalent to the Zaphrentis (Z) Zone and the greater part of the Lower Caninia (γ +C₁) Zone. Vaughan subdivided Z into horizon β at the base, Z₁ and Z₂ subzones. The Lower Fauna (see p. 185) corresponds to the Z Zone of Burrington, but there is little justification for subdividing Z, as is shown by a comparison of the assemblages listed in (Table 1) and those given by Vaughan. For example, it is stated (Reynolds and Vaughan 1911, p. 364) that 'Z. delanouei is diagnostic of the level β throughout the South-Western Province', but in the present collection it was found to range nearly to the top of the beds containing the Lower Fauna.

Vaughan (1905, p. 193; 1906, p. 110) erected horizon γ in the Avon Gorge to cover the overlap between the Zaphrentis and Caninia Zones and defined it as the beds in which Caninia cylindrica is, for the first time, abundant' and associated with Zaphrentids. Sibly (1906, pp. 330–1) used it in this sense in the Mendips. Subsequently, in Burrington Combe, Reynolds and Vaughan (1911, p. 366) redefined γ as marked by the incoming of large vesicular Caniniids, with C. patulum as the diagnostic fossil. As Sibly pointed out (discussion in Reynolds and Vaughan 1911, p. 392) this involved a great expansion of the C Zone at the expense of the Z Zone. The Middle Fauna (see p. 186) with C. patulum corresponds to Vaughan's redefined y while the lower part of the Upper Fauna with C. cylindrica corresponds to Sibly's γ. Elsewhere in the Mendips, however, the ranges of C. patulum and C. cylindrica overlap (Bamber 1924; Bush 1925). The distribution and ranges of these Caniniids needs to be studied in more detail before their relative zonal merits can be assessed. The redefined y was taken as the basal part of the Lower Caninia (C₁) Zone (Reynolds and Vaughan 1911, p. 366). The present study supports this grouping of Horizon γ with C₁ rather than the original grouping with the Z Zone (Vaughan 1905, p. 193).

The beds containing the Upper Fauna of the present account correspond to the lower two-thirds of C₁ (excluding γ) as defined by Reynolds and Vaughan (1911, pp. 366–7) at Burrington Combe and are Tournaisian in aspect. The succeeding dolomites include no stratigraphically significant fossils except for Chonetes (Megachonetes) magnus Rotai near the base (see p. 18) whilst the lowest recognizable Burrington Oolite fauna (36 ft above the base of the formation) includes 'Zaphrentis' konincki aff. kentensis Garwood. Both these forms occur together and in association with an early Viséan fauna in the Vallis Limestone of the Priddy–Binegar area, which is the lateral equivalent of the lowest part of the Burrington Oolite at Burrington (see p. 19). This evidence suggests that the Tournaisian/Viséan boundary at Burrington Combe lies at, or possibly even below, the base of the Burrington Oolite. Vaughan equated the base of C₂ with the base of the Visean at Burrington and placed the boundary about 75 ft above the base of the Barrington Oolite as defined in the present account (i.e. 320 ft below the beds in Plumley's Den, Reynolds and Vaughan 1911, p. 348). It would therefore appear that Vaughan placed the base of the Visean, and by implication the base of C₂, too high in the succession.

The Burrington Oolite in terms of Reynolds and Vaughan's section at Burrington includes C₂, S₁ Upper Caninia Zone) and the lowest 130 ft of S₂. Vaughan (1911, p. 370) defined the lower and upper limits of the S₁ Subzone by the entry points of Lithostrotion and Davidsonina carbonaria (McCoy) respectively thus giving the subzone a thickness of 130 ft. The present survey gives the entry points of these species at 425 ft and 612 ft respectively above the base of the Barrington Oolite, about 15 ft and 60 ft^‡40  higher in the succession. The presence of Carcinophyllum mendipense Sibly has locally been taken as diagnostic of the base of S₁ in the Mendips (Sibly 1906, p. 333; Welch 1929, p. 48). Field observations during the survey of the Wells (280) Sheet show that although the entry points of Lithostrotion and Carcinophyllum mendipense lie within the upper third of the Burrington Oolite, local variations are such as to make the subzonal boundary based on their entry points meaningless. Thus at a large disused quarry only about a mile east of Burrington Combe and in an identical succession, D. carbonaria and L. martini both apparently make their first appearance in the top 15 ft of the Burrington Oolite. It is unlikely that in the distance between Burrington Combe and this quarry the Si Subzone has diminished in thickness from 187 ft (130 ft according to Vaughan) to a few feet.

