Content and licensingview original scan buy a printed copy

Geology of the South Wales Coalfield, Part V, the country around Merthyr Tydfil. Memoir for 1:50 000 geological sheet 231 (England and Wales)

By W. J. Barclay, K. Taylor and L. P. Thomas

Bibliographical reference: Barclay, W. J., Taylor, K., and Thomas, L. P. 1988. Geology of the South Wales Coalfield, Part V, the country around Merthyr Tydfil (3rd edition). Mem. Br. Geol. Surv., Sheet 231, (England and Wales).

British Geological Survey

Geology of the South Wales Coalfield, Part V, the country around Merthyr Tydfil Memoir for 1:50 000 geological sheet 231 (England and Wales)

Authors: W. J. Barclay, BSc British Geological Survey, Keyworth. K. Taylor, BSc, and L. P. Thomas, BSc, PhD Formerly of British Geological Survey

London: Her Majesty's Stationery Office 1988. © Crown copyright 1988. ISBN 0 11 884424 5. Printed in the United Kingdom for Her Majesty's Stationery Office Dd 0240424 6/88 C20 3933 12521

First published 1904 Second edition 1933 Third edition 1988

Other publications of the Survey dealing with this district and adjoining districts

Books

Maps

Preface

Preface to the Third Edition

The district was originally surveyed on the one-inch scale by H. T. De La Beche, W. E. Logan and D. H. Williams, their results being published as Old Series maps 42 SW and 36 around 1845. The primary survey on the six-inch scale was carried out by A. Strahan, T. C. Cantrill and W. Gibson and published in 1900, with an explanatory memoir following in 1904. Revision by T. Robertson between 1927 and 1930 led to a second edition of the memoir in 1933, incorporating much data on the Westphalian sequence. The resurvey was by K. Taylor, L. P. Thomas, G. W. Green, H. C. Squirrell and I. H. S. Hall between 1966 and 1972. Solid and Drift editions of the 1:50 000 map were published in 1979. This short memoir summarises the geology of the district. It was mostly written by W. J. Barclay and is based on a manuscript prepared by K. Taylor and L. P. Thomas prior to their retirement and resignation respectively. A. W. Woodland was responsible for overall editing. For local details the reader is referred to the maps listed in Appendix 1. Carboniferous faunas, full details of which are held in the Biostratigraphy Research Group, BGS, Keyworth, were identified by M. Mitchell (Dinantian macro-fauna), M. J. Reynolds (Dinantian conodonts), J. Pattison (Namurian macrofauna) and N. J. Riley (Namurian and Westphalian macrofauna). Mrs M. Lewis wrote the section on water supply and T. K. Ball provided data on the underground drainage and cave systems.

F. G. Larminie, OBE Director British Geological Survey Keyworth Nottingham NG12 5GG 14th February 1988

Preface to the Second Edition

Revision of this memoir was rendered necessary on the exhaustion of the first edition, by the general advance in geological research and the progress of mining in the district. The re-examination of the ground which was needful for this purpose was carried out between 1927 and 1930 by Dr. T. Robertson, who acted at first under the direction of the late Mr. T. C. Cantrill, and later under that of Mr. J. A. Howe. The present volume has been edited by Dr. Bernard Smith.

The extensive stretch of Old Red Sandstone lying north of the Coalfield has received little revision apart from the checking in the field of the description given in the first edition. With respect to the Carboniferous Limestone, the progress of zonal research elsewhere, and of work in the western part of the district by Dr. T. N. George, has been of help in the detection of the development of an important unconformity within the Limestone Series, and in the elucidation of the sedimentary sequence. The succession in these rocks agrees well with that found in South Wales. The unconformity at the base of the Millstone Grit has been traced westwards from the adjoining Abergavenny district. The sequence in the Shale Group of the Millstone Grit has been studied in the light of information yielded by goniatites and other mineral shells.

The Coal Measures sequence has received considerable attention and the variations of the different parts of the succession from place to place have been studied in detail. In addition, much useful palaeontological material aiding in correlation with more distant parts of the Coalfield has been obtained.

Dr. R. Crookall has studied the Coal Measures flora, an account of which forms Chapter X.

In the chapter on the geological structure the inter-relationship of the various fractures and folds is considered in the light of much underground information which was not available when the original account was written. The chapter on mineral products and water-supply has been considerably enlarged.

We are especially indebted to Mr. W. D. Ware of Ystradgynlais for information, helpful discussion, and carefully collected material.

For identification of fossils and assistance in palaeontological questions we are indebted to Drs. J. Pringle and C. J. Stubblefield, Prof. A. E. Trueman, Mr. W. S. Bisat and Prof. D. M. S. Watson. The photographs were taken by Mr. J. Rhodes. Acknowledgement of the assistance given by the mining firms in the district will be found on p.82.

The published six-inch maps included in the list on p. xii have received some revision, and new editions of most of these have been issued.

John S. Flett, Director Geological Survey Office, 28, Jermyn Street, London, S.W.1. 1st April, 1932.

Preface to the First Edition

The country around Merthyr Tydfil, which is described in this, the fifth part of the Memoir on the Geology of the South Wales Coal Field, is illustrated in Sheet 231 of the New Series one-inch map. The Sheet includes a small tract of the Pennant plateau, and the part of the North Crop of the Coal Field which extends from Dowlais to Swansea Vale. The northern half of the map is occupied by the scarps of the Lower Carboniferous Rocks, and by the long dip-slopes of Old Red Sandstone which form the Fforest Fawr and lead up to the Brecknock Fans.

The original survey was made by Sir H. T. De la Beche, Sir W. E. Logan, and Mr. D. H. Williams, on the Old Series One-inch Maps 42 S.W. and 36. The maps are undated, but appear to have been published about 1845.

The re-survey was made on the six-inch scale under the superintendence of Mr. Strahan, and was published in 1900. The south-eastern part of the map was surveyed by Mr. Gibson, the south-western part by Mr. Strahan and the northern part by Mr. Cantrill. The present volume includes the descriptions furnished by each geologist of the area surveyed by himself, and has been edited by Mr. Strahan.

The Old Red Sandstone includes an upper and impersistent sub-division of grey grits and conglomerates, which is developed round parts of the Carboniferous tract. The grits pass up into the Carboniferous Limestone and are known from fossil evidence obtained further east in Sheet 232, to be of Upper Old Red Sandstone age. The great mass of Brownstones lying below the grey strata has yielded no fossils in this area, and is doubtfully referred to the Lower Old Red Sandstone. The lowest strata exposed are some green and red sandstones and marls which have been named after the Senni Valley, where they were first distinguished.

The Carboniferous Limestone Series includes the Lower Limestone Shales below the main mass of limestone, and a thin impersistent set of shales with thin dark limestones above that mass. The dark limestones have weathered at the outcrop into rottenstone, and have been dug for that material in the Tawe Valley and west of it.

Three sub-divisions are made in the Millstone Grit, the most conspicuous member being a massive quartz-conglomerate at the base of the formation. The fossils yielded by some shaly bands show that marine conditions prevailed during the deposition of strata of Coal Measure aspect, but the marine bands observed in the Lower Coal Series of Monmouthshire (Sheet 232) have not been detected in this part of the Coal Field. Upwards the Millstone Grit graduates into the Coal Measures, and upper and lower limits of the Farewell Rock being ill-defined.

The part of the Coal Field included in Sheet 231 illustrates the change in the seams from the condition of steam-coal, as worked near Dowlais and Aberdare, to anthracite, as worked in and west of the Vale of Neath. The change is gradual and affects all the seams—the lower coals being the first to show it—here as in other parts of the Coal Field.

The base of the Pennant is drawn arbitrarily at the Rhondda No. 2 Seam, which becomes the Ynysarwed Seam, west of the Neath. There are massive sandstones, however, below that seam, which correspond to the Llynfi Rock of the Maesteg district.

In addition to a large number of normal faults running in the N.N.W. direction, which prevails through the whole Coal Field, two great belts of disturbance, known as the Cri-barth and Neath Disturbances, traverse this region from E.N.E. to W.S.W. They form the forerunners of a long series of similar disturbances, which increase in frequency and intensity towards the north-west. All have exercised a marked control upon the river-system, a fact which is well illustrated in the relation of the Tawe and Neath Valleys to the two disturbances named above.

The glacial phenomena shew, both in the distribution of the drifts and in the direction of numerous glacial stria;, that the ice generated on the Old Red Sandstone plateau moved in a general southward direction, surmounting the scarps of the Carboniferous Limestone and Millstone Grit, and crossing deep valleys with little or no deflection. It failed, however, to overtop the great Pennant scarp of Cam Mosyn [Foesen], but split against it, one part of it moving down the Cynon and Taff valleys in the direction somewhat east of south, while another part turned south-westwards down the Neath and Tawe. Some morainic ridges, lying at the foot of the bolder precipices, appear to have been the last work of the retreating ice-sheet. Several enclose tarns.

The principal economic products, with the exception of the coals, which will be separately described, are summarised in Chapter VIII [now IX].

The Map is published in two editions, on one of which (the Solid Edition) glacial deposits are omitted, while on the other (the Superficial Edition), those deposits are shewn by colour, as well as the solid formations where they are not concealed by drift. The six-inch maps illustrative of the Coal Field are in course of publication. MS. copies, both of these and of the maps lying outside the Coal Field, are deposited in this office, where they can be consulted. Copies can be obtained at cost price.

A Geological Bibliography of South Wales and Monmouthshire will be found in "The Country around Cardiff" (Part III of the Memoir on the Geology of the South Wales Coal Field, 1902.)

J. J. H. Teall Director Geological Survey Office 28, Jermyn Street, London. 20th April, 1904.

Summary of geology

This memoir gives a summary of the geology of the district between Merthyr Tydfil in the east and the upper reaches of the Neath and Swansea valleys in the west. The area is part of the North Crop of the South Wales Coalfield, extending northwards from the coalfield escarpment to the Brecon Beacons. 'Merthyr Tydfil was a major industrial centre in the late eighteenth and early nineteenth centuries, its prosperity based on iron-making using locally derived ironstone, coal and limestone.

The oldest exposed rocks are the Devonian Senni Beds, characteristically green sandstones well known for their early vascular plant remains. The succeeding Brownstones, Plateau Beds and Grey Grits occupy a large tract of fine upland country in the north of the district.

The Carboniferous rocks comprise the Dinantian Carboniferous Limestone, a carbonate sequence into which rivers disappear to flow in underground cave systems like that at Dan-yr-Ogof, the Namurian Millstone Grit, spectacularly exposed in the ravines of the upper Neath, and the Westphalian Coal Measures, the coals of which have been much exploited by shallow and deep mining and by opencast working.

The main structural elements of the district are the Neath and Cribarth disturbances, fundamental basement fractures with a prolonged history of movement. Minor, but intense disturbances in the Coal Measures in the west were an important factor in the decline of deep mining of anthracite, machines being unable to cope with the disturbed coals.

In the late Devensian, ice accumulated in the Brecon Beacons, flowed south and funnelled into the valley glaciers which, on melting, left a range of glacial, glaciofluvial and glaciolacustrine deposits. In the periglacial climate that followed, most of the mass movements of the valley sides took place, and the resulting landslips pose engineering and environmental problems in the coalfield valleys today.

Geological succession in the Merthyr Tydfil district (P945450)

(Frontispiece) Scwd-isaf Clyn-gwyn Waterfall in Namurian Basal Grit. [SN 9236 1058] A11981

Chapter 1 Introduction

The southernmost third of the district comprises the central part of the North Crop of the South Wales Coalfield between the Swansea Valley and Merthyr Tydfil. In the north the mountainous tract of Fforest Fawr, marking the outcrop of Old Red Sandstone arenaceous rocks, culminates in the Brecon Beacons, the high summits of which lie mostly just outside the district. Dinantian limestones and Namurian sandstones occupy long gentle dip slopes between these two areas (Plate 1).

The Brownstones form the highest ground, much of Fforest Fawr reaching more than 500 m, with Fan Fawr itself 734 m. The Basal Grit forms almost equally high ground, reaching 558 m on Carreg-goch, west of the Tawe, and 569 m on Cefn yr Ystrad, east of Taf Fechan. The Pennant plateau on the southern margin is generally well over 400 m, and reaches its highest point within the district just south of Llyn Fach at 600 m exactly. The drainage is to the south, and is dominated by the rivers Tawe, Neath, Cynon and Taff.

The population is largely confined to the coalfield, with Merthyr Tydfil and Aberdare the biggest towns, and almost continuous ribbon development along the valley bottoms housing what was formerly mainly a mining population. Dowlais and Hirwaun were early centres of growth because of their proximity to the ironstone, coal and limestone on which the early iron industry was based. Dowlais was the country's major iron-producing centre in the early part of the nineteenth century. The coal mining industry, producing bituminous and steam coals in the east and anthracite in the west, reached maximum production just before the First World War. A rapid decline in recent years has left Tower Colliery at Hirwaun the only major unit still working.

Small private mines have increased in number of late, and forty were in existence in 1986. Opencast mining has also increased in importance since the Second World War with a number of major sites working the main coals along their crop, including Maesgwyn, which produced over seven million tons of coal and was for a time the largest open pit in Britain.

The district is one of great scenic beauty. The view northwards from the top of the Pennant escarpment south of Llyn Fawr and Llyn Fach is among the most spectacular in Wales, and the valley scenery of the upper Neath and its tributaries, the Hepste, Mellte, Pyrddin and Nedd Fechan deserves special mention. Here the rivers flow through narrow ravines in the Millstone Grit, and alternations of sandstones and mudstones have produced a series of beautiful waterfalls. There is fine scenery also in the Carboniferous Limestone country and the many miles of underground passages have made the area a mecca for speleologists and cavers.

Chapter 2 Devonian

Devonian rocks crop out over the northern third of the district, forming the fine upland scenery of Fforest Fawr and the Brecon Beacons. They are almost all of continental Old Red Sandstone facies, deposited by southward-flowing rivers draining the rising Caledonian mountain chain to the north. The red marls, which constitute the lowest division of the Old Red Sandstone in the ground to the north, do not crop out within the district, and the lowest beds exposed belong to the Senni Beds; they are succeeded in ascending order by the Brownstones, the Plateau Beds and the Grey Grits (Figure 2). The Senni Beds and Brownstones are of Lower Devonian age and the Plateau Beds and Grey Grit are Upper Devonian. Uplift and erosion during Middle Devonian times are indicated by the unconformity that truncates the Brownstones. Although no major discordance is apparent, the overlying Plateau Beds bear witness to a major transgression during which marine marginal sediments were deposited intermittently. This was followed by a temporary regression accompanied by some erosion, before further fluvial deposits—the Grey Grits—were laid down.

The lithostratigraphical sequence was estabished by Cantrill (in Strahan and others, 1904) who included the Plateau Beds in the Brownstones. The discovery of Upper Devonian marine faunas (Thomas, 1951) and fish (Croft, 1953) in the Plateau Beds showed that a major unconformity separated them from the Brownstones below. Allen (1964) suggested a Frasnian to early Famennian age for the Plateau Beds and noted that they were overlain unconformably by the Grey Grits. Several new fossil localities were found during the resurvey of 1966–72 (Taylor, 1972; Hall and others, 1973; Taylor and Thomas, 1974, 1975).

Cantrill's original account included petrological descriptions of various sandstones in the sequence, and, more recently, sedimentology studies have been carried out by Allen (1965), Lovell (1978). Allen (in Owen (editor) 1974; 1977) has provided stratigraphical summaries.

Senni Beds

Only the upper half of the formation—about 150 m—are seen in the lower parts of the valleys along the northern margin of the district. Green sandstones are dominant and make up a sequence of fining-upwards cycles. The green colour distinguishes the Senni Beds from the markedly red Brownstones above, but dull purple and red-brown tints are also found. The sandstones range from fine to coarse grained and they are characterised by both parallel lamination and cross-bedding, with mica prominent on the bedding-planes. Intraformational debris in the form of clasts of calcrete ('race'), siltstone and siderite, litter the bases of individual sandstones, and lenticular beds of calcified intraformational conglomerate (' cornstone conglomerate') are common. Interbedded green and purplish siltstone and mudstone layers occur, and calcrete nodules are present in many of them. The argillaceous beds cap the sedimentary cycles and are commonly erosively truncated by overlying sandstone or conglomerate at the base of the succeeding cycle. Loeffler and Thomas (1980) interpreted the Senni Beds as a relatively high discharge, mixed-load, sand-dominated braided stream complex. The conglomerates and sandstones represent in-channel deposits, and the finer-grained lithologies overbank and floodplain deposits. Vascular plants flourished on the shores of ephemeral ponds and lakes, on river banks and in abandoned channels.

The Brecon Beacons (Storey Arms) quarry (9715 2084) is a classic plant locality (Croft and Lang, 1942; Edwards and Richardson in Friend and Williams, 1978; Heard, 1927). Amongst a varied flora, Gosslingia breconensis was first described from here, and Edwards and Richardson gave a detailed description of the quarry, its macroflora and miospore assemblages.

No fish remains have yet been found within the district, but a new pteraspidid ostracoderm, Althaspis senniensis (Loeffler and Thomas), has been recorded in Heol Senni quarry [SN 9145 2210] just to the north (Loeffler and Thomas, 1980). The presence of Rhinopteraspis cornubica (McCoy) in the Abergavenny district to the east (White, 1938), as well as a diverse miospore assemblage at Heol Senni (Edwards and others in Friend and Williams (editors), 1978) suggest a middle to late Pragian age for the Senni Beds.

The beds are exposed in the Tawe Valley [SN 853 205], in Nant Gihirych [SN 889 207], in tributaries of the River Senni, in Nant y Gaseg, Nant y Gwair and Nant Mawr, all tributaries of the River Dringarth, and in the Storey Arms quarry (see above).

Brownstones

The Brownstones crop out over a wide area of the upland Beacons. They are red-brown and purple sandstones with terbedded red siltstones and mudstones, in all about 400 m thick. Intraformational conglomerates occur locally, particularly in the west of the district, but they are less common than in the Senni Beds; calcrete nodules are much less common. Extraformational pebble beds occur sporadically.

The junction with the Senni Beds is generally transitional, being simply a colour change from green to red which takes place over several metres of beds of alternating green and red. The sandstones are mainly fine to medium grained and commonly richly micaceous. Coarse, pebbly, conglomeratic sandstones occur at the base and in the middle of the formation. The argillaceous layers are generally red or mottled red and green, and show evidence of bioturbation.

Tunbridge (1981) has described the sedimentology of the Brownstones in the Brecon Beacons. He recognised five main facies, arranged in three types of facies-sequence. The individual facies are: single-bed fine- to medium-grained sandstone sheets; multistorey channel-form fine- to medium-grained sandstones; intraformational carbonate conglomerate of rare occurrence, but forming minor channel fills at the top of some channel-form sandstone units; massive red siltstone of floodplain or possibly windblown origin; and thin sandstone interbeds, found with the red siltstone facies, and introduced into the flood-basins by crevassing of streams during flooding.

The three facies-sequences represent distal, medial and proximal settings, all marked by an upwards increase in sandstone. The distal and medial sequences are represented in the lower and middle parts of the Brownstones and the sandstone-dominated proximal sequence is confined to the higher part.

The distal sequence comprises alternations of beds of parallel-laminated sheet sandstones, 0.4 to 1.2 m thick and beds of red siltstone, 0.3 to 2.0 m thick, in the ratio 2:3, and appears to be of flood-flat origin. The medial sequence consists of 75 per cent of 3 to 5 m thick multistorey sandstone units of cross-cutting channel fills, and 25 per cent of 2 to 4 m thick siltstones and thin sandstones; the sandstones are interpreted as low sinuosity stream deposits and the siltstones as flood deposits, the two interdigitating on an extensive and unconfined alluvial plain. The proximal sequence consists mostly of cross-cutting channel sandstones in multistorey bodies up to 11 m thick with some siltstone layers, 0.3 to 1.0 m thick. It suggests a higher energy environment, with steeper slopes drained by low sinuosity streams.

The Brownstones have hitherto yielded no animal fossils or stratigraphically useful plants or miospores, although rootlets and other plant fragments have been noted. A late Pragian to Emsian age can thus only be suggested from indirect evidence.

The Brownstones are particularly well exposed on the scarp slopes of Fan Hir [SN 835 195], Fan Fawr [SN 972 193], Craig Byllfa [SO 006 199], Craig Fan-ddu [SO 020 188], Craig Cwm Cynwyn [SO 030 206], Craig Cwm-oergwm [SO 038 200], Craig Cwareli [SO 047 200], Craig Fan Las [SO 052 195], Craig Pwllfa [SO 065 205] and Cwar y-Gigfran [SO 064 195]. Other good sections are in Nant y Tywyni [SN 840 181], Nant y-Wydd [SN 8661 1808], where rootlets and other plant remains occur, between Llwyn rhyn [SN 8490 1809] and Nant y Coedcae [SN 8490 1960] (although this section is interrupted by gaps and faults), and in Nant-y-Gloesydd [SO 0372 1820]. Road cuttings provided sections at Cefn Cul [SN 8533 1775] and Bwlch Brynrhudd [SN 8690 1942]. At the former, plant remains, including rootlets, were noted in sandstones, mudstones and in hard lenticular sandstone with intraformational conglomerates; pyrrhotite, malachite and baryte were noted on bedding planes and in joints. Another section occurs at the southern end of Cray Reservoir [SN 880 207].

Plateau Beds

The Plateau Beds rest unconformably on the Brownstones, and slight angular discordance between the two formations can be seen in the scarp faces of Fan Hir [SN 832 203] (Allen, 1964) and Bannau Sir Gaer [SN 813 218]. They vary in thickness from about 24 m in the upper Tawe Valley to a maximum of 58 m in boreholes in the central part of the district, west of the Taf Fawr; about 26 m are present in the east, where the sequence is incomplete. Fish-bearing sandstones and conglomerates occur at several levels. One such conglomeratic layer—the Afon-y-Waen Fish Bed—is 2.0 to 2.4 m thick at its type locality at the confluence of Afon-y-Waen and Nant y Cwrier [SN 9761 1476]; it has been traced westwards across the district, and has been used to divide the Plateau Beds into Upper and Lower divisions shown on the 1:50 000 map. The Afon-y-Waen Fish Bed is well exposed in Nant Mawr [SN 9507 1568]. Afon Llia [SN 9336 1487], Afon Nedd Fechan [SN 9056 1648] and Nant Byfre [SN 8519 1646] and it was also found in the BGS boreholes drilled on Cefn Esgair [SN 9835 1419]; [SN 9845 1353] (Taylor and Thomas, 1975).

The Lower Plateau Beds begin with a red mudstone (1.2 to 1.5 m thick), rich in quartz granules and commonly overlain by a conglomerate. Above occur up to 18 m of purple to red coarse-grained sandstones exhibiting large-scale trough cross-stratification and passing up into finer-grained, thin-bedded and flaggy red sandstones, which thicken locally where cross-stratified channel-fill sandstones are present.

The Upper Plateau Beds comprise a variable lower unit of trough cross-bedded pebbly sandstones and fish-bearing conglomerates with interbedded mudstones up to about 10 m thick, overlain by a more heterogeneous sequence of thinly interbedded red-brown fine-grained sandstones and mudstones, about 7 m thick. Rippled bedding surfaces, desiccation cracks, abundant bioturbation structures and trails, and mudflake-conglomerate lenses are characteristic of these beds. Cross-stratified channel sandstones are common, and both intraformational siderite clast conglomerates and extraformational quartz-pebble conglomerates occur.

A diverse fauna of brachiopods and fish remains has been collected from the Plateau Beds. D. E. Butler (in Taylor and Thomas, 1975) listed the invertebrate fossils, including Lingula spp.,Cyrtospirifer verneuili (Murchison), cf. Ptychomaletoechia omaliusi (Gosselet). Leptodesma cf. lichas Hall and ?Pterinopecten sp.;vertebrate fragments include Holoptychius sp., Bothriolepis sp.,cf.Pseudosauripterus anglicus (A. S. Woodward) and cf. Sauripterus sp.(Hall and others, 1973; Taylor, 1972; Taylor and Thomas, 1974, 1975). Trace fossils include cf. Planolites and the trilobite track Rusophycus. All the marine fossils have been obtained from the uppermost, heterogeneous unit. The brachiopods cf. Ptychomaletoechia omaliusi and Cyrtospirifer verneuili and the bivalve Leptodesma cf.lichas, together with certain of the fish suggest a Frasnian to Famennian age.

