Palaeogene—Neogene of Wales: Difference between revisions

From MediaWiki
Jump to navigation Jump to search
m (1 revision imported)
m (1 revision imported)
 
(3 intermediate revisions by 2 users not shown)
Line 1: Line 1:
'''From: Howells, M F. 2007. [[British regional geology: Wales|British regional geology: Wales]]. Keyworth, Nottingham: British Geological Survey.'''
'''From: Howells, M F. 2007. [[British regional geology: Wales|British regional geology: Wales]]. Keyworth, Nottingham: British Geological Survey.'''
[[File:P916208.jpg|thumbnail|Palaeogene strata in offshore basins (after Tappin et al., 1994). P916208.]]
[[File:P916208.jpg|thumbnail|Palaeogene strata in offshore basins (after Tappin et al., 1994). P916208.]]
[[File:P916209.jpg|thumbnail|Depositional model for Cardigan Bay, early Palaeogene (after Tappin et al., 1994). P916209.]]
[[File:P916209.jpg|thumbnail|Depositional model for Cardigan Bay, early Palaeogene (after Tappin et al., 1994). P916209.]]
[[File:P662436.jpg|thumbnail|Palaeokarstic surface at the top of the Dinantian Dowlais Limestone overlain unconformably by pebbly grit at the base of the Namurian; the grit also infills large solution cavities in the limestones, Trefil Quarry (MFH P662436).]]
[[File:P662436.jpg|thumbnail|Palaeokarstic surface at the top of the Dinantian Dowlais Limestone overlain unconformably by pebbly grit at the base of the Namurian; the grit also infills large solution cavities in the limestones, Trefil Quarry (MFH P662436).]]
[[File:P662437.jpg|thumbnail|Platform across truncated Dinantian limestones near St Govan’s Head, south Pembrokeshire (MFH P662437).]]
[[File:P662437.jpg|thumbnail|Platform across truncated Dinantian limestones near St Govan’s Head, south Pembrokeshire (MFH P662437).]]
Sea level rose progressively throughout most of Cretaceous time, although in the final stages, Campanian to Maastrichtian, there was a reversal of this trend and most of the basins close to the Welsh coast were inverted and, by early Palaeogene times, were subaerially exposed. The uplift of the basins led to erosion and, for the most part, succeeding deposition was mainly terrestrial. During this period, the climate was tropical or subtropical and it was only later, in Neogene times, that there was a gradual reduction in temperature. Palaeogene sediments have been determined in the St George’s Channel, South Celtic Sea and North Celtic Sea basins. Smaller outliers are preserved in half-grabens in the Cardigan Bay and Bristol Channel basins. The only Neogene sediments recorded are the Lower Miocene sediments in the Mochras Borehole.


The morphology of the main Cainozoic basins was controlled by pre-existing structures; the north-easterly trend of the St George’s Channel Basin is Caledonian, and the more easterly trend of the South Celtic Sea and the Bristol Channel basins is Variscan. In the St George’s Channel Basin, the thickest Cainozoic accumulations coincide with those of the Mesozoic, whereas in the east–west-trending basins they are offset laterally. North of the Variscan Front, the absence of the upper part of the Cretaceous sequence in the St George’s Channel and Cardigan Bay basins may be due to non-deposition or, more probably, uplift and erosion, which would also account for the unconformity between the Jurassic and Cainozoic sequences. However, in St George’s Channel Basin, the unconformity has been ascribed to basin inversion with movement along St George’s and associated faults. To the south of the Variscan Front, the South Celtic Sea Basin was also inverted, although uplift was generally more regional.
Sea level rose progressively throughout most of Cretaceous time, although in the final stages, Campanian to Maastrichtian, there was a reversal of this trend and most of the basins close to the Welsh coast were inverted and, by early Palaeogene times, were subaerially exposed. The uplift of the basins led to erosion and, for the most part, succeeding deposition was mainly terrestrial. During this period, the climate was tropical or subtropical and it was only later, in Neogene times, that there was a gradual reduction in temperature. Palaeogene sediments have been determined in the St George’s Channel, South Celtic Sea and North Celtic Sea basins. Smaller outliers are preserved in half-grabens in the Cardigan Bay and Bristol Channel basins. The only Neogene sediments recorded are the Lower Miocene sediments in the Mochras Borehole.