The Clifton Down Limestone in the Wells Sheet generally is approximately the equivalent of the [main] Seminula (S₂) Zone as used by Reynolds and Vaughan (1911, p. 370) at Burrington; but the entry point of D. earbonaria, which defines the base of the zone, is at Burrington 68 ft below the base of the formation. Field observations over the remainder of Sheet 280 confirm that the entry point of D. carbonaria lies at about, or a little below, the base of the Clifton Down Limestone. It will be noted that this definitive use of D. carbon-aria results in the inclusion of much of what Sibly regarded as S₁ into S₂ (see Reynolds and Vaughan 1911, pp. 370, 392; Sibly 1906, p. 357).

The Hotwells Limestone is synonymous with the Dibunophyllum (D) Zone (Vaughan 1905, pp. 197–9). For discussion of D zone see p. 21.

References

BAMBER, A. E. 1924. The Avonian of the Western Mendips from the Cheddar Valley to the Sea. Proc. Bristol Nat. Soc. (4), 6, 75–91.

BUSU, G. E. 1925. The Avonian Succession of Spring Gardens and Vallis Vale, Frome, Somerset. Proc. Bristol Nat. Soc. (4), 6, 250–59.

CUMMINGS, R. H. 1958. The faunal analysis and stratigraphic application of Upper Palaeozoic smaller foraminifera. Micropaleontology, 4, No. 1,1–24.

GEORGE, T. N. 1952. Tournaisian Facies in Britain. Rep. Int. Geol. Cong. 18, G.B., pt. 10,34–41.

GEORGE, T. N. 1958. Lower Carboniferous Palaeogeography of the British Isles. Proc. Yorks, Geol. Soc., 31,227–318.

HEPWORTH, J. V. and STRIDE, A. H. 1950. A sequence from the Old Red Sandstone to Lower Carboniferous, near Burrington, Somerset. Proc. Bristol Nat. Soc., 28,135–8.

HUDSON, R. G. S. and MITCHELL, G. H. 1937. The Carboniferous Geology of the Skipton Anticline. Sum. Prog. Geol. Surv. for 1935, pt. 2,1–45.

REYNOLDS, S. H. and VAUGHAN, A. 1911. Faunal and Lithological Sequence in the Carboniferous Limestone Series (Avonian) of Burrington Combe (Somerset). Quart. J. Geol. Soc., 67,342–92.

SIBLY, T. F. 1905. The Carboniferous Limestone of Burrington Combe. Proc. Bristol Nat. Soc. (4), 1,14–21.

SIBLY, T. F. 1906. The Carboniferous Limestone (Avonian) of the Mendips Area. Quart. J. Geol. Soc., 62,324–80.

SMITH, S. 1925. Notes upon the small species of Chonetes found in the Lower Carboniferous around Bristol. Geol. Mag., 62,85–8.

VAUGHAN, A. 1905. The Palaeontological sequence in the Carboniferous Limestone of the Bristol Area. Quart. J. Geol. Soc., 61,181–305.

VAUGHAN, A. 1906. The Carboniferous Limestone Series (Avonian) of the Avon Gorge. Proc. Bristol Nat. Soc. (4), 1,74–168.

VAUGHAN, A. 1911. See Reynolds, S. H. and Vaughan, A. 1911.

WELCH, F. B. A. 1929. The Geological Structure of the Central Mendips. Quart. J. Geol. Soc., 85,45–76.

WHITTARD, W. F. and SMITH, S. 1943. Geology of a recent Borehole at Filton, Glos. Proc. Bristol. Nat. Soc. (4), 9,434–50.

Appendix 2 Records of some important boreholes

The following is a selection of important records, hitherto unpublished, of boreholes put down for water in the area of 1-in Wells (280) Sheet. The sites are listed alphabetically under towns and villages.

Banwell

Banwell Moor Borehole

Site, 1900 yd N. 2° E. of Banwell church. Nat. Grid Ref. [ST 3995 6087]. 6-in sheet ST/36 SE. Height above O.D. 19 ft Cored boring for Weston-super-Mare U.D.C., drilled in 1929. Information from Mr. H. A. Brown, Surveyor. Coal Measures generalized, remainder of account only slightly abridged.