The Plateau Beds east of the Taf Fawr are overlain unconformably by the Grey Grits and the highest beds appear to be missing; west of the Taf Fawr a more complete section is preserved. The position of the faulted sequence in Cwm Coedcae Drain is uncertain, but the rocks lithologically resemble the Upper Plateau Beds west of the Taf Fawr.

A fluid mudflow origin has been suggested for the quartzgranule-rich mudstone at the base of the formation. The conglomerates above, most fully developed in the west of the district, are considered to be braided stream deposits (Lovell, 1978). Taylor and Thomas (1975) considered the lower part of the Plateau Beds generally to be fluvial, but Lovell noted that all the cross-bedding in the sandstones was directed towards the north-west and suggested an aeolian origin for much of these beds, with some intercalated fluvial sediments. The uppermost part of the Plateau Beds is marginal-marine (Allen, 1965; Taylor and Thomas, 1975), with some evidence of supratidal, tidal flat and possibly subtidal environments (Lovell, in Friend and Williams, 1978).

Grey Grits

The Grey Grits are grey, greenish and yellow quartzitic sandstones with quartz pebble layers and intraformational mudstone-clast conglomerates and subordinate red, purplish grey and green mudstone layers. The sandstones are generally tabular with sharp interbed contacts; scouring and winnowing is indicated by the presence of intraformational clasts in the bases of beds. Grain-size ranges from fine to medium and individual beds vary in thickness from 0.1 to 1.8 m. Internally, the beds show trough and planar cross-bedding, as well as parallel lamination. A fluvial braided stream origin has been suggested (Allen, 1965; Lovell, in Friend and Williams, 1978).

The Grits attain a maximum thickness of about 19 m in the east of the district and they underlie extensive moorland areas at Cefn Sychbant, Pant-y-Gader, Cefn Car, the southern parts of Waun Lysiog, Cefn Ynys-Fawr and Waun-rydd. West of Afon-y-Waen the Grey Grits are thin or they are absent; locally 1.5 to 1.8 m of green calcareous sandstones are developed and there is some doubt as to their exact correlation; it has been suggested that they may represent an arenaceous development at the base of the Carboniferous Lower Limestone Shale (Lovell, 1978).

The beds are well exposed in the Taf Fechan and Taf Fawr valleys and at Bryniau Gleision (Hall and others, 1973; Taylor and Thomas, 1975). Abercriban Quarry [SO 0638 1275] (Plate 2) provides a complete section (Lovell, in Friend and Williams, 1978).

Chapter 3 Carboniferous: Dinantian

Dinantian (Carboniferous Limestone 'Series') rocks form a distinctive belt of country about 1 to 3 km wide crossing the central northern part of the district, from the uppermost reaches of the Gwys Fawr in the north-west to Llwyn-on Reservoir in the Taf Fawr Valley; here the crop is displaced west-south-westwards about 10 km by the Neath Disturbance (p.32) to Craig y Dinas, near Pont Nedd Fechan, from where it resumes its generally easterly course towards Cefn yr Ystrad on the eastern margin of the district. Several faulted outliers occur along the same disturbance west and east of the Taf Fechan Reservoir. The outcrop, much obscured by glacial drift, forms a series of scarps and dip-slopes between the upland areas of the Old Red Sandstone and the escarpment of the Millstone Grit. Many of the southward-draining streams flow underground for parts of their courses across the limestone outcrop, and large cave systems occur around Dan-yr-Ogof in the Tawe Valley and at Ystradfellte.

The sequence is thickest in the central area, where it reaches some 220 m; about 170 m of beds are present in the west and about 135 m in the east. Two unconformities within the sequence and the unconformable overstep by the overlying Namurian rocks affect the thickness (Figure 3).

The onset of carbonate deposition records a major marine transgression, when the southern margins of the Caledonian massif (St George's Land) were flooded from the south by a shallow shelf sea. The Merthyr Tydfil district lay close to the northern margins of this sea throughout most of the Dinantian, as is shown by the overall thinness of the succession in comparison to the more distal sequences farther to the south, by the predominance of shallow-water limestones, and the number of breaks in the sedimentation.

Following the original six-inch geological survey of 1895–98 (Strahan and others, 1904), Vaughan's (1905) coral-brachiopod zonal scheme was applied to the sequence (George, 1927; Robertson, 1933; Robertson and George, 1929). Robertson (1933, pp.29–33) also interpreted the sequence in terms of major depositional cycles, a theme taken up more recently by Ramsbottom (1973). George (1954)

The classification of the Dinantian rocks of the district is shown in Table 1. The lithostratigraphical nomenclature used in this account follows that of George and others (1976) and its correlation with those shown on the 1979 edition of the 1:50 000 Geological Map and the 1900 edition of the One-inch Map is also given.

Lower Limestone Shale

Over much of the district the Lower Limestone Shale comprises a lower limestone-dominated unit, overlain by an upper shale unit. The limestone thins westwards and is absent west of Carn-yr-onen [SN 884 170]. A thin but widespread conglomerate occurs at the base of the limestone, resting sharply on the Upper Old Red Sandstone. It contains pebbles of quartz, jasper, igneous and metamorphic rocks, black limestone and brown sandstone, as well as intraformational clasts and phosphatised pellets, together with fish and brachiopod fragments. It appears to indicate a break in deposition accompanied by reworking of the topmost Old Red Sandstone sediments during the northward transgression of the sea (Lovell, in Friend and Williams, 1978). The contact between the Old Red Sandstone and the basal limestone in Breconshire is not gradational (Hall and others, 1973; Taylor and Thomas, 1975).

The limestone unit forms a scarp below the flat-lying ground of the overlying shale over most of the central and eastern part of the outcrop, where it is up to 8 m thick. It consists mainly of medium- to thick-bedded oolitic, crinoidal and skeletal calcarenite, generally sandy and capped by a thin micritic limestone and shale. The calcarenite is particularly sandy towards the base, and it passes laterally into grey and greenish calcareous sandstone with honeycomb weathering; quartz pebbles are scattered throughout.

The upper shale division comprises grey mudstones with interbedded thin blue-grey limestones, in all about 15 m thick in the east of the district. Westwards from Carn-yronen it is overstepped by the Dowlais Limestone, and about 9 m are present on the western margin.

The basal calcarenites represent high energy, shallow-water, mixed siliciclastic-carbonate deposition above wave base in barriers and shoals behind which the micritic facies accumulated in restricted intertidal lagoons and flats. The shales represent a second transgressive event, introducing deeper-water conditions, in which muddy sedimentation prevailed but was interrupted by limestone deposition, perhaps due to storms (Burchette, 1981).

The Lower Limestone Shale is very fossiliferous. It contains brachiopods including Cleiothyridina royssii (Davidson), Eumetria carbonaria (Davidson), Leptagonia cf. analoga (Phillips), Macropotamorhynchus mitcheldeanensis (Vaughan), Pugilis vaughani (Muir-Wood), Rugosochonetes vaughani Muir-Wood, Syringothyris principalis North and Unispirifer tornacensis (de Koninck), bryozoans, crinoid columnals and bivalves as well as abundant conodonts. The brachiopods and conodonts indicate an early Courceyan age.

Good sections of these rocks can be seen in Nant Hepste Fechan [SN 9861 1395], in the valleys of the Hepste [SN 9698 1347], the Mellte [SN 9358 1413] and the Llia [SN 9338 1450].

Abercriban Oolite

The Abercriban Oolite consists mostly of pale grey oolitic grainstone, with some fine-grained micritic limestone interbeds. Its type locality is Abercriban Quarry, Pontsticill [SO 0667 1244]. About 28 m of beds are present in the east of the district, and these are progressively overstepped westwards to Cader Fawr (3 km west of Llwyn-on Reservoir in the Taf Fawr Valley), and from there westwards by the Dowlais Limestone as far as Carn-yr-onen where it is overstepped completely.

A thin but persistent fine-grained syngenetic dolomite lies at the base of the Abercriban Oolite; it has a gradational junction with the underlying Lower Limestone Shale, and is truncated by an emergent surface. The dolomite is included in the Lower Limestone Shale on the 1:50 000 Geological Map, but it is correlated with the Sychnant Dolomite of the Abergavenny district (Barclay, in press), which is there included in the Abercriban Oolite. The lower and upper oolite leaves of the Abercriban Oolite in the east of the district are equated with the Pwll-y-Cwm Oolite and the Blaen Onnen Oolite of the Abergavenny district. Below the palaeokarstic surface that truncates the Abercriban Oolite the beds, affected by penecontemporaneous dissolution, become increasingly rubbly and disordered upwards, with irregular oolite masses separated by grey-green mudstone films.

The Abercriban Oolite represents high-energy shallow-water oolite-shoal deposition, interrupted by regressive events which led to the deposition of thin lagoonal micritic interbeds, and ultimately to subaerial exposure of the ooliteshoals.

The beds have yielded brachiopods similar to those in strata of Courceyan age elsewhere in South Wales. Conodont assemblages lack any typical early Courceyan forms, thus suggesting a middle to late Courceyan age for much of the formation.

The best sections are in the Mellte valley [SN 9348 1388], at Ogof Fawr sink-hole [SN 9852 0963], in Abercriban Quarry [SO 0667 1244], Baltic Quarry [SO 0652 1170], Odynau Tyle'r-bont Quarry [SO 0630 1117] and Cwar yr Ystrad Quarry [SO 0820 1395].

Llanelly Formation

The Llanelly Formation consists mainly of thin-bedded fine-grained micritic limestones, forming a narrow bench between the broader outcrops of the Abercriban Oolite and the Dowlais Limestone. It is 6 to 8 m thick on the eastern margin and 10 m thick at Ogof Fawr. Progressive westwards overstep by the Dowlais Limestone reduces its thickness west of Ogof Fawr, and it is cut out completely west of the Hepste valley.

A palaeosol complex lies at the base of the formation, draping and filling hollows on the underlying karstic surface of the Abercriban Oolite. Named the Tyle'r-bont Pedocomplex, it consists of a variable sequence of nodular, rubbly and massive calcretes, green and purple clays, limestone conglomerates and thin sandstones. Its thickness ranges from 1 m up to 6 m at its type locality, Odynau Tyle'r-bont Quarry (Wright, 1982). Up to 4 m of oolitic and oncoidal grainstones with clay interbeds overlie the palaeosol complex, and are capped by a brecciated calcrete, the Cwm Dyar Pedoderm (Wright, 1981, 1982). The overlying Penllwyn Oolite is a thin oolitic grainstone with a bed of large oncoids (the 'Uraloporella Bed') at its base (Wright, 1981). The oolite is overlain by the Gilwern Clay, a purple-and green-mottled palaeosol; this in turn is erosively overlain by the Garn Caws Sandstone, a thin quartzite present in the east of the district.

The Llanelly Formation lacks any stratigraphically useful fossils, but the alga Koninckopora inflata (de Koninck) Lee is present in the Abergavenny district, suggesting a Chadian or younger age.

The best sections are seen at Ogof Fawr, Darren Fawr [SO 0194 1040], Baltic Quarry, Odynau Tyle'r-bont Quarry and Cwar yr Ystrad Quarry.

Dowlais Limestone

The Dowlais Limestone is shown on the 1:50 000 Geological Map as the Cil-yr-ychen Limestone, the name used on the adjoining Ammanford sheet to the west. Re-named the Dowlais Limestone by George and others (1976) it equates with the Upper Seminula Zone (S2) of earlier accounts (Robertson, 1933; Robertson and George, 1929; (Table 1)).

This, by far the thickest division of the Dinantian sequence of the Merthyr Tydfil district, makes up the largest part of the limestone outcrop. It contains most of the caves of the district including Dan-yr-ogof and Ogof Ffynnon Ddu. The formation is thickest in the west of the district, where 120 m are present west of the Tawe Valley, and it thins progressively eastwards to about 105 m north of Penderyn and 60 m around Pontsticill. In the extreme east, at Cefn yr Ystrad it is further attenuated by pre-Namurian erosion, which has cut out all of the overlying Dinantian strata (Figure 3). The type-locality is at Morlais Castle c.[SO 054 698], north of Merthyr Tydfil. The limestones have been extensively quarried, especially in the Pontsticill–Vaynor area, at Penderyn and at Penwyllt in the Tawe Valley.

The Dowlais Limestone rests with sharp unconformity on the beds below, overstepping from east to west successive strata ranging from the Llanelly Formation to, in the western half of the district, the Lower Limestone Shale. Although a variety of lithologies are present, the formation is dominated by tabular-bedded dark grey bituminous bioclastic limestones, mostly of the packstone–grainstone grade. Three members have been recognised (Scott, personal communication 1986). The lowest is 30 to 40 m thick and comprises peloidal and oolitic packstones and grainstones, commonly arranged in fining-upwards cycles capped by peritidal micrites, showing evidence of emergence. Calcareous sandstones and sandy limestones are particularly common in the east of the district; elsewhere the basal beds are Shelly and oolitic grainstones with intraformational clasts of oolite and micrite up to cobble-size. A widespread bed with Siphonodendron martini (Milne Edwards and Haime) and Syringopora lies 3 to 8 m above the base. The middle member is 13 to 23 m thick and consists of dark grey peloidal packstones and grainstones interbedded with thin shales carrying productoids and oncolites. The highest member is between 20 and 38 m thick and contains less micrite. Grainstones predominate, giving the rocks a paler colour, and there are also more oolitic and micritic interclasts, and algal remains than in the underlying members.

A variety of depositional environments has been recognised, from subtidal shelf through shelf lagoon to intertidal backswamp. The oolites that form the bases of the shoaling-upwards units may be flood-tida delta or washover deposits, whereas the thicker oolitic limestones may represent high-energy shoal deposition (Scott, personal communication 1986; Wright, 1981).

The formation contains a typical Holkerian fauna, including Linoprotonia corrugatohemispherica (Vaughan); Davidsonina carbonaria (McCoy), although not recorded from the Merthyr Tydfil district, has been found in the ground to the west. Composita ficoidea (Vaughan) is abundant at several levels in shell-packed banks.

The best sections are at Penwyllt Quarry [SN 857 157], Penderyn Quarry [SN 955 090], Vaynor Quarry [SO 037 095], Morlais quarries [SO 054 098] and Twynau Gwynion quarries [SO 065 105].

Penderyn Oolite

The Penderyn Oolite is shown on the 1:50 000 Geological Map as an unnamed oolite forming the lower part of the Llandyfan Limestone. It was named by George and others (1976, p.22) who equated it with the light oolite of Robertson (1933, p.28). The type-locality is Penderyn Quarry, where it is some 13 m thick, though elsewhere within the district it may reach 25 to 33 m.-It is absent on the eastern margin of the district, cut out by the Namurian overstep. It comprises pale grey coarse-grained massive oolite which forms extensive characteristic dip-slope pavements east of Ystradfellte and north of Merthyr Tydfil. Locally, the basal bed of the formation is a calcareous sandstone, and the contact with the Dowlais Limestone may be marked by a thin mudstone or calcrete. A second calcareous sandstone, up to about 3 m thick, can be traced locally about 7 to 9 m above the base of the formation. Both beds have been referred to as 'Honeycombed Sandstone' on account of their distinctive carious weathering (Plate 3). Locally the top of the the Oolite consists of lenticular sandstone, rubbly limestone and mudstones with a well developed bed with productoid shells; in places coal streaks and seatearths are present.

Formerly assigned to the Lower Dibunophyllum (D1) Zone, the formation contains a limited Asbian fauna in which Linoprotonia corrugatohemispherica is abundant.

There are good sections in quarries at Penwyllt, the best of which is at Pen y Pant Quarry [SN 852 154] (where the junction with the Dowlais Limestone is exposed), and at Penderyn Quarry [SN 955 090].

Penwyllt Limestone

The Penwyllt Limestone is shown on the 1:50 000 Geological Map as the upper non-oolitic part of the Llandyfan Limestone. Equivalent to the Middle Dibunophyllum (D2) limestones of Robertson (1933, pp.28–29), it was named by George and others, 1976, p.22). The type-locality is Penwyllt Quarry [SN 856 153].

The formation consists of a stacked cyclic sequence of dark grey limestones in which shelly and crinoidal grainstones are the major components. The beds are locally sandy, and chert layers and nodules are abundant. Many of the cycles are truncated by subaerial surfaces, below which nodular calcretes ('pseudobreccias') occur. Seatearths and thin coals overlie these surfaces.

The Penwyllt Limestone is absent in the east of the district, being overstepped by Namurian strata 1 km east of Penderyn on the south side of the Neath Disturbance, and at Cader Fawr on the north side. Progressively higher beds appear westwards to Pwll Byfre, where the feather edge of the overlying Upper Limestone Shales lies. Where fully developed, at the type-locality, the Penwyllt Limestone is about 20 m thick.

An abundant and typically Brigantian fauna occurs, including many productoids and the corals Lonsdaleia floriformis (Martin) and Orionastrea sp.

In addition to the type-locality, Pen y Pant Quarry provides a good section.

Upper Limestone Shales

The Upper Limestone Shales are present only in the northwest of the district, being overstepped by the Namurian east of Pwll Byfre. It consists of 2 to 3 m of dark grey thinly bedded impure limestones, micrites and shales. The beds weather to 'rottenstone' at outcrop, a fine siliceous powder that was formerly dug for use as a polishing powder. The old workings extend from Carreg Cadno [SN 8762 1575] westwards to the district boundary, but the outcrop is largely obscured by collapsed Namurian Basal Grit debris.

There are few good sections: the only one west of the Tawe is in Nant y Gwared [SN 8384 1544], while to the east there are small exposures at Penwyllt [SN 8513 1527], in a sink-hole west of Carreg-Lwyd [SN 8594 1523] and at Penderyn Quarry [SN 955 090].

A varied fauna of brachiopods and bivalves was recorded from the Penwyllt area, quite distinct from that of the Penwyllt Limestone. It is assigned to the late Brigantian.

Chapter 4 Carboniferous: Namurian

The Namurian rocks (also referred to on the map as Millstone Grit Series) comprise a paralic sequence of intercalated marine and fluvial deposits. In the previous editions of the memoir the Millstone Grit was divided into three—Basal Grit, Shale Group and Farewell Rock. The last-named division is now placed in the overlying Westphalian Coal Measures, but the first two are retained as informal units indicative of the predominating lithologies. The Basal Grit, which comprises hard quartzitic sandstones with thin mudstones, forms prominent scarps and long dip slopes south of the Carboniferous Limestone outcrop in a strip, up to about 2 km wide, extending from Carreg-lem and Carreg-goch in the west of the district through Cribarth, Pen-y-cae, Carreg-Lwyd and Pant Mawr to Gweunydd Hepste and Garn-ddu, thence, south of the Neath Disturbance from Pontbren Llwyd to Cefn-coed-y-Cymmer and Merthyr Common in the east. The Shale Group, which consists mainly of grey shales and mudstones with subordinate thin sandsones, forms a strip of low-lying drift-covered land about 1 to 1.5 km wide immediately to the south.

(Figure 4) shows generalised vertical sections of the rocks together with their classification, following Ramsbottom and others (1978). Following the primary six-inch geological survey (Strahan and others, 1904), the work of Bisat (1924) in Yorkshire illustrated the value of the goniatites in the correlation of the Millstone Grit Series and Robertson (1933) and Ware (1939) applied Bisat's goniatite zonal scheme to the rocks of the district. Fossil plants were also important in early work (Crookall in Robertson, 1933; Dix, 1933); although long-ranging they have proved useful in local correlation (Jones, 1958; Jones and Owen, 1956). Apart from the earlier work of the Geological Survey, detailed descriptions of the succession in the various headwater tributaries of the River Neath were given by Evans and Jones (1929), Jones and Owen (1956) and by Robertson (1933). Stratigraphical summaries have been provided by George (1970), Jones (1970; in Owen, 1974), Ramsbottom (1978) and Ramsbottom and others (1978). Kelling (in Owen, 1974), has described the sedimentation of the Namurian rocks of South Wales.

The sequence is thickest in the west of the district, where about 260 m of beds are present; there is marked eastwards thinning to about 80 m in the ground north of Merthyr Tydfil. This is brought about partly by attenuation of the Shale Group, but also by successive overstep in the Basal Grit (Figure 4). In the west the basal quartzitic sandstones are of Arnsbergian (E2) age and in the east they are Kinderscoutian (R1). As described in the previous chapter, the Namurian rocks overlie the limestones of the Dinantian unconformably. The unconformity increases in extent eastwards, both by non-deposition of the lower Namurian sediments, as just indicated, and progressive erosion of the higher Dinantian strata. In the west only the basal Pendleian Stage of the Namurian is not represented, and quartzites of Arnsbergian age rest on shales and mudstones of Brigantian age; in the east strata representing the Asbian to Alportian stages are absent and quartzites of Kinderscoutian age directly overlie limestones of Holkerian age. In the east the break may not be due entirely to early Namurian stripping of a complete Dinantian succession, but partly to non-deposition of the latest Dinantian rocks, as suggested by the presence of pebbly rocks of 'Millstone Grit' facies in the topmost limestones (Jones in Owen, 1974; Owen and Jones, 1961; Kelling in Owen, 1974).

After late Dinantian to early Namurian uplift, the later subsidence of the Namurian basin was controlled by differential movements between the relatively stable areas of St George's Land to the north and the Usk Axis to the east. Superimposed on this structural control was a pulsed eastwards expansion of the basin (Jones in Owen, 1974; Ramsbottom, 1978). Ramsbottom interpreted the pulsed expansion in terms of transgressive events caused by eustatic rises of sea level. From examination of the goniatite faunas of the various discrete marine bands, he further concluded that not only were some of the major depositional cycles (mesothems) of the deep basinal Namurian sequences absent in the Merthyr Tydfil district, but that some of those present are incomplete, with the basal mesothemic boundaries representing unconformities (Figure 4). However, as he pointed out, this conclusion is as yet unsupported by any evidence of the age of the sandstones.

Basal Grit

The Basal Grit consists mainly of orthoquartzitic sandstones with thin mudstone interbeds. Up to 150 m of beds are present in the west, but only some 35 m in the east. The sandstones are commonly pebbly and conglomeratic, with pebbles consisting mainly of vein quartz. Locally certain beds comprise very pure quartzite and these have been extensively quarried for the making of refractory bricks near Dowlais, near Penderyn, at Dinas Quarries in the Neath Valley, and at Penwyllt.

Coarsening-upwards cycles within the Basal Grit are best recognised in the west of the district, where they comprise fine-grained clean orthoquartzite passing up into coarser dirtier orthoquartzites and conglomerates. Individual sandstone sequences may be separated by coarsening-upwards marine mudstones, commonly with a coal streak at the base (Robertson, 1933, pp.54–55). This inverse grading of the sandstones has been compared to that in modern beach and barrier sands (Kelling, in Owen, 1974). Towards the east sandstones become increasingly fluvial, with channelled bases, fining-upwards motifs, and with much plant and log debris. Some of the thin mudstone interbeds are marine, and record transgressive events when either subsidence exceeded sedimentation or sea level rose eustatically. The mudstones were deposited in shallow water and contain mainly near-shore faunas of the shelly benthonic and Lingula biofacies (Jones in Owen, 1974).

The Basal Grit forms extensive glaciated pavements in the west and centre of the district, and it is on these that most of the well preserved glacial striae are to be seen. There are few good sections, even where the pavements culminate in the escarpments of Carreg-goch and Carreg Cadno, as the beds are much cambered and foundered. The tributaries of the River Neath provide the best sections (Jones and Owen, 1956; Robertson, 1933). In the Mellte Valley a plant-bearing black shale near the base of the Grit near Mellte Bridge [SN 9108 0794] yielded Lyginopteris porubensis (Stur), which suggests an Arnsbergian age (E2) at least (Jones, in Owen, 1974; Ramsbottom and others, 1978). Farther upstream there are good exposures below and at Clyn-gwynisaf Waterfall [SN 9236 1038], although much of the section is inaccessible. About 34 m of quartzitic sandstones are exposed in the ravine of the River Hepste above Scwd yr Eira Waterfall [SN 9288 0999]. The Reticuloceras circumplicatile Marine Band is exposed on the ledge near the foot of the waterfall, as well as 450 m upstream [SN 9333 0994] and in the Coed Hir fault zone [SN 9405 0973]. The River Sychryd provides almost complete exposure of the Basal Grit at the old Dinas Quarries, Pen-cae-drain [SN 9170 0795], where four quartzite horizons were worked for silica rock (Robertson, 1933, p.229), and one particularly pure bed was worked underground from a mine adit.