Banwell Spring Borehole

Site, 100 yd N.W. of Banwell church. Nat. Grid Ref. [ST 3985 5918]. 6-in sheet ST/35 NE. Height above O.D. 19 ft. Bored for Weston-super-Mare U.D.C. in 1931. Cores examined by F.B.A. Welch.

Bishop Sutton

Site, 200 yd S.E. of White Cross. Nat. Grid Ref. [ST 5916 5840]. 6-in sheet ST/55 NE. Height above O.D. 520 ft. Boring for Mr. W. Harris by J. Wallis Titt & Co. in 1960. Percussion boring, drillers account. Classification by G. W. Green.

Cheddar

Site, 2320 yd N. 28° W. of Cheddar church. Nat. Grid Ref. [ST 4492 5499]. 6-in sheet ST/45 SW. Height above O.D. 194 ft. Boring for Callow Rock Lime Co. Ltd. by F. G. Clements (Easton) Ltd. in 1939. Drillers account.

Site, 1350 yd S. 34° E. of Cheddar church. Nat. Grid. Ref. ST/46645202. 6-in sheet ST/45 SE. Height above O.D. 60 ft. Boring for Mr. A. Bazley by F. G. Clements (Easton) Ltd. in 1963. Percussion boring to 100 ft, cored thereafter. Cores examined by D. E. White and D. B. Courtier.

Draycott

Site, 520 yd W. 40° N. of Draycott church. Nat. Grid Ref. [ST 4727 5159] 6-in sheet ST/45 SE. Height above O.D. 103 ft. Boring for Mr. Sheldon by F. G. Clements (Easton) Ltd. in 1961. Percussion boring to 100 ft, cored thereafter. Cores examined by G. W. Green.

Other boreholes, approximately 0.25 and 0.5 miles to the north (ST 4722 5192], [ST 4714 5245]) were drilled to depths of 170 ft and 125 ft in Triassic deposits, mainly conglomerate, and a third quarter mile to the south-east ([ST 480 510]) to 242 ft.

East Harptree

Site, 950 yd S. 27° W. of Castle of Comfort Inn. Nat. Grid Ref. [ST 5394 5241]. 6-in sheet ST/55 SW. Height above O.D. 920 ft. Boring for Clutton R.D.C. by F. G. Clements (Easton) Ltd. in 1942. Information from Dr. F. S. Wallis.

Hill Grange Borehole

Site, 500 yd N. 37° W. of Castle of Comfort Inn. Nat. Grid Ref. [ST 5407 5356]. 6-in sheet ST/55 SW. Boring for Mr. Hosegood by F. G. Clements (Easton) Ltd. in 1941. Drillers description.

Jumbled cores, examined by G. W. Green in 1949, consisted of 200 ft of massive red conglomerate and 2.5 ft of grey granular limestone with Lithostrotion ( ?Hotwells Limestone).

Castle of Comfort Borehole

Site, 75 yd N.N.W. of Castle of Comfort Inn. Nat. Grid Ref. [ST 5430 5327]. 6-in sheet ST/55 SW. Boring for Mr. Hosegood by G. Clements (Easton) Ltd. in 1941.

Jumbled cores from below the top 18.5 ft examined by G. W. Green, consisted of 22ft of Downside Stone and 125 ft of Clifton Down Limestone (30 ft upper calcite-mudstone beds, 95 ft Lithostrotion beds).

Site, 1350 yd E. 26° N. of Castle of Comfort Inn. Nat. Grid Ref. [ST 5545 5374]. 6-in sheet ST/55 SE. Boring for Mr. J. Masters by F. G. Clements (Easton) Ltd. in 1948. Top 20 ft percussion boring, remainder cored. Cores examined by F.W. Green.

Shepton Mallet (Darshill)

Site, 380 yd W. 4° S. of Bowlish cross-roads. Nat. Grid Ref. [ST 6090 4396]. 6-in sheet ST/64 SW. Height above O.D. 380 ft. Boring for Messrs. Dennett Bros. by J. Wallis Titt & Co. in 1960. Depths according to driller; cores examined by and information from Prof. D. T. Donovan.