Shale Group

The Shale Group comprises a sequence of coarsening-upwards partly marine cycles in which mudstones and silty mudstones predominate. Thin rootlet beds and coal smuts cap individual cycles locally. Although not abundant, sandstones do occur, and are thickest in the west. Two sandstones in particular, the Twelve-Foot Sandstone (immediately below the Anthracoceratites Marine Band; Plate 4) and the Cumbriense Quartzite (immediately above the Cancelloceras (Gastrioceras) cumbriense Marine Band) can be traced thoughout most of the district. The group is about 45 m thick in the east of the district, about 80 m in the Neath Valley and upward of 120 m in the far west. A more distal environment than that of the underlying Basal Grit is suggested by the preponderance of argillaceous rocks and the presence of abundant goniatite-pectinoid faunas within the marine bands. Faunal phases within these beds may represent variations in oxygen amounts, water depth and salinity (Archer, 1968; Jones, in Owen, 1974), and the effects of bottom currents. Silty impure limestones occur within some of the marine strata, which are usually succeeded by non-marine mudstones; a persistent bed with Carbonicola occurs above the Cancelloceras (Gastrioceras) cancellatum Marine Band.

There are good sections of the Shale Group in the River Tawe and its tributaries near Ynyswen [SN 8383 1314] to [SN 8391 1294]; [SN 8415 1313] to [SN 8435 1317], and in Nant-y-ffin [SN 8474 1334] to [SN 8500 1321], where the beds are cut by the Henrhyd Fault [SN 8486 1329]. The sections at these localities are similar in thickness to that in the Twrch Valley (Ware, 1939). A reduced thickness occurs to the east in the upper Neath Valley. The beds from the Twelve-Foot Sandstone to the Cancelloceras cumbriense Marine Band are seen in Nant Cwm-byddar, Nant Llech-isaf and Nant-Llech-pella [SN 8637 1408] to [SN 8635 1292]; [SN 8664 1273]; [SN 8684 1275], and in Nant Hir [SN 8948 1305] to [SN 8918 1232]. The beds above the Cancelloceras cumbriense Marine Band are best seen in the base of Henrhyd Waterfall, Nant Lech [SN 8539 1194]. The ravines of the upper Neath and its tributaries also provide excellent sections in the Shale Group (Jones and Owen, 1956; Robertson, 1933). The BGS Gelli Isaf Borehole provided a detailed record of the Group, and there are good exposures in Nant Melyn and Nant Hir [SN 9797 0698] to [SN 9795 0617]; [SN 9930 0781] to [SN 9896 0720], both tributaries of the River Cynon near Hirwaun. Nant Morlais provides an almost continuous section through the Group [SO 0718 0933] to [SO 0715 0907].

Chapter 5 Carboniferous: Westphalian

Westphalian rocks, or Coal Measures, underlie the southern parts of the district. Up to 1100 m of beds are present, the sequence being thickest in the west. Two broad divisions are present: a lower one, predominantly argillaceous, that crops out north of the coalfield escarpment and comprises the Lower, Middle and lower part of the Upper Coal Measures, and an upper arenaceous one that makes up most of the Upper Coal or Pennant Measures, which form the bold coalfield escarpment and the northern rim of the main coalfield plateau.

(Figure 5) provides a generalised vertical section of the rocks and their classification. Most of the Coal Measures strata are non-marine, but distributed sporadically throughout the lower argillaceous division are a number of thin marine mudstones of widespread extent, which are particularly useful in detailed correlation, and have been widely used throughout north-western Europe.

The Lower Coal Measures (Westphalian A) range between 120 and 210 m, and the Middle Coal Measures (Westphalian B and early C) between 210 and 360 m, the greater thicknesses in each formation being in the west. The Upper Coal Measures (Westphalian late C and D) are represented by about 670 m of strata, largely massive Pennant sandstones. They consist mostly of Lower Pennant Measures, the higher beds, extensively developed to the south and west, having been removed by post-Carboniferous erosion. The overall westwards thickening of the succession is in accord with the general depositional pattern of the coalfield basin as a whole (see e.g. Woodland and Evans, 1964, figs. 6,7 and 8; Thomas, in Owen, 1974).

The Lower and Middle Coal Measures were deposited in a paralic setting in a fluvio-deltaic complex fed with sediment from St George's land to the north (Kelling, in Owen, 1974; Owen, 1964). They consist of rhythmic or cyclic alternations of mudstones, siltstones and sandstones, with interbedded clay ironstones, coals and seatearths. The cyclic, mainly coarsening upwards, nature of the sediments is typical of Coal Measures fluvio-deltaic sequences, described in detail in the adjoining Pontypridd district by Woodland and Evans (1964, pp.20–22).

The cycles commence with medium to dark grey laminated mudstones, with irregularly distributed freshwater 'mussels' or bivalves. Periodic rises in sea level, either due to eustatic causes or to rapid subsidence, resulted in the deposition marine sediments on comparatively rare occasions. At least thirteen distinct and widespread marine bands have been recognised in the central parts of the coalfield. Two of them (Subcrenatum and Aegiranum) have yielded free-swimming goniatites which indicate depositon in deeper near-shore waters, but for the most part they are mainly Lingula bands reflecting fresh water input, and shallow brackish conditions.

The incoming of siltstone streaks and overall coarsening of grain-size accompanied by paler colour, indicate increasing depositional energy and more oxygenated bottom conditions. Sideritic ironstone layers and nodules are common in these finer rocks. 'Mussels', fish fragments and bioturbation may occur in the mudstones and silty mudstones, but they generally decrease upwards to be replaced by comminuted plant debris. Well preserved plants (leaves and stems) are common, and sometimes form recognisable horizons useful in local correlation. With continued increase in depositional energy, the proportion of silty material increases to give initially a 'striped' mudstone–siltstone facies, followed by siltstone-dominated beds in which current ripple lamination is the main structure. Thin sandstone wisps are generally succeeded by thicker sandstone layers, and these in turn pass upwards into sandstone-dominated beds in which cross-lamination and cut-and-fill structures are found. The sandstones mark the culmination of the coarsening-upwards cycle, and the appearance of rootlets indicates that the sediment–water interface was shallow enough for plant colonisation to begin. Decreasing current energy then produced a fining-upwards sandstone–siltstone–mudstone transition accompanied by increasing pedoturbation by rootlets resulting in the typical seatearth or gleyed palaeosol which underlies the coals. Siderite is common in these soils, both as small nodules and as sphaerosiderite ooliths, and pyrite is a common associate. The cycle is capped by a coal seam, representing an extensive period of plant debris accumulation in widespread fresh-water swamps when subsidence was virtually static. Peat growth came to a sudden end when renewed subsidence and drowning brought about the onset of a new cycle of sedimentation.

Sandstones may interrupt the modal Lower and Middle Coal Measures cycle at any level, representing fluvial deposition in several settings. Coarse-grained fining-upwards quartzitic sandstones are the deposits of major distributary channels; their bases are scoured into the underlying surface, and channel into coal seams locally—the so-called genuine 'washouts' of the miner. Finer-grained more feldspathic sandstones are waning-stage channel-fills, levees, overbank sheet flood deposits, or crevasse splay deposits. Where they occur as part of the coarsening-upwards deltaic cycle, the sandstones are the delta-top distributary and mouth-bar deposits of minor deltas. Where the sandstones occur as interruptions in the finer parts of the cycle, they were probably carried into the more distal parts of the lakes during flood periods.

The deposition of the Upper Coal Measures resulted from a major change in palaeogeography when the main source of the sediments switched from north to south and south-west. The rising, northwards advancing Variscan landmass to the south shed its detritus northwards into a rapidly subsiding foreland trough to produce the thick and massive Pennant sandstones (Kelling, in Owen, 1974). These rocks are immature subgreywackes, rich in lithic grains, grey when fresh, but weathering to khaki-green. They are arranged in thick large-scale cross-bedded units, which commonly contain much coalified wood debris and intraformational mudstone, siltstone and siderite clasts as well as quartz pebbles at the base, where the contact is commonly erosive on the underlying rocks. The sandstones were deposited by large, probably braided streams of low sinuosity. Relatively thin but laterally very persistent mudstones occur at numerous levels and give a crude cyclicity to the strata overall; they were probably interdistributary lake and overbank deposits. They are usually associated with one or several comparatively thin coals, which are of only minor importance compared to those of the Lower and Middle Coal Measures.

Lower Coal Measures (Westphalian A)

(Figure 6) gives a generalised section; the lower part of the sequence up to the Garw Seam is more like the Namurian than the overlying upper part of the Lower Coal Measures, in that the coals are thin and generally impersistent, and marine strata recur at a number of horizons, particularly in the bottom half of the sequence. Above the Garw the measures are almost wholly non-marine with a number of thick coals.

Measures below the Garw Seam

The Subcrenatum Marine Band is less than 2 m thick in parts of the west, where it underwent erosion prior to deposition of the overlying 'Farewell Rock' ; it was totally eroded at some places. In the east the band is generally about 14 m thick, and a maximum of over 20 m has been recorded in the Gelli Isaf Borehole<span data-type="footnote">(Figure 7) gives the locations of collieries and boreholes mentioned in the text.</span> (Barclay and Taylor, 1976). It consists mainly of medium to dark grey mudstones, and is abundantly fossiliferous in parts, especially towards the base where Gastrioceras subcrenatum C. Schmidt occurs in a typical goniatite-pectinoid biofacies. The band is exposed in the River Tawe [SN 8274 1265], in Nant Llech [SN 8335 1266]; [SN 8344 1258]; [SN 8588 1238], in tributaries of the River Neath [SN 8864 0624]; [SN 8871 0627]; [SN 8988 0694], in Cwm Gwrelych [SN 8908 0645], in the railway cutting at Pont Walby [SN 8920 0650] (Bloxam and Thomas, 1969), in the gorge of the River Neath at Pont Nedd Fechan [SN 9004 0776] (Bloxam and Thomas, 1969; Owen and others, 1965), in the River Mellte and its tributaries [SN 9058 0773]; [SN 9098 0823]; [SN 9225 0908]; [SN 9225 0973]; [SN 9220 1006], in Nant Melyn [SN 9793 0578] (Jones and Owen, 1956), in Nant Ffrwd [SO 0189 0743], and in Nant Morlais [SO 0728 0908].

The beds above consist of a series of coarsening-upwards cycles with marine bands commonly at their bases, and thick fluvial sandstones above. The sandstones, collectively known as the 'Farewell Rock' occur at two levels (Leitch and others, 1958). West of the Cynon Valley, the 'Farewell Rock' directly overlies the Subcrenatum Marine Band, and comprises both quartzitic and arenitic sandstones (Plate 5). It ranges in thickness from 20 m in the west to 45 m at Glynneath where it contains slumps and other synsedimentary deformation structures (Owen and others, 1965). It thins eastwards and wedges out in the neighbourhood of Hirwaun. In the east, the 'Farewell Rock' is a quartzitic sandstone, pebbly in places, lying at a higher horizon; it is about 30 m thick to the north of Dowlais Top.

There are five marine bands between the Subcrenatum Marine Band and the Garw Seam, named in upwards sequence M1 to M5 (Leitch and others, 1958). They have been correlated with the Springwood, Honley, Listen, Meadow Farm and Langley marine bands (Ramsbottom and others, 1978). In the east the marine bands are rather poorly developed and yield little other than Lingula; locally they are missing, where they were eroded prior to deposition of the overlying quartzitic sandstones. To the west the Ml, M2 and M3 bands yield a more varied fauna of brachiopods, including Productus (Dictyoclostus) sp.and marine bivalves in addition to the ubiquitous Lingula and Orbiculoidea (Leitch and others, 1958, pp.469, 476). The M5 band has yielded only fish scales and teeth.

Cwm Gwrelych, near Pont Walby in the Neath Valley, provides a compete section (Leitch and others, 1958, pp.465 - 466). Plant-bearing shales immediately overlie the 'Farewell Rock' (Dix, 1933). A thin siliceous grit, the 'One-Foot Sandstone', lies within the M1 Marine Band; the M2 Band lies above a 5 cm coal; M3 has a distinctive basal layer with pyrite-filled burrows and large Lingula; M4 is a dark grey shale with Lingula; and M5 has a thin basal layer of quartz granules and fish fragments.

Most of the 'Farewell Rock' was faulted out in the Gelli Isaf Borehole. The M1 and M3 marine bands yielded only Lingula and fish fragments; M2 yielded L. mytilloides J. Sowerby, Orbiculoidea nitida (Phillips) and Serpulites stubblefieldi (Schmidt and Teichmuller), and M4 L. mytilloides, Aviculopecten sp.and Productus carbonarius de Koninck. The marine bands pass up into grey and blue-grey mudstones with Planolites ophthalmoides Jessen and this burrow is particularly abundant throughout much of the mudstones below the Garw.

Two 'mussel' -bands (C1 and C2) lie between the M5 Marine Band and Garw Seam and they contain rich Carbonicola faunas. Horizon Cl was the more fossiliferous in the Gelli Isaf Borehole, yielding C. extenuata Eagar, C. aff. extenuata, C. aff. fallax Wright, C. aff. limax Wright, C. cf. obliqua Wright, C. aff. pillealum Eagar, C. aff. pontifex Eagar, C. aff. Proxima Eagar and Curvirimula belgica (Hind). Numerous clay-ironstone bands occur in this part of the sequence, and were formerly worked in the Merthyr Tydfil area and the Neath Valley as the Rosser Veins.

Nant Llech and Cwm Gwrelych provide the best sections at outcrop (Leitch and others, 1958; Robertson, 1933). There are also good sections in the River Giedd, in the River Tawe at Abercrave [SN 8170 1260], in the River Pyrddin [SN 8906 0935], where the 'Farewell Rock' is about 35 m thick, consisting of rolled and balled-up sandstone, in Nant y Gwyddyl [SN 8841 0727], in Cwm Rhyd-y-gan [SN 8804 0597], and in Cwm Wyrfa [SN 9262 0687]. A thin coal in the Dowlais area carries a mudstone roof with fish debris (?M1 Band) and the local 'Farewell Rock' lies 10 m above.

Garw Seam to Amman Rider

This interval varies from 60 m near Merthyr to over 130 m in the Swansea Valley. It is predominantly argillaceous, comprising a sequence of mudstone- siltstone- seatearth cycles and including several widely worked coals. The westwards expansion of the succession is accompanied by splitting of three of the main seams in the east (the Gellideg, Five-Feet and Seven-Feet) into five in the west (the Lower or New, Grey or Middle, Rhyd, Lower Bluers and Upper Bluers). Precise correlation of these seams remains uncertain.

The Garw, known in the west as the Bryn or Cnapiog, has not been worked extensively. It is about 0.5 m thick in the Swansea Valley, between 0.2 and 0.6 m in the Neath Valley, and about 0.45 m at outcrop in the Cynon and Taff valleys. The roof is typically dark mudstone with fish fragments. In addition, ostracods and Curvirimula fragments were found in the Gelli Isaf Borehole.

Tabular and nodular ironstones distributed through the measures between the Garw and the Gellideg were formerly worked from patchworks and shallow mines, particularly in the area north of Merthyr Tydfil and Aberdare.

Several cycles are recognisable between the Garw and Gellideg. A persistent shell-bed with Carbonicola pseudorobusta Trueman lies at the base of the second cycle, 6 m above the Garw; canneloid shale with fish debris occurs at the base of each of two higher cycles, with Carbonicola also in the lower one.

In the Taff Valley around Merthyr Tydfil the Gellideg is the lowest of three closely spaced coals (the upper two forming the standard Five-Feet), which make up a composite seam, known formerly as the Lower Four-Feet. A typical section, as at Twyn y Waun, east of Dowlais, reads: top coal 0.91 m, clod 0.20 m, middle coal 0.91 m, shale 0.08 m, engine coal 0.18 m, shale 0.76 m, bottom coal (Gellideg) 1.22 m (Robertson, 1933, p.104). At Cwmbach, just east-south-east of Aberdare the separation between the Gellideg and Five-Feet is only 0.20 m, but northwards and particularly westwards the seams split rapidly, the separation being as much as 21 m at Blaennant and Bwllfa. The Gellideg is 0.60 to 0.91 m thick in the Taff Valley; in the Cynon Valley a typical section is coal 1.0 m, dirt 0.4 m, coal 0.15 m, but it thins westwards to 0.6 m or less at Bwllfa and Tower.

In the Neath and Swansea valleys the Gellideg- Seven-Feet group is known as the Bluers Group: the Gellideg is probably represented by two seams, the Lower or New and the Upper, or Grey or Middle. The Lower Gellideg is 0.45 to 0.55 m thick west of the Neath Valley. The Upper Gellideg, which lies 7.6 to 8.0 m above, is 0.4 to 0.6 m thick in the Swansea Valley, and 0.6 to 0.7 m with a parting in the Neath Valley. The roof of the Upper Gellideg or Grey at Rock Colliery carries a 'mussel' band about 0.45 m above the coal, in which internal moulds are common; forms include ?Carbonicola acuta (Sowerby), C. aff. cristagalli Wright, C. aff. polmontensis (Brown), C. rhomboidalis Hind, Curvirimula trapeziforma (Dewar) and the ostracod Geisina arcuata (Bean). Thick persistent ironstones occur some 1.25 m above the coal.

The Rhyd of the Swansea Valley is tentatively correlated with the lower leaf of the Five-Feet, and the Lower Bluers with the upper leaf. The former is about 0.5 m thick in two leaves in the Swansea Valley and about 0.9 m at Rhigos opencast site.

The Lower Bluers (or Upper Five-Feet) lies about 4 to 6 m above the Rhyd in the Neath Valley; it is characterised by the presence of Leaia minima Pruvost and 'Estheria' in the roof mudstones. It is a complex seam with a section in the Glyn-neath Drift: coal 0.24 m, clod 0.5 m, coal 0.41 m, parting 0.05 m, coal 0.08 m. In the west of the district there is much thrusting in the measures of the Bluers group of seams, making the colliery sections in the Dulais and Swansea valleys suspect and the correlation of the individual coals doubtful. At Ynyscedwyn Colliery the Middle Seam (?Lower Five-Feet) is 0.71 to 0.81 m thick and the 'Rhyd' (?Upper Five-Feet) about 0.53 m with a parting.

East of the Neath Valley the two leaves of the Five-Feet converge and in the Cynon and Taff valleys they have been extensively worked as a single seam. At Tower Colliery the section was proved in No.6 Borehole: coal 0.36 m, rashings 0.03 m, coal 0.48 m. The separation increases in the Cynon Valley where a typical section is: coal 0.8 m, dirt 0.71 m, coal 0.43 to 0.61 m. The parting increases still further at the Clynmil Colliery, on the east side of the Taff Valley at Merthyr Tydfil, where the two coals, while maintaining their thickness, are about 2.5 m apart. East of Dowlais the coals again converge: coal 0.76 m, clod 0.30 m, coal 0.84 m, shale 0.08 m, inferior coal 0.15 m.

The Upper Five-Feet and the Seven-Feet converge locally at Abercrave in the Swansea Valley to form the Bluers Seam with a thin intervening parting. In general, 1.2 to 6.0 m separate the seams in the west; they coalesce at Ffyndaff opencast site and are again up to 9 m apart at Tower Colliery. In the east 6 to 18 m separate the seams, the beds comprising a typical upwards-coarsening cycle.

The Seven-Feet Seam is tentatively correlated with the Upper Bluers in the west of the district. A two-leaf section is typical in the Swansea Valley, with each leaf 0.45 to 0.6 m thick. In boreholes in the Dulais Valley three thin coals and a rider are recorded, although it is doubtful that this represents the true section. The seam is up to 1.2 m thick in the Rhigos area, and at Tower Colliery it assumes the typical three-leaf section of the standard Seven-Feet of the Taff and Cynon valleys, with the middle leaf being thickest, 0.60 m. In the Aberdare area the average section is top coal up to 0.60 m, dirt 0.08 to 0.38 m, middle and bottom coals 1.2 to 1.5 m. All three leaves converge at Merthyr Tydfil to give a single three-leaf coal, 2.0 m thick. They separate again to the east, and at Dowlais there are two coals, the Little Vein and Little Vein Rider separated by up to 6.0 m.

In the west the Upper Bluers Seam has a characteristic roof of black canneloid mudstone with Anthraconaia?, and ostracods including Carbonita cf. fabulina (Jones and Kirkby).

About 17 to 25 m separate the Seven-Feet from the Yard Seam in the western and central parts of the district; in the east the separation ranges between 6 and 18 m. In the west several cycles are present, with up to four thin coals. Two cycles are present at Tower, with only one upwards-coarsening cycle in the east.

In the west the Yard Seam is known as the Lower Peacock or Lower Brass and it is about 0.6 m thick. It thickens to between 0.6 and 0.9 m in the Neath Valley and this is maintained eastwards to Tower Colliery and the Aberdare area where the seam was formerly known as the No.2 Yard. Here it shows a fairly consistent two-coal section, each leaf being 0.30 to 0.46 m thick with a parting varying from 0.03 to 0.30 m. The seam deteriorates east of Merthyr Tydfil, and three leaves each generally less than 0.3 m are present at Dowlais, where it was known as the Three Coals.

Dark mudstones everywhere form the roof of the Yard and they yield an uncrushed 'mussel' fauna which is highly characteristic of the early Modiolaris Chronozone: Anthracosia regularis (Trueman), Carbonicola venusta Davies and Trueman and the trace fossil Planolites sp..

About 15 m of measures separate the Yard Seam from the Amman Rider over the western and central parts of the district, thickening to 24 m at Tower Colliery and then thinning again eastwards until, in parts of the Cynon and Taff valleys, the two seams virtually unite. Three coarsening-upwards cycles capped by seatearths and thin coals occur over most of the west. Two cycles are developed around Tower, where the lower is capped by a persistent seatearth and thin coal. The beds comprise one cycle in the east, where a characteristic feature is the presence of standing Lepidodendron trunks.

The Amman Rider is a thin sulphurous coal throughout the district, commonly in two leaves in the east. The top coal is about 0.15 to 0.25 m thick and the bottom coal little more than 0.15 m.

Middle Coal Measures (Westphalian B and early Westphalian C)

(Figure 8) gives a generalised vertical section of the Middle Coal Measures. The beds include many of the important coals of the district, as well as seven marine bands.

Amman (Vanderbeckei) Marine Band to Cefn Coed (Aegiranum) Marine Band

Overall these measures increase in thickness steadily from about 100 m in the east of the district to about 200 m in the west.

The Amman Marine Band is present throughout the district; it is particularly important since it is the only development of marine strata among the main group of coals. Varying in thickness from a few centimetres in the west to about 1 m in the east, it is made up of medium to dark grey mudstone or silty mudstone, containing much disseminated pyrite, which on weathering gives rise to a characteristic yellowish 'sulphury' appearance. Lingula mytilloides, rare Orbiculoidea, crinoid ossicles and fish scales and teeth are almost the only macrofossils, and the burrow Planolites ophthalmoides Jessen increases in abundance upwards. Pyrite-filled burrows and granules are also characteristic. The marine strata are usually succeeded by smooth mudstone containing poorly preserved and often 'ghost'-like 'mussels' including Anthraconaia?, Anthracosia intermediate between aquilina (J. de C. Sowerby) and lateralis (Brown), and A. aff. ovum Trueman and Weir. Ironstones are common in these beds.

About 6 to 8 m of measures lie between the Amman Rider and succeeding Bute Seam over much of the district, increasing to 9 to 12 m in the Neath and Swansea valleys, and also in the Merthyr Tydfil area where boreholes drilled in connection with the new A470 road proved 10 m of beds.

The Bute Seam has proved to be of good quality and thickness throughout the district, and has been widely worked. In the Swansea Valley, where it is known as the Brass or Peacock, it is typically about 0.9 m thick. It deteriorates in the Dulais Valley, the individual leaves in a two-coal section being generally less than 0.6 m. It is a good single coal, 1 m thick, in the Neath Valley, where it has been almost completely worked out on the west side of the valley as far as the Pentreclwydau Fault. From Cwm Gwrelych eastwards towards the Cynon Valley, a single coal, 0.9 to 1.2 m thick is much affected by 'washouts'. The coal splits into two leaves in the Cynon Valley, where it has a consistent section of top and bottom coals each 0.5 to 0.6 m thick, with a parting about 0.5 m thick. This deterioration continues towards the Taff Valley where the parting increases locally to 4.6 m and the two leaves are thin.

The Bute roof carries a highly characteristic 'mussel' fauna over most of the district from the Cynon Valley westwards; the shells include Anthraconaia cf. curtata (Brown), A. modiolaris Hind, Anthracosphaerium affine (Davies and Trueman), A. exiguum (Davies and Trueman), A. aff. fuscum (Davies and Trueman) and Naiadites sp.The roof of the split bottom coal at Merthyr Vale Colliery has yielded A. exiguum and N. aff. subtruncatus (Brown).

The interval between the Bute and Nine-Feet seams ranges from over 18 m in the Swansea and Dulais valleys to as little as 0.25 m in the old Plymouth and Castle pits of Merthyr Tydfil. Thicknesses of 8.5 to 14 m are typical of the Neath Valley, while at Tower Colliery where channel-fill sandstones 'wash out' the Bute Seam locally, a thickness of 21 to 24 m is common, and 9 m is a typical figure in the Cynon Valley.

Over much of the district the 'mussel' -rich roof mudstones of the Bute lie at the base of a typical coarsening-upwards cycle. In parts of the Tawe Valley, e.g. at Ystradgynlais Colliery, an intervening cycle develops with 'mussel' -bearing mudstone resting on a thin coal and seatearth about 6 m below the Nine-Feet. The channel-fill deposits of the Tower area are arranged in fining-upwards sandstones–siltstones subcycles.

The Nine-Feet Seam (Big Vein of the west of the district) is the most important coal of the district and has been extensively mined, particularly in the Cynon Valley, where it is up to 3 m thick with a typical two-leaf section: top coal 1.07 m, clod or rashings 0.15 to 0.3 m, bottom coal 1.52 m.

Over much of this area the coal is very disturbed and abnormally thick sections of mixed coal and rashings up to 5 m or more are common. The same two-coal section is seen over much of the west of the district, though a dirt parting develops locally in the upper coal; at Onllwyn No.1 Pit the average section reads: coal (inferior) 0.35 m, clod or shale 0.45 to 0.85 m, coal 0.38 to 0.48 m, clod or rashings, coal 1.52 to 2.35 m. At Cefn Coed Colliery, just west of the district boundary, the upper of the two main coals splits away, is known as the Dulais and has a cannel band 5 cm below its top. Abnormal seam thicknesses are also common throughout the Dulais and Neath valleys. The seam was especially thick at Ynyscedwyn Colliery, where a large lenticular mass up to 21 m has been recorded (Robertson, 1933, p.126). The two main leaves also diverge towards Dowlais in the extreme east of the district, which presages a major split into the Lower Nine-Feet (or Rhaslas) and Upper Nine-Feet (Red or Polka) of the Abergavenny district to the east, and the Lower and Upper Nine-Feet Seams of the Pontypridd district to the south.

A characteristic feature of the lower leaf of the Nine-Feet is the occurrence of persistent bands of 'brass', consisting mainly of lenticular and nodular masses of pyrite, up to 0.15 m thick, sometimes associated with sphaerosiderite or massive siderite and calcite.

Rich 'mussel' faunas have been recorded from the roof of the Nine-Feet Seam in the Swansea and Dulais valleys and also in the Cynon and Taff valleys. In the west the shell-bearing sediments have been described as 'greasy' grey shales; in the east they are dark smooth carbonaceous shales. The following forms were recorded by Davies and Trueman (1927, p.241) from Seven Sisters and Ynyscedwyn collieries: Anthraconiaia curtata, Anthracosia aquilina U. de C. Sowerby), A. nitida (Davies and Trueman), Anthracosphaerium bellum (J. de C. Sowerby), Naiadites carinata (J. de C. Sowerby) N. obovatus (Hind), and N. modiolaris (J. de C. Sowerby). In the Cynon and Taff valleys the following have been identified: Anthraconaia aff. salteri (Leitch), Anthracosia aff. beaniana King, A. sp. cf. disjuncta Trueman and Weir, and variants of the group represented by the species A. aquilina, A. ovum (Trueman and Weir), A. phrygiana (Wright), and A. retrotracta (Wright) (Woodland and Evans, 1964, p.106). Over a large area of the central part of the district the 'mussel'-band is absent, and the Nine-Feet roof consists of plant-bearing mudstones, siltstones and sandstones. Where the seam is split east of the Taff Valley, 'mussels' are present in canneloid mudstone above the lower (Rhaslas) coal, though none has been found in the roof of the upper coal.

The strata between the Nine-Feet and the Six-Feet are amongst the most variable in the Middle Coal Measures, due for the most part to the very irregular behaviour of the Red Vein group of coals, which split and thicken as the measures generally expand westwards. In the Maerdy–Cwmaman area just south of Aberdare the Red Vein is a single composite seam lying some 14 to 17 m above the Nine-Feet; it usually shows three distinct leaves, e.g. coal 0.38 m, dirt 0.78 m, coal 0.68 m, dirt 0.7 m, coal 0.55 m. In the Aberdare area itself the dirt partings thicken and the coal has been very little worked. It also deteriorates rapidly eastwards and is poorly developed in the Taff Valley, where it lies close above the Nine-Feet. Farther east in the Abergavenny district the seam should not be confused with the Red Vein of the Rhymney Valley, which equates with the Upper Nine-Feet. Westwards into the Neath and Dulais area two seams occur between the Nine-Feet and the Six-Feet; they are known as the Harnlo (or Lower Cornish) and the Cornish or Cornish Four-Feet seams. By analogy with the behaviour of the Red Vein group in the Pontypridd district to the south, it is possible that these two seams represent major splits from the compound Red Vein of the area south of Aberdare.

In the Neath, Dulais and Swansea valleys about 12 to 25 m of measures intervene between the Nine-Feet and the Harnlo. The Harnlo Seam has a characteristic section west of the Neath Valley with a persistent band of cannel or 'cornish', up to about 0.2 m thick in the roof of the coal or up to 0.4 m above it. An abundant fauna of large crushed shells has been recorded from canneloid mudstones in the roof of the coal, including Anthraconaia cf. robertsoni (Brown), Anthracosia cf. aquilina, A. cf. aquilinoides (Chernyshev), A. aff. beaniana King, A. aff. ovum, A. aff. phrygiana, A. cf. regularis (Trueman), A. sp.intermediate between beaniana and phrygiana, A. sp.intermediate between phrygiana and regularis; Carbonita cf. fabulina Jones and Kirby has also been found.

About 5 to 12 m of mudstone and silty mudstone usually lie between the Harnlo and the Cornish Four-Feet, in places comprising two cycles. The Cornish Four-Feet has been widely worked at the head of the Neath Valley and in both the Dulais and Swansea valleys. The section varies generally from 0.8 to 1.1 m in thickness, locally with a median parting, and with a band of cannel or 'cornish' up to 0.2 m thick at the top. Over a large part of the Dulais Valley the seam has been largely 'washed out' and the channels filled by a thick pebbly and quartzitic sandstone. Where not cut out by overlying sandstones, the roof mudstones contain an abundant fauna: Anthraconaia aff. curtata, Anthracosia aquilina (sensu lato), A. nitida, A. simulans Trueman and Weir, Naiadites sp.; and Belinurus spp.Plants are also abundant in places and the shark egg cases Palaxyris helicteroides (Morris) and Vetacapsula kidstoni Crookall have been recorded (Dix, 1934).

The measures between the Cornish Four-Feet or Red Vein and the Six-Feet Seam are variable in thickness, ranging from 25 to 40 m in the Taff Valley, 1 to 25 m in the Cynon Valley, 9 to 11 m in the Neath Valley, 10 to 29 m in the Dulais Valley and 9 to 21 m in the Swansea Valley. In the east there are several minor rhythmic units present, generally coarsening upwards. In the west two cycles are usually present, with a thin coal recorded locally at the top of the lower one.

Throughout the district the Six-Feet Seam has proved second only to the Nine-Feet in economic importance. A single though compound seam of two main coals occurs in the Merthyr Tydfil area and again in the area extending from Tower Colliery to the western side of the Neath Valley. Elsewhere the seam is split into its two component parts, the Lower and Upper Six-Feet seams, the separation ranging up to nearly 15 m in the Swansea and Dulais valleys and south of Aberdare.

In the Swansea and upper Dulais valleys the Lower Six-Feet was known (mistakenly) as the Cornish and the Upper Six-Feet as the Black, Upper Black, Eighteen-Feet or Little. In the lower part of the Dulais Valley, at Cefn Coed, the Lower Six-Feet is known as the Lower Eighteen-Feet or White, and the Upper Six-Feet as the Upper Eighteen-Feet. The lower coal appears to be 0.9 to 1.0 m thick, and the upper 0.6 to 0.75 m. Higher up the Dulais Valley the Six-Feet consists of three coals. The middle coal is only a few centimetres thick, but carries a 'mussel' fauna in its roof, which probably correlates with that in the roof of the Lower Six-Feet (Lower Eighteen-Feet) at Cefn Coed.

The natural separation between the Upper and Lower Six-Feet seams is 12 to 15 m in the Dulais Valley. At Onllwyn Colliery, at the head of the valley, the separation is variable; overall the two leaves are much closer than to the west, the separation being 0.15 to 3.0 m, and each leaf 0.6 to 0.9 m thick. They continue to converge eastwards, and in the eastern part of Aberpergwm Colliery only a thin parting intervenes.

The Six-Feet and the overlying Four-Feet converge over a small area on the western side of the Neath Valley, where the combined seam was known as the Eighteen-Feet. It was extensively worked at Aberpergwm Colliery and at Maesgwyn opencast site. In the west of the Maesgwyn site, the Six-Feet, 2.4 to 3.0 m thick, lies 12 m below the Four-Feet which is 0.9 to 1.2 m thick. The two coals converge eastwards and coalesce within 275 m to give a seam, 2.4 to 4.0 m thick with only minor partings.

In the central part of the district from the Neath Valley to Tower Colliery, the Six-Feet consists of one coal averaging 2.5 to 3.0 m in thickness, with a maximum of 3.7 m at Tower. It has been extensively worked in the opencast sites between Rhigos and Cwmgwrach.

The Six-Feet splits south-eastwards across the Tower Colliery area into a lower coal 1.5 to 2.1 m thick and an upper coal 0.75 to 0.9 m thick, known locally as the No.1 Yard. Tower No.6 Borehole proved a section showing the Upper Six-Feet 12 m above the Lower Six-Feet. There were traces of 'mussels' in the roof of the lower coal. The seams continue to be split when traced from the Cynon Valley into the Taff Valley. At Bryn Pica opencast site the Lower Six-Feet is 1.6 m thick and the Upper Six-Feet 0.8 m, the two seams being separated by 10 m of dark grey carbonaceous mudstones with plants passing upwards into silty mudstones with lenticular sandstones. North-east of Merthyr Tydfil, the two coals converge giving a single seam 2.4 to 3.0 m thick. There are at least four partings in the section at Trecatty opencast site, dividing the coal into two main leaves about 0.9 m thick, with several minor leaves. The two leaves also converge southwards in the Taff Valley, the line of convergence passing through Merthyr Vale Colliery and the compound seam being 2.4 to 3.0 m thick. Up to five thin closely separated coals lie immediately beneath the Lower Six-Feet in some of the pits in the Aberdare area of the Cynon Valley, the interleaf separation being greater at Werfa and Tunnel pits.

Within the present district roof 'mussels' associated with the Six-Feet have been recorded only in the west and at Merthyr Vale Colliery in the extreme south-east (Woodland and Evans, 1964, p.107). In the west the best collections are those of Ware (1930, p.474) at Cefn Coed Colliery just outside the district boundary. From the roof of the Lower Eighteen-Feet he listed Anthraconaia sp., Anthracosia aquilina, A. simulans, Naiadites triangularis, and fish scales. Small Naiadites sp.and Euestheria sp.are generally present above the Upper Six-Feet in the Neath and Dulais valleys, and Anthraconaia cf. lanceolata (Hind) and 'Estheria' sp.were found on the combined Eighteen-Feet Seam at Ynysarwed Colliery. To the east of the Neath Valley, only 'mussel' fragments have been recorded from the roof of the Upper Six-Feet, and Lioestheria striata Pruvost occurs in the roof of the Lower Six-Feet south of Aberdare.

The Six-Feet and the Four-Feet are close together at Maesgwyn, and attain a maximum separation of 21 m at Tower Colliery, though this figure may also indicate repetition by thrusting, since 9 to 15 m is more usual over most of the Cynon area. In general the interval is represented by a coarsening-upwards mudstone cycle. Carbonaceous silty mudstones with plant debris form the base of the cycle in the west and at Bryn Pica opencast site.

The Four-Feet is a complex seam. It was known as the Stwrin at Cefn Coed in the Dulais Valley, where it reaches 1.5 to 1.8 m in two or three leaves. Higher up the Dulais Valley and in the Neath Valley it was known as the White Four-Feet west of its convergence with the Six-Feet at Aberpergwm and Maesgwyn. Although much disturbed in this area, it can generally be recognised as a fairly clean coal 1.0 to 1.3 m thick, and it has been widely worked in the upper Neath Valley. The Four-Feet is split in the Rhigos-Tower area into the Upper Four-Feet (or Driver), about 0.6 m thick, and the Lower Four-Feet, a two-leaf seam 0.9 to 1.1 m thick. The two coals converge south-eastwards to form a single three-leaf coal with only thin partings, which was extensively worked throughout the Cynon-Taff area. A typical section was that at Werfa Colliery: coal 0.45 m, clod 0.07 m, coal 1.22 m, coal 0.25 m. Locally both top and bottom coals separate somewhat from the main middle leaf, the worked section being either the two top coals together or the two bottom ones.

The Four-Feet (Stwrin) roof carries a 'mussel' fauna at Cefn Coed Colliery, including Anthracosia cf. aquilina, A. atra (Trueman), A. fulva (Davies and Trueman), A. simulans, and Naiadites productus (Brown), as well as fish remains (Ware, 1930, p.474). A similar fauna occurs above the White Four-Feet at Ynysarwed and Glyncastle collieries and on the east side of the Neath Valley. It fails eastwards, and only 'Estheria' has been recorded at Aberpergwm, nor have any 'mussels' been found from the Four-Feet in the ground to the east.

About 12 to 23 m of coarsening-upwards measures separate the Four-Feet from the Two-Feet-Nine in the west of the district. In the central parts of the district the interval is 3 to 8 m, and this thickens again eastwards to 9 to 12 m where sandstone is prominent in the thicker sections.

In the west of the district the Two-Feet-Nine Seam is represented by four thin coals, the lowest two of which are known locally as the Stwrin and Penny Pieces, and the topmost as the Soap. These coalesce eastwards at Rhigos, and the Two-Feet-Nine is a clean coal, 0.6 to 0.9 m thick in the Bwllfa- Tower area, and in the Cynon Valley, where it thickens northwards to 1.2 m. It deteriorates rapidly east of the Cynon, and in the Taff Valley is rarely more than 0.6 m thick.

The four seams of the west of the district can be traced from Rhigos to Maesgwyn, and again in boreholes in the Dulais Valley. They are all less than 0.6 m thick, and span about 12 m of beds. The lowest coal, the Stwrin, is 0.15 m thick and has a cannel roof. This seam is not to be confused with the coal named the Stwrin at Cefn Coed Colliery. It can be traced throughout the Neath Valley, and extends westwards to the Abercrave opencast site, where it is represented by cannel. The canneloid roof has yielded 'mussels' including Anthraconaia cf. librata (Wright), Anthracosia cf. lateralis (Brown) and A. planitumida (Trueman). The roof of the Penny Pieces is a valuable marker throughout the west of the district, containing an abundant 'mussel' fauna in mudstones rich in sideritic ironstone layers that have been worked in the past for iron ore. The coal equates with the lower-leaf of the Two-Feet-Nine in the north of the Rhondda valleys, where the 'mussel' fauna was also found (Woodland and Evans, 1964, p.58). The fauna listed by Ware (1930, p.473) includes Anthraconaia adamsii (Salter), An. cf. confusa (Trueman), An. obscura (Davies and Trueman) An. rubida (Davies and Trueman), An. wardi (Hind non Salter), Anthracosia acutella (Wright), A. aquilina, A. bipennis (Brown), A. cf. concinna (Wright), A. fulva, A. planitumida, A. retrotracta (Wright), A. cf. simulans and Naiadites triangularis. The Soap Vein, the topmost of the Two-Feet-Nine group of coals in the Swansea and Dulais valleys, also carries an abundant 'mussel' fauna (Ware, 1930, p.472) including Anthraconaia confusa, A. wardi, Anthracosia cf. aquilina, A. aff. atra, A. concinna, A. gibbosa (Hind), A. simulans, Naiadites subtruncatus (Brown), as well as Belinurus sp.and 'Estheria' sp.

About 60 m of measures separate the Two-Feet-Nine group of coals from the Cefn Coed Marine Band in the Dulais Valley. Five main cycles, each associated with thin coals, equate with the Three Coals and the Hafod Heulog and Cefn Coed marine horizons of the Pontypridd district (Woodland and Evans, 1964, pp.56–60). 'Mussel' faunas were collected by Ware (1930, pp.470–471) from the three lower coals during the Cefn Coed sinkings. Two very distinctive 'mussel' bands occur above the highest of the three seams, which lies about 24 m above the Two-FeetNine, both containing numerous shells belonging to the Anthracosia aquilina group. These two horizons stretch as far as the Swansea Valley where Ware located similar assemblages. Only the two higher coals were located in the Dulais boreholes, with a rich 'mussel' fauna in the roof of the higher one. At Ynysarwed, the lowest coal has a 'mussel'-bearing roof, but the higher beds are faulted out; at Aberpergwm, a thin coal about 30 m above the Two-Feet-Nine carries 'mussels', but in a nearby borehole 'Estheria' occurs in a pyritic mudstone roof. Three thin coals are present at Maesgwyn, the highest of which lies 15 m above the TwoFeet-Nine and carries a 'mussel' fauna. Three thin coals also represent the Three Coals at Bryn Pica; fining-upwards fluvial channel-sandstones separate the two higher coals, and a similar sandstone, equivalent to the Lower Cockshot Rock overlies the top coal. Here a thin two-leaf coal above the sandstone carries the Hafod Heulog Marine Band in its roof.

In the Cynon Valley the Three Coals group has a very irregular development. It commonly forms a single though compound seam of inferior quality, lying 9 to 15 m above the Two-Feet-Nine. Three to five leaves of coal, separated by varying amounts of rashings or seatearth may be present. In the Taff Valley sandstones occupy much of the ground between the Two-Feet-Nine and the Cefn Coed Marine Band and the Three Coals group seems largely to have been washed out. Two fining-upwards cycles are present at Trecatty opencast site, and are capped by thin coals.

The Hafod Heulog (Haughton) Marine Band lies in the roof of 0.25 m of coal at Cefn Coed, and comprises 2.5 m of dark laminated shales with large Lingula, Orbiculoidea and a productoid near the base, tiny Lingula in the middle and Euestheria in the upper part (Ware, 1930, p.479). The band is poorly fossiliferous over most of the west of the district, containing only Lingula and fish fragments, and locally Orbiculoidea. L. mytilloides is present throughout 2 m of mudstones in the Dulais No.2 Borehole, succeeded by silty mudstones with 'Estheria' and Planolites ophthalmoides, but only dark grey barren pyritic mudstones occur in other Dulais boreholes. Similarly at Aberpergwm, a shallow borehole proved the marine horizon to consist of pyritic mudstone with 'Estheria' lying on seatearth. The marine band lies on a 0.2 m coal some 21 m above the top coal of the Two-Feet-Nine group at Maesgwyn, where it contains L. mytilloides and fish fragments. It has not been seen in the east of the district. The Upper Cockshot Rock overlies the

Hafod Heulog Marine Band. It is a quartzitic sandstone generally, and in the Swansea Valley resembles a ganister. About 15 m thick at Cefn Coed it fines upwards through siltstone into a seatearth capped by a coal 10 cm thick. It is less than 6 m thick in the upper Dulais Valley, and in the Neath Valley it ranges from 15 m on the west side to 2 m on the east. It is absent at Aberpergwm and Maesgwyn and over most of the ground to the east, though there is a sandstone, up to 4 m thick, in the same stratigraphical position in the Cynon Valley (Robertson, 1933, p.148).

The 10*cm coal at Cefn Coed carries Naiadites and Belinurus in its roof, which lies at the base of a 15 m coarsening-upwards cycle. The same thin coal is present throughout the Dulais and Neath valleys, where P. ophthalmoides occurs in its roof mudstones. This is the horizon of the Britannic (Sutton) Marine Band of the Pontypridd and Maesteg district (Woodland and Evan, 1964, p.59) and of the Trimsaran Marine Band in the Gwendraeth Valley (Archer, 1968, pp.100–101). It is exposed in Nant Gwrelych [SN 9155 0466] where it rests on 10 cm of coal and consists of dark shale with ?Lingula, 'mussels', ostracods and fish fragments. It has not been proved to the east.

Cefn Coed (Aegiranum) Marine Band to Upper Cwmgorse (Cambriense) Marine Band

These strata range from 195 m in thickness in the west of the district to 90 m in the east. The tributary streams of the Neath provide the best sections, particularly Pentreclwydau Brook on the north side of the valley, 2 miles west of Cwmgwrach.

The Cefn Coed Marine Band is present throughout the district, and was identified by Robertson (1933, pp.147–156), following Ware's (1930, p.478) description of it at its type locality at Cefn Coed Colliery. There it consists of 0.33 m of mudstones with an abundant and varied fauna of corals, crinoids, bivalves, brachiopods, goniatites, and trilobites overlain by mudstones with a restricted fauna of Lingula. A non-marine phase of mudstones with 'Estheria' and sporadic 'mussels' lies above; this is 9 to 12 m thick and extends throughout the western and central parts of the district. P. ophthalmoides also occurs in these beds. The following fauna was collected at Maesgwyn opencast site from 0.45 m of dark grey shaly pyritic mudstone, separated from the underlying thin coal by 7 cm of shaly mudstone with fish scales and 'mussels': crinoid columnals; Archaeocidaris sp.; sponge spicules; Coleolus sp.; Dictyoclostus sp., Lingula mytilloides, Orbiculoidea cf. nitida, productoid indet, Tornquistia diminuta (Demanet); Nuculopsis gibbosa (Fleming), Palaeoneilo sp.?, Phestia sharmani (Etheridge), Posidonia sp?; euryomphaloid indet., orthocone indet., turreted gastropods; Donetzoceras sp., Gastrioceras sp; Holinella ulrichi Knight and Asturiella fernandezi Bless. The marine band is thin and impoverished in the Neath Valley, as for example in a stream near Pentreclwydau [SN 8540 0540], where it is a Lingula-bearing mudstone with plants. A rich fauna was recorded from a borehole at Pentrebach in the Taff Valley.

The cycle above the Cefn Coed Marine Band varies in thickness from about 25 to 30 m in the Dulais Valley, 15 to 25 m in the Neath Valley, 9 to 12 m in the Cynon Valley, and 12 to 18 m in the Taff Valley. Sandstone in the upper part of the cycle contains persistent slumped layers in some of the Dulais Valley boreholes. In the east the beds are for the most part argillaceous and rich in ironstone layers and nodules, which were worked extensively in the Taff Valley as the Black Pins Mine Ground. At Trecatty opencast site, however, only 3 m of beds are present, including the Cefn Coed Marine Band, before the normal cycle is truncated by up to 20 m of large-scale cross-bedded sandstones infilling a major distributary channel.

The Gorllwyn Seam is a thin but distinctive coal in the west of the district. At Cefn Coed the No.2 Shaft section showed: shale and coal streaks 0.15 m, dirt 0.28 m, coal 0.64 m. It carries a characteristic roof of 2.75 m of black carbonaceous shale with ironstone bands rich in 'mussels' and known as the Blackband Ironstone. The following fauna has been recorded (Robertson, 1933, p.157) Anthraconaia cf. adamsii (Salter), cf. An. adamsii var. expansa (Hind); aff. An. dolabrata (J. de C. Sowerby), aff. An. hindi (Wright), An. cf. librata, Naiadites cf. productus, N. cf. subtruncatus (Brown), N. cf. triangularis, and Euestheria simoni (Pruvost).