Wells

Athletic Ground No. 2 Borehole

Site, 1200 yd. S. 33° W. of Wells Cathedral. Nat. Grid Ref. [ST 5455 4497]. 6-in sheet ST/54 SW. Height above O.D. 101 ft. Boring for Wells City by F. G. Clements (Easton) Ltd. in 1953–4. Top 83 ft percussion boring, remainder cored. Cores examined by G. W. Green.

No. 1 Borehole was drilled to a depth of 202 ft in 1928–9, 150 yards to the north-north-east of No. 2 Borehole. The succession, recorded by Mr. H. E. Balch, was similar to the above but lay 16 ft higher above O.D. Celestine nodules were reported from a horizon equivalent to that at 173 to 178.5 ft in No. 2 Borehole.

Vowles Borehole, Silver Street

Site, 630 yd S.S.W. of Wells Cathedral. Nat. Grid Ref. [ST 5498 4544]. 6-in sheet ST/54 NW. Height above O.D. 129 ft. Boring for Messrs. Vowles by F. G. Clements (Easton) Ltd. in 1937. Dug well to 21 ft, percussion boring 21 ft to 65 ft, cored thereafter. Cores examined by Mr. H. E. Balch on whose description the following summary is based.

Rookham Borehole

Site 2540 yd. N. 15° W. of Wells Cathedral. Nat. Grid Ref. [ST 5458 4813]. 6-in sheet ST/54 NW. Height above O.D. 514 ft. Cored boring for Wells City by F. G. Clements (Easton) Ltd. down to 206 ft in 1924–5 and deepened by Le Grand, Sutcliffe and Gel Ltd. in 1927 to 574 ft. Cores below 206 ft examined and described by Mr. H. E. Balch. Particulars communicated by Bristol Waterworks Co. The ensuing account classified and slightly abridged by G. W. Greenafter Mr. H. E. Balch.

Comment: Of the top 207.5 ft it is probable that the uppermost 60 ft is Black Rock Limestone. Cores from depths of 361 ft and about 550 ft were recently recovered from the site by Mr. R. W. Jones (Bristol Waterworks Co.). M. Mitchell identified Camarotoechia aff. mitcheldeanensis, and Cleiothyridina cf. royssii from the higher core. Dr. R. Dearnley reports on a thin rock slice (E30096) from the lower core as follows: 'A pale grey dolomitic fine-grained sandstone with thin carbonaceous wisps. Consists mainly of subangular grains of quartz (0.10 mm) and a few fragments of chert, perthitic feldspar, albitic plagioclase, quartzite, recrystallized rhyolite and granophyre. Detrital epidote occurs. Euhedra of dolomite and some poikilitic dolomite grains make up about half of the rock.' He further comments that similar rock types are known from the Old Red Sandstone (Portishead Beds) of the Portishead–Kingsweston area (Sheet 264). In the hand specimen the rock is traversed by numerous tiny calcite veins with traces of galena.

West Harptree

Keat's Cottage Borehole

Site 1350 yd N. 12° W. of Castle of Comfort Inn. Nat. Grid Ref. [ST 5410 5444]. 6-in sheet ST/55 SW. Height above O.D. 850 ft. Bored for Mr. Hosegood by F. G. Clements (Easton) Ltd. in 1942. Lowest 96 ft of core examined by G. W. Green.

Appendix 3 List of Geological Survey photographs one-inch Wells (280) sheet

Copies of these photographs are deposited for reference in the Library of the Geological Survey, South Kensington, London, S.W.7. Prints and lantern slides can be supplied at a fixed tariff.

In addition to the above, views of Cheddar Gorge (A9756-A9757) are available in colour as well as black and white.

Figures, plates and tables

(Figure 1) Sketch-map showing probable distribution of exposed and concealed Palaeozoic formations in the Wells area and the surrounding country.

(Figure 2) Sketch-maps showing isopachytes of (A) Tea Green Marl (Triassic) and (B) Clifton Down Group (Carboniferous Limestone Series).

(Figure 3) Comparative vertical sections showing the lithological variations of the Clifton Down Group of the Carboniferous Limestone Series.

(Figure 4) Sketch-map of the Geology of the Carboniferous rocks of the Cheddar Gorge district.

(Figure 5) Sketch-map of the Geology of the Ebbor Rocks area, near Wookey Hole.

(Figure 6) Generalized vertical sections in the Coal Measures.