The seam is variable in the Neath Valley: on the west side at Glyncastle it is 0.25 m thick and in a borehole at Ynysarwed it consists of three leaves, totalling 1.5 m; in a nearby drift two leaves are present with a section: coal 0.08 m, dirt 0.05 to 0.08 m, coal 0.33 m (Robertson, 1933, p.157). A two-leaf section was proved in the Treforgan boreholes, and another borehole at Maesgwyn proved a complex 0.46 m thick section of coal and mudstone. At Bwllfa, on the west side of the Cynon Valley, coal up to 0.84 m thick is recorded but the seam splits into two leaves in many places, only one, 0.41 to 0.56 m, providing the working section. At the nearby Graig Pit, the section was: coal (with 0.05 m parting) 0.46 m, Soap Vein Ironstone 1.2 m, coal 0.38 m (Robertson, 1933, p.156). It was also worked in the Cynon Valley pits with a similar two-leaf section. It is patchily developed in the Taff Valley, and appears to be absent at South Dyffryn Colliery, the coal named as the Gorllwyn by Robertson (1933, p.156) probably being the Eighteen-Inch, with the Gorllwyn seatearth 4.95 m below. In boreholes at Pentrebach the Gorllwyn was 0.08 m thick, and at Trecatty it is up to 0.3 m.

The Gorllwyn Rider, where present, lies up to 12 m above the Gorllwyn, but its development is patchy. It is present at Cefn Coed and in the Dulais Valley and Treforgan boreholes, where it ranges from 0.08 to 0.27 m, but it is absent at Ynysarwed and Maesgwyn. A coal 0.10 m thick lying 6 to 8 m above the Gorllwyn in the Rhigos area is presumably the Gorllwyn Rider. It is known as the Gorllwyn Fach in the Cynon Valley, where it lies between 12 and 22 m above the Gorllwyn, is up to about 0.6 m thick, and has been worked on a small scale. Carbonaceous mudstone some 3 m above the Gorllwyn at Pentrebach, in the Taff Valley, may mark the position of the Gorllwyn Rider. At Trecatty, where it is known as the Black Pins, it is up to 0.3 m thick and is separated from the Gorllwyn by 8 to 9 m of quartzitic sandstone.

Naiadites sp.and large specimens of Anthraconaia were recorded in the roof of the Gorllwyn Rider at Cefn Coed (Ware, 1930, p.466), and the same fauna is present in the Dulais Valley and Treforgan boreholes. In the east dark grey carbonaceous mudstones with fish fragments lie at the base of the coarsening-upwards cycle, which is capped by the Eighteen-Inch Seam.

The Eighteen-Inch Seam lies 12 to 15 m above the Gorllwyn Rider in the Taff-Cynon area, where it is about 0.3 m thick. It is 0.15 m thick at Pentrebach, south of Merthyr Tydfil, but it thickens north-eastwards, and is a two-leaf seam at Trecatty where it was worked as the Soap Vein. The canneloid mudstone roof carries Naiadites alatus Trueman and Weir and N. melvillei Trueman and Weir, ostracods and fish fragments. The seam's identity is in doubt over the west of the district, though it is generally considered to be the lowest of a group of four thin coals in the Dulais Valley.

The Lower Pentre is generally the bottom coal of a complex group of thin coals, the detailed correlation of which is obscure apart from the Pentre Rider at the top which carries the Foraminifera Marine Band in its roof. At Cefn Coed four thin coals lie between the Gorllwyn Rider and Pentre Rider compared with three (Eighteen-Inch, Lower Pentre and Pentre) in the Cynon-Taff area, and it is not clear how these correlate; it is possible that in the west an extra cycle is developed between the Pentre and Pentre Rider, since a thin coal is present at this level near Maesteg (Woodland and Evans, 1964, p.62). If this correlation is correct, the only significant coal at Cefn Coed is the Pentre, 0.9 m thick, lying at a depth of 472 m in No.2 Shaft. 'Mussels' including Anthraconaia dolabrata and Naiadites spp., as well as fish scales, were recorded by Ware (1930, p.466) from its roof. At Ynysarwed and Maesgwyn only a single coal is present in this part of the sequence. In both the Dulais and Neath valleys much of the succession consists of sandstones, the other coals of the Pentre group are probably absent because of washout. The Pentre Rider is locally 0.75 m thick, and was worked in the Neath Valley under the name Ynysarwed Red Vein. The Lower Pentre, Pentre and Pentre Rider are all present throughout the Cynon and Taff valleys, but only the Pentre is of significant thickness, being 0.6 m at Pentrebach. The Pentre Rider averages about 0.45 m.

The Foraminifera (Edmondia) Marine Band is a distinctive marker band throughout the district. At Cefn Coed Ware (1930, p.478) recorded 6 m of mudstones which were marine in aspect, with Lingula confined to a layer in immediate contact with the Pentre Rider coal. The marine mudstones are everywhere characterised by an abundance of arenaceous foraminifera, along with Planolites ophthalmoides, Spirorbis sp.and ostracods. The mudstones also contain small spherical ironstone nodules as well as pyritic granules, fucoid' smears and burrow-fills. The marine band lies at the base of a coarsening-upwards cycle up to 20 m thick. In one of the Dulais Valley boreholes, this cycle is interrupted by a thin fine pyritic mudstone with Euphemites sp., lying some 10 m above the base of the Foraminifera Marine Band; this layer probably equates with the Five Roads Marine Band of the Gwendraeth Valley (Archer, 1968, pp.110 -111), which was also found in the Margam boreholes (Woodland and Evans, 1964, p.63).

The Foraminifera Marine Band cycle is capped by a seatearth and a persistent unnamed coal throughout the west of the district, but which is absent in the east where the cycle continues upwards uninterrupted to the Abergorky Seam.

The Abergorky, known as the Red Vein throughout the west of the district, lies about 15 to 28 m above the Pentre Rider. It has been widely worked in the west, where it averages 0.8 to 0.9 m of clean coal. It is about 0.45 m thick in the Neath Valley, and is known as the Welsh Vein at Ynysarwed. Several thin coals are present on the west of the Cynon Valley, extending over about 3.5 m, including one of workable thickness, 0.66 m, known as the Graig at Bwllfa.

One or two closely spaced thin coals are present at this level in the Taff Valley and the ground to the east.

The roof of the Abergorky contains an abundance of 'Estheria' west of the Cynon Valley, which makes it a valuable marker band. 'Estheria' was rare at Cefn Coed but Anthraconaia aff. oblonga (Wright), Naiadites spp.and an insect wing were recorded by Ware (1930, p.466). Sandstones are present in the higher parts of the cycle along the outcrop between the Dulais and Neath valleys, channelling down into the underlying mudstones. The cycle, which ranges from 17 m in the west to less than 5 m in the east is overlain by two closely spaced thin coals in the Taff Valley, the upper of which carries the Lower Cwmgorse Marine Band.

The Lower Cwmgorse (Shafton) Marine Band is 10 m thick at Cefn Coed, from which Ware (1930, pp.476–477) obtained Lingula, Orbiculoidea and Dunbarella sp.from several levels; in addition, dark shale at the base yielded Nuculana, Posidoniella, numerous fish scales and ostracods. It is generally thinner and less fossiliferous elsewhere in the west, being no more than a Lingula bed up to 6 m thick. In the Dulais Valley boreholes, L. mytilloides and Planolites ophthalmoides occur in pyritic mudstones 3 to 6 m thick and 'Estheria' usually occurs in the beds immediately above. About 4 m of marine mudstones are present at Treforgan, where pyritised crinoid columnals, Orbiculoidea sp., fish scales and P. ophthalmoides were collected. Robertson (1933, p.164) noted Dunbarella sp.at Ynysarwed. The band is exposed in Pentreclwydau Brook [SN 8343 0504], in Nant Ysgwlfa [SN 8586 0694], in the brook above Ynys-yr-allor Farm [SN 8595 0422] (Robertson, 1930, p.162), in Nant Pen-y-Cae [SN 8850 0394], and in Nant Llyn Fach [SN 9050 0439]. At Rhigos opencast site 0.9 m of silty mudstones with L. mytilloides and pyritised worm traces were seen, and brachiopods, bivalves, fish fragments and worm traces have been recorded in the Cynon Valley, where the band is 10 m thick. Crinoid debris has been found in the Taff Valley.

The Lower Cwmgorse Marine Band lies at the base of a coarsening-upwards cycle, which in the west ends with a coal locally named the Welsh Vein (not to be confused with the Welsh Vein at Ynysarwed, see above). The cycle ranges from 19 m at Cefn Coed to 22 m in the Dulais Valley boreholes and at Ynysarwed. It decreases to about 17 m in Pentreclwydau Brook, and still further to about 10 m near Maesgwyn. The Welsh Vein is present only in the west of the district; it is 0.58 m thick at Cefn Coed, 0.2 m in Dulais Valley No.2 Borehole, and was present in only one of the Treforgan Boreholes (No.3), where it was 0.31 m. The cycle above the Welsh Vein is up to 18 m thick in the Tawe Valley, thinning eastwards to 12 m and northwards to 8 m, and is capped by a thin coal carrying the Upper Cwmgorse Marine Band. At Cefn Coed, Ware (1930, p.465) recorded a few specimens of Anthraconauta phillipsii (Williamson) and some insect wings in the roof of the Welsh, and A. phillipsii and ostracods in the canneloid shale roof of a thin coal, 5.8 m above.

In the east of the district the Welsh Vein is absent and only one cycle, about 24 m thick, intervenes between the Pentre Rider and Upper Cwmgorse Marine Band. It culminates in the Hafod Seam, which is best developed in the Cynon Valley area, where it was known as the Clay, a seam about 0.65 m thick. In the Taff Valley it was called the Welsh and is generally thinner.

The Upper Cwmgorse (Cambriense) Marine Band, though ubiquitous, is variable in its development throughout the district. Locally it comprises three distinct layers of marine strata spread over as much as 18 m, similar to the occurrence at its type locality of Cwmgorse (Dix and Trueman, 1928). At Cefn Coed it is only 0.6 m thick resting on a thin seatearth with coal streaks, but contains a rich fauna in soapy blue-grey mudstone including crinoids, brachiopods (including Lingula and Orbiculoidea), gastropods (including Loxonema and Euphemites), marine bivalves (including Phestia and Dunbarella) nautiloids (including orthocones and Metacoceras), and well preserved goniatites (including the eponymous Donetzoceras cambriense Bisat). In only one of the Dulais Valley boreholes was the fauna as rich, and in the others the paucity of fossils made precise correlation uncertain. In the Neath Valley also, the marine band is impoverished, and shows one to three separate marine phases. The tributary ravines provide good sections; as for example above Craig Nedd [SN 8235 0326], with two marine phases separated by 9 m of barren siltstones; in Rheola Brook [SN 8321 0577], one Lingula-bed resting on plant-bearing mudstones; Pentreclwydau Brook [SN 8468 0570], [SN 8468 0644], one marine bed; and Crugau Brook [SN 8290 0437], three marine beds in 11 m of measures. At Rhigos the marine band is 4 m thick with Lingula, 'Estheria', P. ophthalmoides, fish debris and fucoids'. It occurs about 10 m above the correlative of the Hafod Seam, the roof of which has yielded Anthraconauta phillipsii and P. ophthalmoides. In the Cynon Valley the marine roof of the Clay Seam is about 7.6 m thick. The base is marked by a thin dark siltstone which passes up into dark shaly mudstone with pyritised burrows, paling upwards into grey mudstone with worm trails; 3.6 m above the base dark coarse silty mudstone with L. mytilloides again passes successively into dark grey and then paler grey mudstones with L. mytilloides, P. ophthalmoides and other burrows. Boreholes in the Taff Valley have proved 1.8 m of beds in which a lower phase contained crinoid debris, fish fragments and worm burrows, and a higher one contained P. ophthalmoides.

Upper Coal Measures or Pennant Measures (Late Westphalian C and Westphalian D)

Up to 670 m of Pennant Measures are present within the district. A large part of the sequence is made up of thick developments of massive Pennant sandstones, which form the high ground north-west of the Neath Valley and the bold escarpment between the Neath and Cynon valleys near the central southern margin of the district. To the south of the escarpment lies the high plateau of the central part of the coalfield, only the northern part of which lies within the district. Pennant Measures also crop out on the interfluve between the Cynon and Taff valleys, between Aberdare and Merthyr Tydfil, and on Mynydd Cilfach-yr-encil in the extreme south-east of the district. (Figure 9) shows the generalised vertical section.

Llynfi Beds

The Llynfi Beds are almost 200 m thick in the Dulais Valley in the south-west of the district where they consist largely of Pennant sandstones. They thin eastwards as the proportion of sandstone decreases, and they are about 120 m thick on the east side of the Neath Valley, south of Cwmgwrach. Further reduction to about 90 m takes place northwards towards Banwen and eastwards to the Cynon Valley, and this thickness is maintained over the Taff Valley where the beds are predominantly argillaceous.

In the west, two groups of coals lying towards the base and top of the formation are separated by a thick development of Pennant sandstone—the Llynfi Rock. Here the Lower Pinchin group of coals, below the Llynfi Rock, equates with the Blackband-Tormynydd group of coals in the east of the district. The Upper Welsh and Upper Pinchin coals lie above the Llynfi Rock in the west, and together with the Paynes and Pantrhydydwr seams, are broadly equivalent to the No.3 Rhondda-Taldwyn-Gilfach group of coals in the Taff Valley.

Ware (1930) described the sequence at Cefn Coed, and the Dulais Valley boreholes provide a more recent detailed record of the beds. The Lower Pinchin Seam lies about 18 m above the Upper Cwmgorse Marine Band at Cefn Coed, and 8 m above it in the Neath Valley. It is about 0.75 m thick, and is currently worked in a number of private small mines on the western side of the Neath Valley. It may correlate with the White Seam of the Cwmafan-Maesteg area (Woodland and Evans, 1964, p.66) in which case the two thin coals at 333 in and 326 m in the Cefn Coed No.2 Shaft would be the Jonah and Tormynydd. A flora in the roof of the Lower Pinchin is particularly rich in neuropterids. The Llynfi Rock, which lies between the Tormynydd and the No.3 Rhondda in its type locality in the Maesteg area, is about 86 m thick at Cefn Coed, thinning to 58 m in the Dulais Valley No.2 Borehole. It is a typical Pennant subgreywacke and is overlain by the Upper Welsh Seam, 0.57 m thick in Cefn Coed No.2 Shaft, followed by an 8 m coarsening-upwards cycle ending in seatearth and a coal, 0.18 m thick. In the Dulais Valley No.2 Borehole the Upper Welsh is associated with argillaceous strata some 26 m thick, and has a three-leaf section totalling about 0.6 m; Anthraconauta phillipsii occurs in mudstones both above and below the seam. The roof mudstones contain a rich flora at Cefn Coed, where neuropterids and Sphenophyllum emarginatum Brongniart are particularly common.

Massive Pennant sandstones, 15 m thick, overlie the rider seam to the Upper Welsh, and these are succeeded by the Upper Pinchin Seam. West of the Neath Valley this seam ranges from 0.75 to 1.1 m in thickness, and it has been worked from small adits in the Rheola Forest from both the Dulais and Neath valley sides. The uppermost beds of the Llynfi group are much obscured at outcrop by thick head and landslip from the overlying Pennant scarp, and detailed sections are scarce. In the Cefn Coed shafts three coals including the Pantrhydydwr, which may be a split from the No.2 Rhondda above, were recorded, in 45 m of mainly argillaceous strata. These seams have been seen intermittently at outcrop in the Dulais and Neath valleys, but none has been worked to any extent.

The Llynfi Beds crop out under the steep Rhondda Beds escarpment from the Neath Valley to Aberdare, but they are generally hidden beneath a thick cover of scree and head. Two streams below Craig Isaf [SN 8798 0349] provide sections, and the Llyn Fach Borehole [SN 9045 0364] proved the sequence in the central part of the district. Of the 100 m of Llynfi Beds drilled, the upper 40 m were mainly mudstones and the lower 60 m mainly sandstones.

In the Cynon-Taff area the Llynfi Beds are mainly argillaceous and correlation of several generally poor coals with those in the sequence in the west is obscure. No Pennant sandstone equivalent to the Llynfi Rock appears to be present on the east side of the Cynon Valley or in the Taff Valley. Here, however, a 6 m quartzitic sandstone lies close above the Upper Cwmgorse Marine Band, but this lenses out westwards at Bryn Pica, and southwards, indicating a north-easterly source, in contrast to the Pennant sandstones above, which are derived from the south and south-west. The Blackband Seam of Merthyr probably equates with the White Seam of Maesteg, and is about 0.3 m thick; two thin coals lying about 6 m above may equate with the Tormynydd and No.3 Rhondda. The 30 m of measures extending upwards to the Taldwyn Seam are mainly mudstones and appear to be devoid of coals. The Taldwyn has been worked in the Taff Valley, the most extensive workings being under Mynydd Cilfach-yr-encil [SO 0729 0308], where it is 1.5 m thick in several thin leaves. The overlying Gilfach Seam is thin, but has been worked on a small scale. Sandstones are present between the Gilfach and the top of the Llynfi Beds east of the Taff.

Rhondda Beds

These include the thick massive Pennant sandstones which form the great coalfield escarpment between the Neath and Cynon valleys, as well as the high plateau immediately to the south. They are about 210 m thick throughout most of the district, thinning towards the Dulais Valley where about 185 m are present.

Over the central and western parts of the district the beds consist of a number of discrete developments of typical Pennant sandstones, interrupted by comparatively thin but persistent mudstones, most of which carry thin and generally uneconomic coals towards their tops. This alternation of thick hard sandstone and thin soft mudstone gives rise to a very characteristic landscape of 'scarp and slack' features and the distinctive terraced aspect of the South Wales valley sides. In the east of the district, and particularly in the Taff Valley the upper part of the formation is much more argillaceous in character and mudstones with fireclays and thin coals predominate. 'Red beds', comprising grey-green, red-brown and purple mudstones with thin interbedded sandstones are present to a minor extent among these uppermost beds in the Taff Valley where they represent the western limit of a more extensive development of 'red beds' in the Llynfi Beds and Rhondda Beds of the east crop (Blundell and Moore, 1960; Downing and Squirrell, 1965; Kelling, in Owen, 1974). Many of these mudstones are oxidised palaeosols with a seatearth fabric, containing sphaerosiderite but lacking discernible rootlets.

The No.2 Rhondda Seam is a single coal, 0.6 to 0.8 m thick, in the east of the district, where it was worked as the Castell-y-Weiver or Saron coal in the Troedyrhiw area south of Merthyr Tydfil. It appears to be absent along the escarpment south of Hirwaun for about 9 km between Aberaman and Llyn Fawr, presumably cut out by the overlying conglomeratic sandstone. It was worked extensively beneath the precipitous crags of Craig Isaf, west of Llyn Fach, where it had the following section at the British Rhondda Colliery: rock roof, coal 0.38 m, stone 0.02 m, inferior coal 0.10 m, clod 0.18 m, coal 0.71 m, fireclay floor. To the west the seam appears to split, but whether into two or three coals is not clear. According to Robertson (1933, p.172) the No.2 Rhondda is represented by three distinct seams at Glyncastle No.2 Pit: 'No.2 Rheola' Seam, with three thin partings, 1.0 m; argillaceous measures 10.7 m; 'No.2 Rhondda' Seam, with two partings, 0.66 m; argillaceous measures 8.33 m; 'No.2 Rhondda (Tyra Level)' Seam, coal 0.56 m, clod 0.2 m, coal 0.20 m. The top coal is known as the Ynysarwed on the west side of the Neath, and the bottom coal is probably the Pantrhydydwr. Evans (in Woodland and Evans 1964, pp.188, 209) states that the No.2 Rhondda at the head of the Corrwg Valley (which has the same section as at British Rhondda) splits southwards and south-westwards into the Field Vein and Wernddu, two distinct seams separated by 10 m or more of mudstones. The Wernddu is underlain by a thin coal, the Wernpistyll Rider, which appears to take no part in the split, and which Evans correlates with the Pantrhydydwr at Resolven. There is little doubt that in the ground west of the Neath the Ynysarwed Seam, lying at the base of massive Pennant sandstones, equates with the upper part of the No.2 Rhondda and for this reason is taken as the base of the Rhondda Beds. The seam is over 0.6 m thick in Rheola Forest, and it has been extensively worked. The section is variable at Cefn Coed, but in No.1 Shaft 1.2 m of coal with a 5 cm parting was recorded •by Ware (1930).

Thick massive Pennant sandstones everywhere overlie the No.2 Rhondda Seam. They range from about 40 m in thickness in the Taff Valley, to about 30 m on the escarpment south of Hirwaun to over 60 m in the south-west of the district. The base of the sandstone, in particular, is very conglomeratic with abundant white quartz pebbles, up to 5 cm across, together with many intraformational clay-ironstone and mudstone pellets and 'rafts' and fragments of coal.

Above the No.2 Rhondda sandstone in the east of the district there are almost 125 m of mainly argillaceous measures within which three seams, the No.1 Rhondda, the No.1 Rhondda Rider and Tyla Court have been mapped in the Cynon -Taff area. Boreholes at Aberfan proved several thick seatearths below the Tyla Court for some 46 m, with reddish brown and purple palaeosols prominent in the basal 27 m. Pennant sandstones above the Tyla Court are succeeded by 4.8 m of mudstones with Anthraconauta phillipsii, ostracods, fish debris and Spirorbis. The seatearth of the Brithdir Seam lies above.

Immediately above the top of the escarpment crags above Llyn Fawr, a 'slack' with about 25 m of largely argillaeous beds has been mapped. These measures include four coarsening-upwards cycles, the lowest of which carries a 0.7 m coal which may correlate with the Daren Rhestyn Seam of the Pontypridd district. A. phillipsii, associated with ostracods occurs in mudstones near the base of the cycles. This group of measures is faulted down into the face of the scarp above Llyn Fach.

The No.1 Rhondda is thin above Llyn Fach, comprising only 0.13 m of coal with a thin parting, but it thickens westwards and, in old crop workings around Mynydd Resolfen an average section is: coal 0.6 m, parting 0.07 m, coal 0.3 m. The roof, where mudstone, contains A. phillipsii and ostracods. Locally sandstone, 6 to 9 m thick, overlies the coal.

On the west side of the Neath Valley 6 m of mudstone above the No.2 Rhondda sandstone contain two thin coals evidently equivalent to the Daren Rhestyn and associated minor cycles above Llyn Fawr. About 21 m of Pennant sandstone separate these coals from a thin two-leaf seam, which is presumably the No.1 Rhondda.

The sandstone between the No.1 Rhondda and No.1 Rhondda Rider is lenticular to the west of the Neath Valley, but eastwards to the Taff it is continuous and upwards of 30 m thick. The No.1 Rhondda Rider can be mapped throughout the district by virtue of its associated 'slack' feature, and is exposed in a few places in Rheola Forest, where it is up to 0.36 m thick. It is thick enough to have been worked on a small scale in the Taff Valley. In the west of the district the No.1 Rhondda Rider is separated from the Brithdir Seam by about 70 m of measures, mainly Pennant sandstone.

Brithdir Beds

Brithdir Beds are present in their entirety only over a small area between the Gyncorrwg Fault and the River Neath, where they are about 175 m thick and consist predominantly of Pennant sandstones.

The Brithdir Seam has been worked intermittently and on a small scale along its outcrop in Rheola Forest between the Neath and Dulais valleys, but there are no details of its section. it is absent over much of Twyn Corrwg Fechan to the east of the Glyncorrwg Fault, where it is washed out, and it does not appear again until the east side of the Cynon Valley at Cefn-pennar. It crops on both sides of the Taff Valley above Troedyrhiw where it has been widely worked. On the west side of the valley the seam consists of 'two thin coals separated by a thick clod' worked from small day-levels, 1200 m north-east of Cnwc. On the east side, where it has been worked out beneath Mynydd Cilfach-yr-encil and Penddeugau, it is 1.2 m thick with two partings.

About 50 m of beds, largely Pennant sandstone, separate the Brithdir from a thin coal, known as the Graig to the west of the Neath Valley. Old crop-workings mark the position of this coal above Sam Helen. This seam is not present east of the Neath Valley where about 50 m of Pennant sandstones separate the Brithdir from the Brithdir Rider, a seam which lies about 27 m above the Graig west of the Neath.

The Brithdir Rider is a thin and impersistent coal, overlain west of the Glyncorrwg Fault by about 85 m of Pennant sandstone. Above lie about 9 m of mudstones underlying the Wenallt Seam, which are seen only in a small area west of the fault.

The Brithdir Beds have been proved by drilling on the west side of the Taff Valley above Aberfan (just to the south of the district). About 145 m of Pennant sandstones above the Brithdir are overlain by 73 m of mudstones containing up to 9 thin coals, named the Cefn Glas group, the approximate equivalent of the Wenallt and its associated seams. 'Mussels' including A. phillipsii and A. tenuis occur in the roofs of most of these thin coals, and ostracods and Planolites are common above the lower ones. The top of the Brithdir Beds is placed at the top of these mudstones.

Hughes Beds

These are incomplete and present only in a small area west of the Glyncorrwg Fault. They consist largely of Pennant sandstones, but contain two thin coals with associated mudstones and seatearths—the Wenallt Seam at the base, and the Mountain Seam lying 50 m above. Up to 11 m of mudstones underlie the Mountain Seam immediately west of the fault. No details are known of these coals within their limited outcrop in the district, but Woodland and Evans (1964, pp.223–225) gave details for the area immediately to the south. Pennant sandstones overlying the Mountain Seam and forming the summit of Cam Brynllydan are the youngest strata preserved within the district.

Chapter 6 Structure

Earth movements belonging to four different periods can be recognised in the rocks of the Merthyr Tydfil district. The first three are represented by unconformities of limited extent. The first may be regarded as a dying phase of the Caledonian orogeny, and resulted in widespread uplift and erosion in Middle Devonian times, followed by renewed sedimentation in the Upper Devonian. Echoes of this movement are seen in the relatively minor unconformity at the base of the Grey Grits. In early Carboniferous times Sudetic movements presaged the main Hercynian (or Armorican) orogeny, which was responsible for most of the tectonic disturbance throughout the district. Comparatively small movements are indicated by the unconformities within the Dinantian, and there is evidence of more widespread uplift and erosion, especially in the east, towards the end of the Dinantian and before the deposition of the main mass of Namurian rocks.

The Hercynian orogeny post-dates all the rocks seen in the district. It probably commenced in the late Carboniferous and culminated in the early Permian, with consequent erosion continuing into the early Triassic. By the time sedimentation was renewed to the south in late Triassic times, the main tectonic shape of the coalfield had been established and its present topographic outline largely formed.

The Merthyr Tydfil district lies on the broad northern limb of the synclinal basin of the South Wales Coalfield. (Figure 10) shows the main structural elements. The predominant feature is the general southwards dip, interrupted by two sets of contrasting structures lying approximately at right angles to one another:

1.      the entire district is cut by swarms of steeply inclined so-called 'normal' cross-faults, aligned in general NNW–SSE.

2.      two zones of disturbance, made up of tight steep folds, associated with confined lines of parallel faults, aligned approximately WSW–ENE—the Neath Disturbance and the Cribarth Disturbance. Major folding is confined to a broad anticline in the Cwmgwrach–Rhigos area, and the complimentary Cwmgwrach Syncline.

In addition to these large-scale structures, the Lower Coal Measures and the Middle Coal Measures are affected by multitudinous relatively small-scale folds and dislocations, resulting from the generally incompetent nature of the argillaceous sequence.

A general variation of dip can be observed across the district, changing from west-south-westerly in the west, through southerly throughout the central area to south-south-easterly in the east. In general the regional dips range from 5° to 10°, locally they steepen to as much as 20°, and there are considerable aberrations of both direction and angle in the neighbourhood of faults and folds. On the flanks of the Cribarth Disturbance dips change from south-southwest to south-east, at right angles to the main structure, and this is especially marked on the south side, where southeasterly or south-south-easterly dips are seen for several kilometres away from the axis. This is also the case on the south side of the Neath Disturbance, where several shallow folds lie parallel to the line of disturbance. On the north side of the Cribarth Disturbance the steep south easterly dips on the immediate flank of the structure quickly revert to the general south-south-westerly pattern when traced away from the disturbance.

Cross-faults

East of the Tawe Valley these faults are mainly aligned in parallel swarms in a NNW–SSE direction. In the extreme east they have a more south-easterly trend, and in the northwest, they strike N–S. For the most part they throw down west. Throws up to 150 m occur, but usually they range from 15 to 75 m. In general throws diminish from south to north, and most of them die out in the Brownstones. The faults are almost all steep with hades ranging from 50° to 70°. Except for two examples in the upper Neath Valley, where horizontal slickensides were seen on mineralised north-northwesterly faults in Basal Grit, all the faults observed showed vertical slickensides.

There is evidence of differential folding along some of the cross-faults. For example, the Cwmgwrach–Rhigos fold belt is confined to the west of the Hirwaun faults; gentle folding is similarly confined to the west of the Glyncorrwg Fault at Maesgwyn. However, most of the movement on the faults was vertical, with little strike-slip component. Some connection between these faults and the coalfield basin evolution is suggested by their decreasing throws away from the Coal Measures, and by their consistent downthrows westwards into the deeper parts of the basin.

The Neath Disturbance

This is the most important structure in the district. It is a narrow zone of intense mainly monoclinal folding accompanied by trough faulting, along a straight course from the Neath Valley east-north-eastwards to Cwm Callan on the north-east margin of the district. (Strahan and others, 1904, pp.82–92; Owen, 1954; Owen and Weaver, 1983, Robertson, 1933, pp.206–209). Owen (1954, pp.333–358) concluded that it was the result of Armorican compressional forces from the south being resolved along a deep-seated sub-Devonian Caledonoid structure, and that the main late-stage faulting was at least partly transcurrent with sinistral shift detectable by displacement of some of the cross-faults.

Over most of its course the net effect of the dislocation is a downthrow to the north with major displacement of the Carboniferous rocks in the centre of the district. This, however, is reversed in the Neath Valley south-west of Aberpergwm, where the net displacement is down to the south.

The disturbance is simplest where it crosses the Old Red Sandstone in the east. At Bryniau Gleision ((Figure 11), section 1) it consists of a NE–SW curved fault, cut off by the north-westerly Tredegar Fault. It continues west-southwestwards as two diverging master faults—the Dinas and Coed Hir faults. The southerly Dinas Fault, with a downthrow to the north, forms a shatter-belt in the Old Red Sandstone along Cwm Callan. The Coed Hir Fault throws south and the combined effect is to produce a graben of uppermost Old Red Sandstone and Dinantian flanked by Brownstones in a belt up to a kilometre wide extending some 9 km to the west-south-west. Steep northerly dips up to 80° are seen in the Brownstones adjacent to the Dinas Fault.

At Penderyn the major element is still the Dinas Fault, with the Coed Hir Fault some 750 m to the north. ((Figure 11), section 2). At Craig y Dinas the main fault is a shatter-belt in the limestone, associated with a tight anticlinal fold with steep or vertical limbs (Plate 6) giving way rapidly to near flat-lying beds to north and south ((Figure 11), section 3).

A short distance west of Craig y Dinas the limestone dips beneath the Namurian, successively higher beds appearing in the Mellte as they curve around the nose of a west-southwesterly pitching anticline with tight folds along its axis. The anticline dies out in Farewell Rock at Ynyscambwll and is replaced by a complex syncline in the Lower Coal Measures. The Dinas Fault appears to be cut off 300 m west of Craig y Dinas by a subparallel fault throwing down north and to the west it is this fault that forms the southern boundary of the Neath Disturbance.

South-west of Glyn-neath the line of dislocation is hidden beneath the drift in the valley bottom, but the effect of the disturbance is clearly seen in the acute monoclinal folding in the Lower and Middle Coal Measures on the lower slopes of the valley above and below Cwmgwrach. These folds appear to combine to form a sharp faulted arch across the valley, not unlike that at Craig y Dinas, though broader.

In the Resolven area there is clear evidence of major dislocation. The crop of the No.2 Rhondda is mapped on the south side of the valley, reaching the valley floor just north of Resolven village; on the north side it descends to the valley bottom near Clyne, nearly 4 km to the south-west. This indicates a throw of about 175 m to the south.

The Cribarth Disturbance

The Cribarth Disturbance is a linear compressional stucture similar to the Neath Disturbance. It is named after Cribarth, the conspicuous anticlinal ridge of Carboniferous Limestone that lies to the south-west of the Tawe Valley at Craig-y-nos Castle [SN 8406 1530] (Plate 7). North-eastwards the disturbance can be traced along a line of faults and associated folds into the Old Red Sandstone country of Fan Gihirych and beyond into the upper reaches of the Senni Valley. Southwestwards from Cribarth it passes into the coalfield as the Tawe or Swansea Valley disturbance (Trotter, 1947; Weaver, 1975).

At its north-eastern end the disturbance runs through a hollow between Fan Gihirych and Fan Fraith, where a block of Plateau Beds and Grey Grits is let down into the Brownstones between two faults about 200 m apart. The overall throw from one side of the trough to the other is scarcely discernible.

To the south-west the disturbance is seen in the tributaries to Nant Twyni [SN 864 173], where a number of faults intersect and give rise to variable dips in the Old Red Sandstone rocks. Farther south-west the faulting splays out to form another block of Plateau Beds let down south against Brownstones. In Nant Byfre the faults are associated with sharp monoclinal and synclinal folds over a width of about 0.7 km ((Figure 12), section 1). Between Nant Byfre and the Tawe Valley at Craig-y-nos a sharp monocline in the Dowlais Limestone has an almost vertical north limb form ing a steep ridge plunging into the river, and a gentle southern limb with dips of 15° to 20°. A complementary syncline forms low ground to the north, and this is cut by faults continuous with those on the north side of the Nant Byfre complex. The northern limb of this syncline is almost monoclinal, with southward dips of 80° seen below the Danyr-ogof caves.

To the south-west of Craig-y-nos the high whaleback ridge of Cribarth ((Figure 12), section 2) is a continuation of the north-facing monocline, east of the river. It has a steep northern limb with dips of 65° or more, and a more gently dipping southern limb, interrupted locally by minor folds. The Llandyfan Oolite, which forms the main ridge, is shattered and recrystallised along numerous low-angle shears. The syncline to the north is an extension of the monoclinal fold below Dan-yr-ogof, though it is a tighter structure with vertical limestone on either limb and Basal Grit cropping out along its axis.

The tight fold pattern ends abruptly against the Pwllau Bach Fault [SN 823 136] which crosses the disturbance north of Abercrave. To the south-west, the structural pattern is markedly different. Here there are a series of major faults, which splay out and increase in throw towards Ystradgynlais, accompanied by some folding and numerous small-scale incompetence structures. The main fault, called the Tawe Valley Disturbance, bounds the complex on the north-west and has a downthrow south of about 140 m at Ystradgynlais. Another fault, the Cwmddu Fault, lies not far to the south, and throws north, producing a tilted graben ((Figure 12), section 3). At Abercrave the graben has little net effect on the measures on either side, but to the south-west the overall southward throw grows to about 120 m.

There is no strong evidence in either the Neath or Cribarth disturbances of late transcurrent movement, which would require consistent matching of cross-faults across them (e.g. Owen, 1954). Instead, the faults along the disturbances and the cross-faults form a reticulate pattern and appear to be of about the same age. The simple graben structures in the Old Red Sandstone are replaced southwestwards by monoclinal folds in the Carboniferous. It is probable therefore that these folds are the expression in the Carboniferous cover rocks of mainly vertical movements in the basement rocks.

Minor structures in incompetent strata

A multiplicity of comparatively small-scale low-angle thrusts, overlaps and lag-faults affects the weak predominantly argillaceous strata of the Lower and Middle Coal Measures, chiefly that part of the sequence containing the thick coals. Individually the structures range from little more than bedding plane slips to dislocations with displacements rarely up to 60 m. These structures are arranged in two main directions, one set striking WNW, and the other WSW, with the throw of the overthrusts to the north and that of the lag faults to the south. There is also a minor set of north-south structures at Rhigos and in the Dulais Valley. The two main sets are similar to structures in the Pontypridd district which Woodlands and Evans (1964, p.263) considered to have formed during the earlier phases of Armorican compression.

Although developed in varying degree throughout the district these crush-structures tend to be more abundant, and to affect more of the sequence generally, towards the west. In many areas they produce such wholesale dislocation of the coal seams as to render them virtually impossible to work, particularly by modern automated mining methods. Commonly movement confined more or less to the plane of a seam has resulted in wide bands of abnormally thick coal, all too often unworkable underground by virtue of its weakness and the incorporation of large amounts of 'rashings' (highly comminuted slickensided mudstone) derived from roof or floor; individual 'pods' of coal have been known to reach several tens of metres in thickness. A complementary feature is the squeezing out of coal along the bedding, so that it becomes attenuated or pinched out. This has produced numerous examples of elongated 'wants', areas where the coal is missing altogether, and which were described on old mine-plans as 'washouts'. Overlap thrusts and lag-faults produce rapid variation in the overall thickness of groups of measures, with thrusts resulting in thickening and lag-faults thinning.

Incompetence compressional structures have least effect in the Taff and Cynon valley areas, where many of the coals have long since been worked out. 'Washouts' and areas of thick coal are, however, associated with the Nine-Feet Seam.

In the opencast workings of the Rhigos–Cwmgwrach area the measures between the Nine-Feet and the Two-Feet-Nine were intensely disturbed. Northward-directed thrusts and rolls associated with lag-faults had maximum effect in the Six-Feet and Four-Feet seams, and the two coals were transformed into an indistinguishable series of structural slices. Not only were the seams converted into dirty mixtures of coal and rashings, but the associated mudstones were so highly slickensided that normal bedding was hard to recognise. Lag-faulting in places produced wide 'wants', as for example in the Nine-Feet, where a gap 25 m wide proved devoid of coal with the strata above folded into rolls and box-type anticlines. Overthrusts and rolls in the Two-Feet-Nine group are directed east, almost at right angles to those in the Four-Feet and Six-Feet. In the lowest coal, a shear zone 5 m thick lay between undisturbed measures over much of the site, and further bedded disturbed zones occurred at higher levels.

In Cwmgwrach colliery, a lag-fault in the Nine-Feet is overlain by a series of overthrusts in the Cornish, as at Rhigos. North-directed WSW and WNW thrusts affected the seams from the Cnapiog to the Lower Peacock in the north of the Rhigos area.

On the west side of the Neath Valley and in the Dulais Valley serious disturbances occur throughout the Lower, Middle and basal Upper Coal Measures, with maximum development in the Six-Feet. These are well exposed in the ravines of the Rheola and Pentreclwydau brooks. At Rheola [SN 8375 0467] a WNW-trending 500 m wide zone of steep, almost vertical beds includes most of the measures between the Cefn Coed and Lower Cwmgorse marine bands. Similar zones occur in the Pentreclwydau ravine. In the Dulais Valley the structures include a north-directed monoclinal roll in the Four-Feet at Seven Sisters, where the vertical limb contains almost 40 m of coal (Robertson, 1933, p.197).

At Maesgwyn opencast site [SN 8600 0835] a series of north-directed WSW-trending rolls with amplitudes up to 15 m were encountered in the Six-Feet. Step-faults in the Four-Feet terminated upwards and downwards on sheared bedding planes. A discontinuity lay 24 m above, the beds below being broken and jumbled and meeting the discontinuity at an acute angle, the beds above being parallel to the break.

In Drym opencast site [SN 8347 0977] the Cornish, Lower Six-Feet and Upper Six-Feet were affected by independent sets of minor north–south whale-backed rolls with steep east-directed limbs. Bedding-controlled disturbances, mainly overthrusting to the east, occurred in the beds below the White Four-Feet and above the Stwrin.

Onllwyn opencast site [SN 8395 1148] exposed a NNW-trending thrust zone in the Bluers seam with up to thirty sheets and pods of coal stacked in a northerly direction. The strata above were locally broken into a mélange of blocks and appeared to be overstepped by the rootlet bed of the Lower Peacock. The thrust and lag-fault swarms are considered to be the result of the compression of incompetent measures sandwiched between competent formations during the Armorican orogeny in an area north of the main orogenic front. However, there are features not easily explained in terms of simple south-to-north compression. First, there is no consistent trend to the structures. Both 'Caledonoid' (WSW) and 'Variscan' (WNW) trends occur, the former predominating north of the Neath Disturbance, the latter to the south. North-south and NNW trends are also common, as at Rhigos. Secondly, the association of low-angle normal or lag faults and thrusting is difficult to explain entirely in terms of a compressional regime. Gravitational sliding prior to the main Armorican compression, as in the case of the Fernhill and Jubilee slides (Woodland and Evans, 1964, p.272), may provide a better explanation for many of the thrusts and lag-faults, extensive thrusting being associated with the Jubilee Slide at depth. Reactivation of synsedimentary growth faulting may also account for some of the strata-bound disturbances of the Merthyr district.

Chapter 7 Quaternary

Glacial deposits

In the Merthyr Tydfil district there is clear evidence of only the last, late Devensian glaciation. This virtually obliterated all signs of any previous ice activity, but left behind widespread deposits, mainly of melt-out till (Boulder Clay), now preserved over all but the highest ground (Plate 8).

The main gathering ground for the ice was the high tract of Old Red Sandstone rising to between 600 and 886 m in the Brecon Beacons along the northern margin of the district. From here the ice-sheet spread mainly south towards the coalfield, little affected by local topography, until it met the bold escarpment of the Pennant at Craig-y-Llyn, south and west of Hirwaun (Plate 9). Here the ice split into two major lobes, one passing SSE into the coalfield through the gaps of the Cynon and Taff valleys, and the other WSW down the Neath Valley. Everywhere it left great spreads of till, characterised by its Breconshire imprint of reddish colour and abundant Old Red Sandstone boulders and cobbles.

On the high ground in the north the till is composed wholly of Old Red Sandstone debris, with boulders only little rounded and with a sandy and gravelly matrix. Followed southwards it progressively takes on features of the younger rocks over which the ice moved. First, Carboniferous Limestone and Millstone Grit quartzitic erratics were incorporated, and finally the mudstones of both the Namurian and Coal Measures imparted a stiff clayey, more grey, character to the matrix. Over the long dip slopes of the Old Red Sandstone, Carboniferous Limestone and Basal Grit a thin veneer of unsorted till was laid down, but in the hollows and valleys of the Lower and Middle Coal Measures, drift up to 30 m or more was deposited. Some morainic form is preserved in places, as near Hirwaun, and drumlins occur between the Neath and Tawe valleys, but more generally the till is featureless and smoothed by periglacial action. Fresh striations, particularly on the pavements of the Basal Grit, testify to the general southwards movement of the ice.

Local drift was deposited by tongues of ice moving downslope from the Pennant scarp to merge with the main ice, and similar tongues emerged from Craig Clwyd Fechan and Craig Clwyd in Rheola Forest on the north side of the Neath Valley above Resolven.

As the glaciation waned, recessional halt moraines were left in the Neath Valley at Tonna and Clyne, just south of the district, and a glacial lake was impounded behind them as far up-valley as Cwmgwrach (Anderson and Owen, 1979). The lake clays fill a rock-basin more than 64 m deep at Resolven. Lateral morainic gravels (Glacial Sand and Gravel) extend up the valley on the south side as far as Resolven. At Hirwaun Ponds thin banded clays may indicate the former presence of a glacial lake.

The upper Taff valley is filled with poorly sorted gravelly drift to 15 m below river level at Llwyn-on Reservoir and outwash gravels and possibly kame terrace deposits line much of the valley sides. There are also small deposits of glacial gravels in the Mellte Valley at Ystradfellte, and in the upper Tawe Valley between Abercrave and Craig-y-nos.

Corrie glaciation marked the final stages of ice activity. Morainic ridges of angular material are piled in concentric form across the cwm exits, backed by marshy hollows and even small lakes, as at Llyn-fawr and Llyn-fach beneath the Pennant scarp.

Periglacial phenomena are widespread. The till commonly shows a crude downhill stratification in its uppermost layers, indicating solifluction. Gravity accumulations of 'Head', ranging from true end-glacial solifluction deposits to still-forming hillwash and scree, are ubiquitous on all the valley sides, commonly reaching 8 m on the lower hill-slopes and in hollows on the Old Red Sandstone. Head is equally widespread on the Carboniferous outcrops. In the Pennant uplands, sandy head covers all but the upstanding crags, where it merges with thick scree. This down-slope instability reaches its maximum expression in the numerous landslips described below.

Landslips

Landslips occur on the glacially oversteepened sides of the valleys, usually where thick, massive open jointed sandstones of the Pennant Measures and Brownstones overlie impermeable mudstones. They occur notably in Coal Measures rocks on both sides of the Neath Valley between Resolven and Pont Nedd Fechan, and in Cwm Gwrelych near Glyn-neath, in the predominantly argillaceous Namurian succession in the Pyrddin and Mellte valleys, along the main Pennant scarp south-west of Aberdare, and on the east side of the Cynon and Taff valleys at Blaen-nant and Troedyrhiw. The coalfield landslips have been described by the BGS Engineering Geology Unit (Conway and others, 1980).

Swallow-holes and solution-collapse features

Swallow-holes, sink-holes or dolines are inverted conical depressions common over much of the basal Namurian and Dinantian outcrops; some of them reach nearly 100 m across and are of considerable depth. They are caused by collapse into underground cavities formed by solution of the limestone. Those on the limestone outcrop commonly contain blocks of Basal Grit, indicating that they were formed while there was still a cover of that rock, while both these and those on the Basal Grit outcrop may contain glacial till. A few swallow-holes on the outcrop of the Shale Group must extend down to about 30 m. Individual swallow-holes are commonly aligned along the course of faults.

Major linear collapse features, up to 200 m wide and 1.6 km long, occur along faults on the Basal Grit moorland between the Tawe and Giedd valleys. These hollows filled with jumbled quartzite and peat are areas where the Basal Grit has collapsed into cavern systems in the limestone. Immediately to the north of the Basal Grit scarp a mass of collapsed quartzite fills an irregular hollow up to 400 m wide and 3 km long, south of Waun-Fignen-Felen and near Pwll Byfre.

Underground drainage and cave systems

Extensive caverns occur in the Carboniferous Limestone, chiefly in the Dowlais (Cil-yr-ychen) Limestone. The best known is the Dan-yr-ogof system with entrances on the west side of the Tawe, about 1 km south-west of Glyntawe [SN 8339 1602]. With a total passage length of 10 km, its course is marked on the surface by a line of shallow sinks along a fault. A large collapse extending north-westwards probably marks the course of a major underground stream carrying water from the Waun-Fignen-Felen sink to the resurgence at Danyr-ogof. Sink-y-Giedd [SN 8105 1790] also connects with the Dan-yr-ogof system. The Ogof Ffynnon Ddu system with a resurgence to the east of the Tawe [SN 8477 1508] extends 3.2 km to Pwll Byfre, which is the system's sink; the total passage length is in excess of 37 km. A plethora of smaller caves exists, the best known of which occur in the Nedd Fechan between Pant Mawr [SN 891 162] and Pwll Du [SN 912 123], and in the Mellte Valley below Porth yr Ogof [SN 928 124].

Fluvial deposits

Most of the larger valleys contain spreads of poorly sorted gravelly alluvium in their lower reaches; in the smaller valleys alluvium is much more sporadic, and in many of the smaller ravines none occurs. River terraces, consisting of gravel and sand, generally coarse with little matrix, occur intermittently along the rivers Taf Fawr, Mellte, Neath and Tawe. Alluvial fans occur in the main valleys, where steeply graded tributary streams have deposited their storm debris on entering the flat-bottomed valley. In the Neath Valley they occur about 6 m above flood level and may have formed as lake deltas rather than as alluvial fans (Anderson and Owen, 1979).

Peat

Hill peat is widespread on the high ground of the Old Red Sandstone and Pennant sandstones. The most extensive deposits lie on the mountain tops on either side of the Taf Fechan above the Taf Fechan Reservoir. On the west side the bog is virtually continuous over 6 km from Waen Wen through Waun Lysiog and Gwaun Nantddu to Cefn Crew, while on the east side large spreads cover Garn Fawr, Gwen Cerrig-llwydion and Waun-rydd. The peat is noticeably wasting, leaving exposed the bleached white sandstones of the Plateau Beds. Peat is equally widespread on the coalfield plateau, and includes wasting peat on the open moorland and still-growing peat in the hollows (Woodland and Evans, 1964, p.280). Patches of peat also occur sporadically in hollows on the glacial till. On Hirwaun Common peat is almost continuous, though it rarely exceeds 2 m.

Calcareous tufa

Glacial gravels have been cemented by calcareous tufa in the Tawe, Taf Fawr and Taf Fechan areas, where they flow over Carboniferous Limestone, producing strongly cemented aggregates. In the Taf Fawr, limestone scree from Darren Fawr has also been cemented.

Chapter 8 Economic geology

Coal

The coals of the district were described in some detail by Robertson (1933, pp. 213–220) and little can be added to this account. Adams (1967) and Thomas (1961) have reviewed the coal mining industry in South Wales.

The world demand for steam coal for ships' bunkers led to a rapid expansion of the coal mining industry in the mid and late 19th century, particularly in the Cynon and Taff valleys, from whence the coal was transported by canal and rail to Cardiff, Penarth and Barry for export to coaling stations the world over. Maximum production was attained just prior to the First World War, since when there has been a steady decline in deep mining. This decline has accelerated in recent years, and now only Tower Colliery, near Hirwaun remains in production. The anthracite field achieved maximum production in 1934 and with the increasing demand for smokeless fuel, there have been attempts to work the highly disturbed seams of the Neath and Dulais valleys from drift mines at Aberpergwm and Treforgan, but both were closed in 1986. There has recently been an expansion in small-scale private licenced mining from day-levels and drifts, and there are about forty small mines currently in work in the district.

Since the Second World War opencast mining has expanded along the outcrop of the Lower and Middle Coal Measures, and the last twenty years have seen dramatic developments in the scale of operations. Within the Merthyr Tydfil district a series of large open-pit complexes have been exploited, including Dunraven (Plate 10), Maesgwyn, Rhigos, Bryn Pica and Trecatty. Three sites are currently in production: Nant Helen, Maesgwyn and Ffyndaff.

Types of coal in the district (Extract from Robertson 1933)

This part of the coalfield yields house, steam and anthracitic coals. In the main productive portion of the Lower Coal Series the percentage of volatile matter varies from 12–14 at the eastern side of the district, to 6–7 at the western side. In the same distance and in the same direction the No. 2 Rhondda Coal shows a reduction from about 18 per cent. to 10 per cent.'

In the Taff and Cynon valleys the coals above the Two-Feet-Nine are distinctly bituminous in character, whilst those below are steam coals. Between Aberdare and the Vale of Neath all the seams are steam coals, being semi-bituminous above the Gorllwyn and semi-anthracitic in the lowest portion of the sequence. From the Vale of Neath westwards all the coals are anthracitic, the lower seems decidedly so ; whilst those from the Red Vein upwards are low-grade or semi-anthracites. Several of the seams in the Tawe valley are suitable for malting. The coals in general appear to reach the anthracitic composition before attaining the physical properties of anthracite, so that there is a tract lying between steam coal and anthracite in which the coal, though anthracitic in analysis, is brittle and has not the crushing strength of anthracite. It therefore collapses on the fire, breaking up into small fragments when the volatile matter is expelled, and will fall through the bars of an ordinary grate. This type of coal is illustrated more or less by the lower coals in the Aberdare area, for example the No. 2 Yard, Seven-Feet, and to a less extent the Gellideg. The difference in crushing strength of the coals in the Neath and the Tawe valleys respectively is very appreciable, though not reflected in any noticeable degree by analysis.

The variation in volatile content of the coals can be indicated in a general way by means of the accompanying diagrams, prepared from a considerable number of analyses (Figure 47) and (Figure 48). For comparison the ground east of the present district is included. It will be seen that the rate of decrease of volatile matter in a westward direction takes place in a similar way in all the seams. West of the point showing flattening of the curves at about 3o per cent. volatile (Sheet 232, Abergavenny Memoir, p. 119) there is an area in which the volatile content decreases rapidly westwards. This is followed by a flattening of the curve in the Rhymney and Dowlais area at a volatile content of somewhere about 12–14 per cent. Thence there is little further drop in volatile matter until the west side of the Aberdare area is reached, except in the lowest seams. Between Bwllfa and the Vale of Neath, however, there is a distinct steepening of the gradient (though information is rather scanty in this part) to an average of about 8 per cent, volatile matter. Thence, westwards, there is a gentle but steady drop to about 6 per cent. in the Ystradgynlais area.

The graph of the seams above the Two-Feet-Nine (Figure 48)—comprising the Gorllwyn, Graig or Red Vein and No. 2 Rhondda—is similar, the flattening out at a volatile content of 13–14 per cent. being noticeable in this case also. The corresponding features in the curves for the higher and lower measures are 4–5 miles apart, showing the general application of Hilt's Law as regards these two divisions. The difference between the higher and lower seams in the measures below the Two-Feet-Nine is small, though in general the lower seams have a smaller volatile content in any one vertical section. The main coals, however, are contained in a comparatively small thickness of strata.

1 For detailed discussion of the problem of anthracitization see A. Strahan and W. Pollard, The Coals of South Wales ' (Mem. Geol. Surv.), ed. 2, 1915 ; also sundry papers by other authors in Proc. S. Wales Inst. Eng.

Character and uses of seams

The following general notes on the various coals indicate their character and use.

Gam or Cnapiog Coal

In general a clean, good coal, but too thin to work at a profit at the present time. It is a steam coal in the Taff and Cynon valleys, but is one of the first seams to show anthracitic qualities in a westward direction. In the Vale of Neath, however, it is no more anthracitic than the overlying seams.

Lower Four-Feet, Gellideg, Bluers Group

A coal of good quality, and much worked in the Taff valley. The lower portion (the Gellideg) becomes very thin in parts of the Cynon valley, and is semi-anthracitic in the north-west part of that area, where it thickens out again. The upper portion (the Five-Feet) at Aberdare is decidedly more bituminous than the Gellideg, but on the west side of the unworked ground south of Hirwaun its equivalent, the Bluers, is anthracitic. It is a satisfactory coal from thence westwards, but thins to less than 2 ft. in places. The lower members of the group, namely the Rhyd and Grey, though poor in the Vale of Neath, improve and thicken to the north-west.

Bute, Peacock or Brass Coa

The coal of this seam is good in the Taff and Cynon valleys, but there is a clod parting which tends to make it dirty. The top portion is the better coal. Farther west it is a good, clean coal without definite parting. It is usually bright throughout, but shows some streakiness near its bottom in the Tawe valley, due to the presence of fusain. At Onllwyn also the top coal is usually the brighter of the two.

Nine-Feet or Big Vein Coal

This coal is of good quality and section in the Taff and Cynon valleys, but is practically worked out. Of all the seams in the sequence this is the one most subject to disturbances and barren ground. Near Hirwaun, where it enters the region in which overthrusting becomes more common, it gets rather softer, though it is still fairly good. The top coal is better than the bottom in this area. In the Vale of Neath the Big Vein is distinctly anthracitic, but not very hard, and subject to ' backs ' or ' riders ' (p. 220). Fusain is only rarely present in any quantity, and then mainly in the bottom coal. Westwards the seam becomes harder, the top portion still remaining the better coal, but this is in places absent through contemporaneous denudation. There is a little fusain in the bottom coal.

Red or Four-Feet-Cornish Coal

This coal is good just east of the present district, but poor in the Taff and Cynon valleys and little worked. In the Vale of Neath it has much fusain throughout, making it very streaky, but it improves somewhat westwards. The occurrence of cornish, which has a tendency to stick to the coal, makes it rather dirty in places, and the slips frequently run into the roof. At the west side of the district the coal is rather hard and streaky, and is not up to the Nine-Feet and Peacock in quality. It has, however, been extensively worked.

Six-Feet or Eighteen-Feet Coal

This is a good coal in the Aberdare and Merthyr area, though with a soft roof and some partings. Farther west it deteriorates and is very streaky in the Vale of Neath. The bottom coal is good here, but has much fusain near the floor ; the top coal is in general poor, irregular and very slaty, apparently owing to movement and intermixture of shaly matter during consolidation. This persists in the thick coal area between Glyn Neath and Onllwyn. Farther west, where the seam thins again, it retains these characters, being soft, shaly and streaky.

Upper Four-Feet or White Coal

This is one of the best coals in the Merthyr and Aberdare areas ; clean, bright and of good section. At the north end of the Vale of Neath it was formerly much worked as a steam coal. It becomes dirty with partings in places, but is quite a good coal elsewhere, though less anthracitic than the Nine-Feet. West of the Vale of Neath it deteriorates rapidly, being poor in quality, soft and high in ash.

Two-Feet-Nine Coal

This seam is decidedly more bituminous than those underlying it in the Abergavenny district (Abergavenny Memoir, p. 123, Elled Coal), and it retains this distinctive character westwards at least as far as Hirwaun. It is a good steam coal in the Merthyr and Aberdare area, and is a satisfactory house coal also. West of Hirwaun the seam (Stwrin Coal) is very poor in quality and thickness, and is not worked.

Gorllwyn Coal

This is a very persistent coal, yet it is only between the Taff and Neath valleys that it is thick enough to be worked. It is bituminous in character, but is also used as a steam coal. There is a clod parting in the middle of the seam in many places, tending to make the coal rather dirty. The overlying Gorllwyn-f ach Coal is also a good bituminous coal, used either for house or steam-raising purposes. It has only been found thick enough to work, however, in one small area.

Graig or Red Vein Coal

At Aberdare this is an excellent house coal, but it thins both east and west of that area. Where it again becomes of workable thickness, under Hir Fynydd, it is a semi-anthracite, rather soft, and streaky with fusain.

No. 2 Rhondda and Ynisarwed Coals

In the Merthyr area the No. 2 Rhondda is a bituminous coal, fairly good in quality but with a number of partings. On the east side of the Vale of Neath it is a poor quality steam coal, soft and dirty. Under Hir Fynydd the Ynisarwed is similar, but semi-anthracitic.

The accompanying table shows the character of each of the more important seams in the different parts of the district:

Seam Colliery or Locality Type Quality
No. 1 Rhondda Vale of Neath Steam Poor, dirty
No. 2 Rhondda Llewellyn Plymouth Bituminous —
No. 2 Rhondda C. and E. Steam Poor
No. 2 Rhondda Ynisarwed Steam Fair
No. 2 Rhondda Llwynon Semi-anthracite Sulphury
Castell-Weiver Llewellyn Cyfarthfa Bituminous Clean
Taldwyn Llewellyn Plymouth Bituminous —
Hafod Penydarren House Good
Hafod S. Dyffryn House Good
Graig Aberdare House Good
Red Vein Brynteg Semi-anthracite Fair, sulphury
Gorllwyn-fach Windber House or Steam Good
Soap Vein Dowlais Bituminous Poor
Gorllwyn Llewellyn Plymouth House —
Gorllwyn Werfa House Good
Gorllwyn Aberdare Steam or Smith's Good
Gorllwyn Bwllfa House or Steam Good
Elled Dowlais Steam or House Good
Upper Two-Feet-Nine Ll. Pl. and Ll. C. Steam Good
Two-Feet-Nine Werfa Steam Rather ashy
Two-Feet-Nine Aberdare Steam Good
Two-Feet-Nine Bwllfa Steam
Two-Feet-Nine Hirwaun Steam
Stwrin W. of Vale of Neath Anthracitic Very poor
Black Vein Dowlais Steam (loco.) Good
Driver Hirwaun Steam —
Driver N. British Rhondda Semi-anthracitic —
Upper Four Feet or Four-Feet Dowlais to Hirwaun Steam Good
Upper Four-Feet N. British Rhondda Steam to Anthracitic Good
Four-Feet White C. and E. Anthracitic Dirty
Four-Feet White Aberpergwm Anthracitic Good
Four-Feet White Dulais and Tawe Valleys Anthracitic Soft and ashy
Six-Feet Ll. Pl. and Ll. C. Steam Good
Six-Feet Aberdare Steam Good
Six-Feet C. and E. Anthracitic Fair
Six-Feet Vale of Neath Anthracitic Bottom good, Top fair
Eighteen-Feet or Upper Black Abercrave Anthracitic Fair, streaky
Red Taff Valley Steam Good
Red Aberdare to Hirwaun Steam Poor
Four-Feet-Cornish Neath Valley Anthracitic Fair
Four-Feet-Cornish Dulais and Tawe Valleys Anthracite Fairly good, hard
Rhaslas or Nine-Feet Taff Valley Steam Good
Nine-Feet Aberdare Steam Good
Nine-Feet Hirwaun Steam Fairly good, soft, top better than bottom
Nine-Feet Vale of Neath Anthracite Good
Nine-Feet or Big Vein Dulais and Tawe Valleys Anthracite Good, top better than bottom
Bute Taff Valley Steam Good
Bute Aberdare Steam Good
Bute Bwllfa Steam Fair
Yard, Peacock or Brass Neath, Dulaisand Tawe Valleys Anthracite Good
Lower Two-Feet- Nine Taff Valley Steam Fair
No. 2 Yard Aberdare Dry Steam Fair to good
Five-Feet-Six Taff Valley Steam Fair to good
Seven-Feet Aberdare Dry Steam Fair to good
Lower Yard Bwllfa Dry Steam Good
Lower Four-Feet Taff Valley Steam Good
Gellideg Aberdare Steam —
Seven-Feet Bwllfa and Hirwaun Very dry Steam —
Gellideg Bwllfa Semi-anthracitic Rather soft
Bluers Neath, Dulaisand Tawe Valleys Anthracite Good

Cornish

Associated with several of the coal seams there is a black, horny substance usually referred to as ' cornish.' In the western part of the district and in the lower coals generally, it shows no great amount of carbonaceous matter and is useless as a fuel. Even as far west as Llwynhelig and Bwllfa, however, the cornish associated with the Gorllwyn Seam burns with a luminous flame. Cornish usually occurs on top of the coal seam, though more rarely it is found at the bottom. It has a tendency to stick to the coal, and occurs in patches and lenses. It also contains fish remains, lamellibranchs, ostracods, etc. In all these characters and in general appearance it resembles cannel so closely that there can be no doubt that it is its equivalent, but the original cannel has been devolatilized in a similar way to the ordinary coals and in an increasing degree westwards. With increasing devolatilization, however, the cornish has become less cannel-like in appearance, losing its horny lustre to some extent and appearing more like a dense mudstone flecked with tiny specks of bright carbonaceous material. The seams with which it is most usually associated are the Red or Four-Feet-Cornish and the Gorllwyn or Blackband, but it is by no means absent in other parts of the succession.

Slips, Riders, etc

One of the most interesting features of the coal seams themselves is the development in them of a series of joints commonly known as slips. These occur in all the seams, at a distance apart which varies from 1–6 ft., but is most usually xxx ft. The distance varies with the nature of the seam. There is in general no displacement, crushing or slickensiding along the slip faces, although the strike of the jointing, except in proximity to structural disturbances such as the intra-Coal Measures movements, is always parallel to that of the main north-north-west to south-south-east normal faulting. The angle of the dip of the slips with reference to the bedding planes in areas of normal dip of the measures is 55°-57° to the west. This angle was measured in numerous seams at many different localities and found to be surprisingly constant in amount except near structural disturbances. In some instances, e.g. the Four-Feet-Cornish at Ystradgynlais, the slips are much lined with calcite or ankerite, but this is not general. In this seam the slips have a tendency to curve over to a flatter slope on top and to pass into the cornish or bast roof. Where the coal approaches the fold in Seven Sisters Colliery (Fig. 41, p. 198) the slips are closer together (1–1.5 ft.) but remain at the same angle ; whilst at every  second slip, on an average, there is a step down of 3–4 in. The cornish above the Cornish Coal at Seven Sisters is veined with calcite or ankerite, but not the coal itself.

The direction of the working face of the seam is usually determined by the slips, the face being kept parallel to the strike of the slips. In general, working in the opposite direction to that of the dip of the slips, i.e. on-the-back,' is preferred, as the coal overhangs if the seam is worked in the direction of dip of the slips, i.e.' on-the-face.' For the coal-cutter faces, e.g. in the Gorllwyn-f ach at Windber, working on-the-end ' is preferred, the face trending at right angles to the strike of the slips.

The ordinary fine cleavage or ' cleat ' of coal is not so well developed in the seams of this district as in the more bituminous coals, and it becomes still less conspicuous westwards. It does exist, however, to a certain degree, and appears to coincide approximately in direction with the slips, but not always. A sufficient number of precise observations on this point was not made, however, owing to the poor development of the structure.

Another type of structure is illustrated in the ' backs ' or ' riders ' which are common in this part of the coalfield. These are crush-fractures and frequently are inclined at an angle of about 45° to the roof and floor of the seam. They are much more irregular than the slips in development, in direction and in inclination. They generally occur, however, in two more or less distinct sets striking at right angles to one another. The strike of one of these sets frequently forms an angle of 350–45° with the strike of the slips. There is usually much powdery crushed coal in the ' backs,' and they frequently pass into, or are associated with, an irregular type of cone-in-cone structure that is very prevalent amongst the semi-anthracitic coals. These structures seem to be connected with the brittleness of the transitional coals which has been referred to already (p. 213), and they are most marked in the thicker coals such as the Nine-Feet and Six-Feet, where the ' backs ' may mask the slips almost completely.

Coal resources

A large part of this area has been extensively worked, but there are still considerable reserves. Lines between areas which have been largely worked and those in which development is only commencing are shown in the sketch-maps of the various seams (Figs. 23, 24, 27, and 28). In general it may be said that the thickest and best seams are approaching exhaustion in the Merthyr and Aberdare area ; and almost all the good seams have been freely worked for some distance in from the crop between the Vale of Neath and the western boundary of the district. Much of the ground between the Cynon and Neath valleys, however, is almost untouched except along the coal crops, and the greater part of the south-western margin of the district is virgin ground.

The depth of the seams in the unexploited ground along the southern edge of the district does not exceed 2,000 ft., as measured from the valley bottoms, except possibly in the extreme southwestern corner.

There is no likelihood of trouble from excessive water in future working, but the potential existence of belts of disturbance, washouts, etc., must not be ignored. The particulars given in the chapters on Structure and Coal Measures are of considerable value in the estimation of the position and character of possible disturbances.

In general the rank of all the coals increases rapidly westwards and it also increases significantly with depth at any one locality (George, 1970, fig.30). In the east of the district the seams above the Two-Feet-Nine are bituminous and those below are all steam coals. Between Aberdare and the Neath Valley the coals range from semi-bituminous above the Gorllwyn Seam to semi-anthracite in the lower seams. West of the Neath Valley all the coals are anthracitic, the lower seams having distinctly the higher ranks. The cause of the westwards devolatilisation remains unresolved. In addition to Hilt's depth of burial theory (e.g. Jones, 1951), and Trotter's (1948) conjecture that it was related to the Carreg Cennen Disturbance, recent work on the geochemistry of the coals has suggested that an emanative source, possibly a deeply buried granite, centred on the north-western margins of the coalfield, may have been reponsible (Davies and Bloxam in Owen, 1974).

Limestone

The Dinantian limestones have been extensively quarried for flux, lime and crushed aggregate, though nowadays the last-named is the main product. The Abercriban Oolite and the Penderyn Oolite are the purest rocks and were used as flux for steel-making. The Cwar yr Ystrad and Cwar yr Hendre quarries supplied stone for the Ebbw Vale steelworks, but they were abandoned when the furnaces ceased to operate.

The physical properties of the limestones were described by Harrison and others (1983). Most have adequate strengths for constructional purpose and are highly resistant to abrasion, but they are not suitable for road surfacing by virtue of their low resistance to polishing. Three quarries are working at present, mainly for aggregate: Penwyllt [SN 856 158] works Dowlais Limestone, Penderyn Oolite and Penwyllt Limestone, and has a rail link to supply flux to the Abbey Steelworks at Margam; Penderyn Quarry [SN 953 092] works Dowlais Limestone and Penderyn Oolite; and Vaynor Quarry [SO 038 147] works the Dowlais Limestone.

Ironstone

The nodules and bedded sideritic iron ores of the Lower and Middle Coal Measures provided the raw material on which the iron and steel industry of south Wales was founded, and this in turn gave impetus to the Industrial Revolution during the late eighteenth and early nineteenth centuries. Initially the ores were smelted using charcoal derived from the deciduous forests of the area, but Merthyr, and particularly Dowlais, may have been the birthplace of the modern iron industry when Anthony Bacon first used coal for smelting (Strahan and others, 1920). The ironstones are of little more than historical interest today and there is nothing to add to the accounts of Robertson (1933, pp.220–222), Rogers (1861) and Strahan and others (1920, p.101).

Silica rock and silica sand

The highly siliceous orthoquartzites of the Namurian Basal Grit have been worked for the manufacture of silica bricks for refractory purposes, mainly by quarrying, but also by mining from the Dinas levels at Glyn-neath. However the decline of the old traditional steel-making industry in south Wales has reduced demand for refractory bricks, and the only recently working quarry in the district, at Cefn Cadlan, Penderyn [SN 957 113] (Highley, 1977) has been abandoned. The quality of the rock is variable because of iron-staining and shale partings; also, contained quartz pebbles may be explosive during firing of the bricks.

Deposits of silica sand occur in collapse hollows in the Basal Grit, and Thomas (1961) considered that these might prove to be an important source of glass sand.

Brick clays and fireclays

Although in the past the mudstones of the Coal Measures have been used extensively for brick-making elsewhere in the coalfield, the only brickworks in the district at present is at Merthyr where a disused ironstone working shale-tip is being dug. Small amounts of soft clay seatearth (fireclay) have been exploited in the past for refractory bricks, and at some of the opencast sites in the district it has been extracted for use in ordinary brick-making elsewhere in South Wales.

Rottenstone

The Dinantian Upper Limestone Shales weather at outcrop to fine-grained rottenstone, which was formerly dug for use as a polishing powder (Robertson, 1933, pp.225–228). Thomas (1961) reported that the last working took place in the nineteen-thirties.

Sandstone

The sandstones of the Old Red Sandstone, Dinantian, Namurian and the Coal Measures (including in particular the Pennant), have been quarried extensively in the past for building and walling purposes. However, there are now no quarries in the sandstones for these purposes. There is nothing to add to Robertson's (1933, pp.224, 225) account, when the Old Red Sandstone Grey Grits were being quarried at Abercriban, and Adlam and others (1984) study of hard rock resources in the Caerphilly district.

Water supply

Groundwater is of limited importance. There are no abstractions recorded from the Devonian, but water levels are generally close to the surface and yields of up to 0.5 l/s are obtained from boreholes and springs in adjacent areas. Carboniferous rocks form the main aquifer in the district. The Dinantian limestones have minimal primary permeability, and groundwater movement is concentrated along joints and fissures enlarged by solution. These fissures are generally large and groundwater flow can be rapid, issuing from a limited number of springs whose flow rates vary with rainfall. At Dan-yr-Ogof [SN 8380 1600] flow rates in excess of 1000 l/s have been recorded. Small springs also issue at the top of the Lower Limestone Shale. Drilling into the limestones is speculative as the fissure systems tend to be widely distributed and not necessarily extensively interconnected. A borehole at Penderyn [SN 9522 0775] into the limestone beneath superficial deposits and Namurian strata, yielded 64 l/s when constructed. Water from the limestone is hard, but of good quality under low flow conditions; however it may be turbid and polluted after heavy rain. Springs occur at the bases of the Upper Carboniferous sandstones; several in the Namurian at Garw-leisiau [SN 8269 1331] yield up to 100 l/s. The Gelli Isaf borehole [SN 9912 0427] yielded overflowing artesian water at 0.6 l/s from a basal Westphalian sandstone. A borehole at Merthyr Tydfil [SO 0471 0639] produced 0.7 l/s from the Westphalian. Mining subsidence can affect groundwater movement in the Pennant Measures. Concentration of flow in zones of tensional stress may lead to the enlargement of groundwater catchments, and to an increase in the groundwater component of streams draining such areas. The water is generally of reasonable quality but is locally polluted by coal mining. Water from the superficial deposits is generally in hydraulic continuity with the adjacent river. Five boreholes in river gravels at Rheola [SN 8407 0370] have together yielded 62 l/s.

References

ADAMS, H. F. 1967. The seams of the South Wales Coalfield. Inst. of Min. Eng., London. 38pp.

ADLAM, K. A. M., HARRISON, D. J. and WILD, J. B. L. 1984. The hard rock resources of the country around Caerphilly, S. Wales. Miner. Assess. Rep Inst. Geol. Sci., No. 141.

ALLEN, J. R. L. 1964. Pre-Pickwell Down age of the Plateau Beds (Upper Devonian) in South Wales. Nature, London, Vol. 204, 364–366.

ALLEN, J. R. L. 1965. Upper Old Red Sandstone (Farlovian) palaeogeography in South Wales and the Welsh Borderland. J. Sed. Petrol., Vol. 35, 167–195.

ALLEN, J. R. L. 1977. Wales and the Welsh Borders. In A correlation of Devonian rocks of the British Isles. HOUSE and OTHERS. Special Report Geol. Soc. London, No. 7. 110pp.

ANDERSON, J. G. C. and OWEN, T. R. 1979. The late Quaternary history of the Neath and Afan valleys, South Wales. Proc. Geol. Assoc., Vol. 90, 203–211.

ARCHER, A. A. 1968. Geology of the South Wales Coalfield, Gwendraeth valley and adjoining areas. Spec. Mem. Geol. Surv. G.B.

BARCLAY, W. J. In press. The geology of the South Wales Coalfield, Part II, Abergavenny (3rd edition). Mem. Br. Geol. Surv.

BARCLAY, W. J.and TAYLOR, K. 1976. The Gelli Isaf borehole. 26 in IGS boreholes 1975. INSTITUTE OF GEOLOGICAL SCIENCES. Rep. Inst. Geol. Sci., No. 76/10. 47pp.

BISAT, W. S. 1924. The Carboniferous goniatites of the north of England and their zones. Proc. Yorkshire Geol. Soc., Vol. 20, 40 –124.

BLOXAM, T. W. and THOMAS, R. L. 1969. Palaeontological and geochemical facies in the Gastrioceras subcrenatum Marine Band and associated rocks from the North Crop of the South Wales coalfield. Q. J. Geol. Soc. London, Vol. 24, 239–281.

BLUNDELL, C. R. K. and MOORE, L. R. 1960. Mid-coal measures 'red beds' in the South Wales Coalfield. C. R. 4me Congr. Strat. Carb. Heerlen 1958, Vol. 1, 41–48.

BURCHETTE, T. P. 1981. The Lower Limestone Shales. In A field guide to the Lower Carboniferous rocks near Abergavenny. WRIGHT, V. P., RAVEN, M. and BURCHETTE, T. P. 76pp. (Cardiff: Unversity College of Wales.)

CONWAY, B. W., FORSTER, A., NORTHMORE, K. J. and BARCLAY, W. J. 1980. South Wales Coalfield Landslip Survey. Rep. Eng. Geol. Dep. Inst. Geol. Sci. No. EG 80/4 [Unpublished].

CROFT, W. N. 1953. Breconian: a stage name of the Old Red Sandstone. Geol. Mag. Vol. 90, 429–432.

CROFT, W. N. and LANG, W. H. 1942. The Lower Devonian flora of the Senni Beds of Monmouthshire and Breconshire. Philos. Trans. R. Soc. B231, 131–163.

DAVIES, J. H. and TRUEMAN, A. E. 1927. A revision of the non-marine lamellibranches of the Coal Measures and a discussion of their zonal sequence. Q. J. Geol. Soc. London, Vol. 83, 210–259.

DIX, E. 1933. The succession of plants in the Millstone Grit and the lower portion of the Coal Measures of the South Wales Coalfield (near Swansea) and a comparison with that of other areas. Palaeontographica, No. 78B, 158–202.

DIX, E. 1934. The sequence of floras in the Upper Carboniferous with special reference to South Wales. Trans. R. Soc. Edinburgh, Vol. 57, 789–821.

DIX, E. and TRUEMAN, A. E. 1928. Marine horizons in the Coal Measures of South Wales. Geol. Mag., Vol. 65, 356–363.

DOWNING, R. A. and SQUIRRELL, H. C. 1965. On the red and green beds in the Upper Coal Measures of the eastern part of the South Wales Coalfield. Bull. Geol. Surv. G.B., No. 23, 45–56.

EVANS, D. G. and JONES, R. 0. 1929. Notes on the Millstone Grit of the north crop of the South Wales Coalfield. Geol. Mag.. Vol. 68, 164–177.

FRIEND, P. F. and WILLIAMS, B. P. J. (editors). 1978. International symposium on the Devonian system (P.A.D.S. 1978) September 1978. A field guide to selected outcrop areas of the Devonian of Scotland, the Welsh Borderland and South Wales. 106pp. (The Palaeontological Association.)

GEORGE, T. N. 1927. The Carboniferous Limestone (Avonian) succession of a portion of the north crop of the South Wales Coalfield. Q. J. Geol. Soc. London, Vol. 83, 38–95.

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

GEORGE, T. N. 1970. British regional geology: South Wales (3rd edition). (London: HMSO for Institute of Geological Sciences.)

GEORGE, T. N. JOHNSON, G. A. L., MITCHELL, M., PRENTICE, J. E., RAMSBOTTOM, W. H. C., SEVASTOPULO, G. D. and WILSON, R. B. 1976. A correlation of Dinantian rocks in the British Isles. Spec. Rep. Geol. Soc. London, No. 7. 87pp.

HALL, I. H. S., TAYLOR, K. and THOMAS, L. P. 1973. The stratigraphy of the Upper Old Red Sandstone in South Breconshire. Bull. Geol. Surv. G.B., No. 44, 45–62.

HARRISON, D. J., MURRAY. D. W. and WILD, J. B. L. 1983. Reconnaissance Survey of Carboniferous Limestone Resources in Wales. Intern. Rep. Inst. Geol. Sci. 79pp. [Unpublished].

HEARD, A. 1927. On Old Red Sandstone plants showing structure from Brecon, South Wales. Q. J. Geol. Soc. London, Vol. 83, 195–207.

HIGHLEY, D. E. (compiler) 1977. Silica. Miner. Dossier Miner. Resour. Consult. Comm. No. 18. 57pp.

JONES, D. G. 1958. A note on new Namurian plant localities at the head of the Neath Valley, South Wales. Geol. Mag., Vol.95, 77–81.

JONES, D. G. 1970. The base of the Namurian in South Wales. C. R. 6th Cong. Intern. Strat. Geol. Carb. Sheffield 1967, Vol. 3, 1023–1029.

JONES, D. G. and OWEN, T. R. 1956. The rock succession and geological structure of the Pyrddin, Sychrid and Upper Cynon Valleys, South Wales. Proc. Geol. Assoc., Vol. 72, 239–250.

JONES, O. T. 1951. The distribution of coal volatiles in the South Wales Coalfield and its probable significance. Q. J. Geol. Soc. London, Vol. 107, 57–83.

LEITCH, D., OWEN, T. R. and JONES, D. G. 1958. The Basal Coal Measures of the South Wales Coalfield from Llandebie to Brynmawr. Q. J. Geol. Soc. London, Vol. 113, 461–486.

LOEFFLER, E. G., and THOMAS, R. G. 1980. A new pteraspidid ostracoderm from the Devonian Senni Beds formation of South Wales and its stratigraphic significance. Palaeontology, Vol. 23, 287–296.

LOVELL, R. W. W. 1978. The sedimentology and stratigraphy of the Upper Old Red Sandstone and Lower Limestone Shales of the South Wales Coalfield. Unpublished Ph.D. thesis, University of Bristol.

OWEN, T. R. 1954. The structure of the Neath disturbance between Bryniau Gleision and Glynneath, South Wales. Q. J. Geol. Soc. London, Vol. 109, 333–365.

OWEN, T. R. 1964. The tectonic framework of Carboniferous sedimentation in South Wales. 301–307 in Deltaic and shallow marine deposits. VAN STRAATAN, L. M. J. U. (editor). (Amsterdam: Elsevier.)

OWEN, T. R. (editor). 1974. The Upper Palaeozoic and Post Palaeozoic rocks of Wales. 425pp. (Cardiff: University of Wales Press.)

OWEN, T. R. and JONES, D. G. 1961. The nature of the Millstone Grit-Carboniferous Limestone junction of a part of the North Crop of the South Wales Coalfield. Proc. Geol. Assoc., Vol. 72, 239–249.

OWEN, T. R. and WEAVER, J. D. 1983. The structure of the main South Wales Coalfield and its margins. In The Variscan Fold Belt in the British Isles. HANCOCK, P. L. (editor). 217pp. (Bristol: Hilger.)

OWEN, T. R. RHODES, F. H. T., JONES, D. G. and KELLING, G. 1965. Summer (1964) field meeting in South Wales. Proc. Geol. Assoc., Vol. 76, 463–496.

RAMSBOTTOM, W. H. C. 1973. Transgressions and regressions in the Dinantian; a new synthesis of British Dinantian stratigraphy. Proc. Yorkshire Geol. Soc., Vol. 39, 567–607.

RAMSBOTTOM, W. H. C. 1978. Namurian mesothems in South Wales and northern France. J. Geol. Soc. London, Vol. 135, 307–312.

RAMSBOTTOM, W. H. C. CALVER, M. A., EAGER, R. M. C., HODSON, F., HOLLIDAY, D. W., STUBBLEFIELD, C. J. and WILSON, R. B. 1978. A correlation of Silesian rocks in the British Isles. Spec. Rep. Geol. Soc. London, No. 10, 81pp.

ROBERTSON, T. 1933. The geology of the South Wales Coalfield, Part V, Merthyr Tydfil (2nd edition). Mem. Geol. Sum. 283pp.

ROBERTSON, T. and GEORGE, T. N. 1929. The Carboniferous Limestone of the North Crop of the South Wales Coalfield. Proc. Geol. Assoc., Vol. 49, 18–40.

ROGERS, E. 1861. Iron ores of Great Britain, Part III. Iron ores of South Wales. Mem. Geol. Surv. G.B. 236pp.

STRAHAN, A. H., GIBBON, W. and CANTRILL, T. C. 1904. The geology of the South Wales Coalfield, Part V, Merthyr Tydfil (1st edition). Mem. Geol. Surv. 132pp.

STRAHAN, A. H., GIBSON, W., CANTRILL, T. C., SHERLOCK, R. L. and DEWEY, H. 1920. Iron ores. Pre-Carboniferous and Carboniferous bedded ores of England and Wales. Mem. Geol. Surv. Spec. Rep. Miner. Resour. G.B., Vol. XIII.

TAYLOR, K. 1972. New fossiliferous localities in the Upper Old Red Sandstone of the Ystradfellte-Cwm Taff district of Breconshire. Bull. Geol. Surv. G. B. No. 3, 11–14.

TAYLOR, K. and THOMAS, L. P. 1974. Field meeting in the Upper Old Red Sandstone of south Breconshire. Proc. Geol. Assoc., Vol. 85, 423–432.

TAYLOR, K. 1975. Geological Survey boreholes in south Breconshire and their bearing on the stratigraphy of the Upper Old Red Sandstone and the Carboniferous-Old Red Sandstone boundary east of the Afon Hepste. Bull. Geol. Surv. G.B. No. 54, 1–39.

THOMAS, G. 1951. The stratigraphy of the Lower Limestone Shales and the Devonian-Carboniferous junction in South Wales: Unpublished PhD thesis, Imperial College, London.

THOMAS, T. M. 1961. The mineral wealth of Wales and its exploitation. 248pp. (Edinburgh and London: Oliver and Boyd.)

TROTTER, F. M. 1947. The structure of the Coal Measures in the Pontardawe-Ammanford area, South Wales. Q. J. Geol. Soc. London, Vol. 103, 89–113.

TROTTER, F. M. 1948. The devolatilisation of coal seams in South Wales. Q. J. Geol. Soc. London, Vol. 104, 387–437.

TUNBRIDGE, I. P. 1981. Old Red Sandstone sedimentation-an example from the Brownstones (highest Lower Old Red Sandstone) of south central Wales. Geol. J. Vol. 16, 111–124.

VAUGHAN, A. 1905. The palaeontological sequence in the Carboniferous Limestone of the Bristol area. Q. J. Geol. Soc. London, Vol. 61, 181–307.

WARE, W. D. 1930. An account of the geology of the Cefn Coed sinkings. Proc. S. Wales Inst. Eng., Vol. 46, 453–494.

WARE, W. D. 1939. The Millstone Grit of Carmarthenshire. Proc. Geol. Assoc., Vol. 59, 168–204.

WEAVER, J. D. 1975. The structure of the Swansea Valley disturbance between Clydach and Hay-on-Wye, South Wales. Geol. J., Vol. 10, 75–86.

WHITE, E. I. 1938. New pteraspids from South Wales. Q. J. Geol. Soc. London, Vol. 94, 85–115.

WOODLAND, A. W. and EVANS, W. B. 1964. The geology of the South Wales Coalfield, Part IV, Pontypridd and Maesteg. Mem. Geol. Sun,. 391pp.

WRIGHT, V. P. 1981. The Llanelly Formation. In A field guide to the Lower Carboniferous rocks near Abergavenny. WRIGHT, V. P., RAVEN, M. and BURCHETTE, T. P. 76pp. (Cardiff: University College of Wales.)

WRIGHT, V. P. 1982. Calcrete palaeosols from the Lower Carboniferous Llanelly Formation, South Wales. Sediment. Geol. Vol. 33, 1–33.

Appendix 1 Six-inch or 1:10 000 maps

The National Grid six-inch or 1:10 000 maps included wholly or in part in the Merthyr Tydfil sheet are listed below, together with the initials of the surveyors and the dates of survey. The officers involved were A. A. Archer, R. A. Downing, W. B. Evans, G. W. Green, I. H. S. Hall, H. C. Squirrell, J. V. Stephens, K. Taylor, L. P. Thomas and A. W. Woodland. Copies of the maps are available for public reference in the libraries of the British Geological Survey. Most can be purchased, either as published maps or as dye-line copies.

SN 70 NE Ystalyfera GWG 1959–60, 1964
SN 70 SE Cilfrew AAA, JVS, WBE, GWG 1950, 1959
SN 71 NE Esgair Hir GWG, HCS 1964, 1966
SN 71 SE Ystradgynlais GWG 1960, 1964
SN 72 SE Llandeusant GWG, HCS 1964, 1966
SN 80 NW Seven Sisters GWG, KT 1961–2, 1968–9
SN 80 SW Resolven WBE, KT 1947–50, 1967–8
SN 80 NE Glyn-neath KT 1966
SN 80 SE Blaengwrach WBE, KT 1947–50, 1967–8
SN 81 NW Fan Hir KT 1971–2
SN 81 SW Abercrave GWG 1966
SN 81 NE Fan Gihirych KT 1971- 2
SN 81 SE Coelbren KT 1966
SN 82 SW Traianglas KT 1972
SN 82 SE Cray KT 1972
SN 90 NW Rhigos KT 1968
SN 90 SW Blaenrhondda WBE, KT 1946–8, 1968
SN 90 NE Hirwaun LPT 1970
SN 90 SE Aberdare LPT 1970
SN 91 NW Fan Nedd IHSH 1971–2
SN 91 SW Ystradfellte KT 1970
SN 91 NE Fan Fawr IHSH 1971–2
SN 91 SE Cader Fawr KT 1971
SN 92 SW Senny IHSH 1972
SN 92 SE Glyn IHSH 1972
SO 00 NW Merthyr Tydfil HCS 1968–9
SO 00 SW Aberaman AWW, HCS 1946, 1968–9
SO 00 NE Dowlais HCS 1967–8
SO 00 SE Pentrebach-Aberfan AWW, RAD, HCS 1945,1952, 1967–8
SO 01 NW Twyn Mwyalchod LPT 1972
SO 01 SW Vaynor KT 1971
SO 01 NE Glyn-collwn LPT 1971–2
SO 01 SE Cefn yr Ystrad LPT 1971
SO 02 SW Modrydd LPT 1972
SO 02 SE Llanfigan LPT 1972

Appendix 2 Selected boreholes and shafts

SN70NE/34 Dulais Valley No. 2 B.H. [SN 7967 0527] 694.94m. Upper, Middle and Lower Coal Measures to below Garw.
SN70NE/48 Treforgan East Development No. 2 B.H. (Oaktree) [SN 7948 0789] 360.7 m. Middle Coal Measures to Amman Rider.
SN70NE/49 Treforgan East Development No. 3 B.H. (Tynwern) [SN 7990 0696] 395.89 m. Basal Upper Coal Measures and Middle Coal Measures to Amman Rider.
SN80NW/30 Dulais Valley No. 1 B.H. [SN 8024 0663] 523.32 m. Basal Upper Coal Measures, Middle and Lower Coal Measures to below ? Gellideg.
SN80NW/41 Treforgan East Development No. 1 B.H. (Dillwyn) [SN 8125 0742] 345.65 m. Middle and topmost Lower Coal Measures.
SN80NW/42 Treforgan East Development No. 4 B.H. (Hir Fynydd) [SN 8180 0672] 479.8 m. Middle Coal Measures to topmost Lower Coal Measures.
SN80NW/44 Seven Sisters Pit [SN 8214 0914] 218.1 m. Middle Coal Measures to Bute.
SN80NW/46 Treforgan East Development No. 5 B.H. (Gelli-Benuchel) [SN 8074 0550] 778.5 m. Middle Coal Measures and Lower Coal Measures to below Garw.
SN80SW/120 Glyncastle No. 2 Pit [SN 8462 0203] 610.26 m. Rhondda and Llynfi Beds, Middle and Lower Coal Measures to below Garw.
SN90SW/47 Llyn Fach B.H. [SN 9022 0399] 160.93 m. Llynfi Beds and Middle Coal Measures to below Upper Cwmgorse Marine Band.
SN90SE/54 Gelli Isaf B.H. [SN 9912 0427] 182.22 m. Lower Coal Measures above Garw to top of Namurian Basal Grit (Barclay and Taylor, 1976).
SN90SE/56 Powell's Pit (Bwllfa No. 3) [SN 9860 0251] 209.3 m. Middle and Lower Coal Measures to Gellideg.
SN90SE/62 Graig/Dare Pits [SN 9952 0208] 287.4 m. Basal Llynfi Beds. Middle and Lower Coal Measures to below Garw.
SN90SE/107 Bwllfa collieries: composite section through Middle and Lower Coal Measures to Garw.
SN91SE/1 Cefn Esgair Carnau North B.H. (231/B) [SN 9835 1419] 94.95 m. Lower Limestone Shale to topmost Brownstones (Taylor and Thomas, 1975).
SN91SE/2 Cefn Esgair Carnau South B.H. (231/C) [SN 9845 1353] 114 m. Lower Limestone Shale to topmost Brownstones (Taylor and Thomas, 1975).
SO00SW/49 Abernantygroes Pit (air shaft) [SO 0267 0218] 150.52 m. Middle Coal Measures to Nine-Feet seam.
SO00SW/61 Werfa Pit, Aberdare [SO 0198 0316] 155.32 m. Middle and Lower Coal Measures to Five-Feet seam.
SO00SW/63 George Pit [SO 0428 0175] 449 m. Upper, Middle and Lower Coal Measures to Seven-Feet seam.
SO00SW/64 Blaennant Pit [SO 0212 0451] 224.76 m. Middle and Lower Coal Measures to Gellideg.
SO00SW/65 River Level Pit, Aberdare [SO 0088 0344] 98.55 m. Middle and Lower Coal Measures to Gellideg.
SO00SW/68 Blaengwawr Pit, Aberdare [SO 0061 0200] 147.32 m. Middle and Lower Coal Measures to Gellideg.
SO00SW/69 Tunnel Pit, Aberdare [SO 0255 0263] 149.38 m. Middle Coal Measures to Nine-Feet.
SO00SW/70 High Dyffryn Pit, Cwmbach [SO 0182 0184] 122.12 m. Middle and Lower Coal Measures to Seven-Feet seam.
SO00SW/71 Park Pit, Aberdare [SO 0038 0354] 121.79 m. Middle and Lower Coal Measures to below Garw.
SO00SW/72 Ysguborwen Pit, Aberdare [SO 009 045] 177.52 m. Middle and Lower Coal Measures to Gellideg.
SO00NE/57 Penydarren No. 1 Pit, Merthyr Tydfil [SO 0798 0523] 225 m. Upper and Middle Coal Measures to Six-Feet seam.
SO00NE/65 Isaac Morgan Pit, Penydarren Colliery [SO 0812 0625] 205.7 m. Middle and Lower Coal Measures to Five-Feet seam.
SO00SE/165 South Dyffryn No. 3 Pit, Merthyr Tydfil [SO 0703 0314] 239.75 m. Middle and Lower Coal Measures to Five-Feet.
SO00SE/166 Castle Pit, Merthyr Tydfil [SO 0647 0262] 392 m. Upper, Middle and Lower Coal Measures to below Gellideg.
SO00SE/167 Abercanaid Pit, Abercanaid [SO 0542 0408] 127 m. Middle and Lower Coal Measures to Gellideg.

Figures and plates

Figures

(Figure 1) General geology of the Merthyr Tydfil district

(Figure 2) Schematic cross-section showing relationships of the formations of the Upper Devonian to the older and younger beds

(Figure 3) Schematic cross-section of the Dinantian rocks described in some detail the strata between the Lower Limestone Shale and the Dowlais Limestone.

(Figure 4) Generalised vertical sections of the Namurian rocks and their classification. Chronozones and mesothems are from Ramsbottom (1978) and Ramsbottom and others (1978). Inset (after Ramsbottom, 1978) shows the full basinal Namurian sequence, with its attenuated equivalent in the Merthyr Tydfil district, as interpreted by Ramsbottom (1978). Stipple represents deposition, blank areas represent non-deposition

(Figure 5) Generalised vertical section of the Westphalian rocks

(Figure 6) Generalised vertical section of the Lower Coal Measures (Westphalian A)

(Figure 7) Location of boreholes, shafts and opencast sites

(Figure 8) Generalised vertical section of the Middle Coal Measures. See (Figure 6) for key

(Figure 9) Generalised vertical section of the Upper Coal Measures. See (Figure 6) for key

(Figure 10) The main structural elements of the Merthyr Tydfil district

(Figure 11) Sections across the Neath Disturbance (from Robertson, 1933, fig. 45). See (Figure 10) for their locations

(Figure 12) Sections across the Cribarth Disturbance. See (Figure 10) for their locations

(Figure 47) Charts showing variations in the volatile content of the main seams in the lower portion of the Lower Coal Series.

(Figure 48) Chart showing variations in the average volatile content of the  coals in the lower and upper portions of the Lower Coal Series.

Plates

(Plate 1) Gently dipping Namurian sandstones and Dinantian limestones, west of Craig-y-nos Castle in the Upper Tawe Valley. A11998

(Plate 2) Topmost Grey Grits and basal Lower Limestone Shale, Abercriban Quarry. A11993

(Plate 3) Carious weathering of the Honeycombed Sandstone in the dry bed of the River Neath [SN 9114 1376]. A11978

(Plate 4) Llech Sychryd Waterfall formed by the Twelve Foot Sandstone [SN 9913 0733]. A11991

(Plate 5) The 'Farewell Rock', Pont Nedd Fechan [SN 9010 0767]. A11957

(Plate 6) Steeply dipping Asbian limestones on the northern flank of the Neath Disturbance, Craig y Dinas [SN 9118 0793]. A11965

(Plate 7) View looking south-west along the axis of the Cribarth Anticline Craig-yr-Rhiwarth, Cribarth [SN 845 158]. A12000

(Plate 8) Glacial till overlying striated bedrock, Nant Plas y Gors [SN 9287 1543]. A12006

(Plate 9) Craig y Llyn escarpment formed by Pennant sandstones, with Llyn Fawr, a late glacial cirque [SN 9175 0346]. A11972

(Plate 10) Dunraven opencast coal site in 1969, exploiting coals of the Middle Coal Measures [SN 898 049]. A11974

Tables

(Table 1) Present and previous classifications of the Dinantian rocks. George and others (1976) placed the top part of the Abercriban Oolite and the Llanelly Formation in the Chadian.