The sedimentary character of the Cainozoic sequence is best known from the Mochras Borehole, sited close to the faulted margin of the Cardigan Bay Basin, where 524 m of Middle Oligocene to Lower Miocene strata were proved. The eastward-dipping sequence is contained in a half-graben [[Media:P916208.jpg|(P916208)]], and rests unconformably on Lower Jurassic strata. The sequence has been divided into three lithostratigraphical units. The Basal Red Unit comprises coarse-grained, cobble conglomerates, up to 14 m thick, with interbedded fining-upward sand to clay sequences, 1 to 2 m thick. The conglomerates are intepreted as alluvial debris is marked by a gradual change in colour from red to grey, and is characterised by upward-fining cycles of silt and clay with a carbonaceous and lignite component. The facies associations are characterisitic of a floodplain environment. The contact with the overlying, uppermost Lignite and Clay Unit is marked by a gradual increase in the proportion of lignite. The olive-green sediments form sharp-based, upward-fining sand to clay sequences laid down in brackish water or on well drained, vegetated surfaces. In many places, internal structures are obliterated by bioturbation, organic debris is scattered throughout the upper parts of the cycles and spherulitic siderite is common. The sequence suggests that swamp conditions, with possible seasonal variations, were established at the height of the water table.
The morphology of the main Cainozoic basins was controlled by pre-existing structures; the north-easterly trend of the St George’s Channel Basin is Caledonian, and the more easterly trend of the South Celtic Sea and the Bristol Channel basins is Variscan. In the St George’s Channel Basin, the thickest Cainozoic accumulations coincide with those of the Mesozoic, whereas in the east–west-trending basins they are offset laterally. North of the Variscan Front, the absence of the upper part of the Cretaceous sequence in the St George’s Channel and Cardigan Bay basins may be due to non-deposition or, more probably, uplift and erosion, which would also account for the unconformity between the Jurassic and Cainozoic sequences. However, in St George’s Channel Basin, the unconformity has been ascribed to basin inversion with movement along St George’s and associated faults. To the south of the Variscan Front, the South Celtic Sea Basin was also inverted, although uplift was generally more regional.


Lithologically similar sequences were intersected by a borehole at Tonfannau, sited on marine alluvium between the Mawddach and Dyfi estuaries, and in two offshore boreholes close to the northern edge of Cardigan Bay. Farther offshore, in St George’s Channel Basin [[Media:P916208.jpg|(P916208)]], the thickest Cainozoic sequence is intersected by the St George’s Fault along part of which a salt wall has been emplaced. North-west of the fault, in a faulted syncline, over 1500 m of Palaeogene (Eocene to Oligocene) strata lie unconformably on Cretaceous strata, and four boreholes have shown that the overall characteristics of the sediments are closely comparable with the Mochras sequence. A similar environment is envisaged — a floodplain with fluvial channel sands and overbank deposits. Between St George’s Channel Basin and the Welsh mainland, a small outlier of Eocene to mid-Oligocene sediments has been preserved in the Teifi basin, where 10 m of dark green and pale brown laminated clay, with plant fragments and glauconite, indicate possibly the most northerly marine incursion about Wales. The general disposition of Palaeogene facies across Cardigan Bay reflects a widely meandering river encroaching a swamp-dominated flood plain [[Media:P916209.jpg|(P916209)]].
The sedimentary character of the Cainozoic sequence is best known from the Mochras Borehole, sited close to the faulted margin of the Cardigan Bay Basin, where 524 m of Middle Oligocene to Lower Miocene strata were proved. The eastward-dipping sequence is contained in a half-graben [[Media:P916208.jpg|(P916208)]], and rests unconformably on Lower Jurassic strata. The sequence has been divided into three lithostratigraphical units. The Basal Red Unit comprises coarse-grained, cobble conglomerates, up to 14 m thick, with interbedded fining-upward sand to clay sequences, 1 to 2 m thick. The conglomerates are intepreted as alluvial debris is marked by a gradual change in colour from red to grey, and is characterised by upward-fining cycles of silt and clay with a carbonaceous and lignite component. The facies associations are characterisitic of a floodplain environment. The contact with the overlying, uppermost Lignite and Clay Unit is marked by a gradual increase in the proportion of lignite. The olive-green sediments form sharp-based, upward-fining sand to clay sequences laid down in brackish water or on well drained, vegetated surfaces. In many places, internal structures are obliterated by bioturbation, organic debris is scattered throughout the upper parts of the cycles and spherulitic siderite is common. The sequence suggests that swamp conditions, with possible seasonal variations, were established at the height of the water table.


In the South Celtic Sea Basin, over 500 m of Palaeogene sediments have been dated as mid-Eocene to Oligocene on the basis of a sparse microfloral assemblage. One borehole penetrated finely laminated, green, micaceous, glauconitic and lignitic sandstone with interbeds of silty clay and dark brown to black lignite. Lignite is more abundant near the base of the sequence and is thought to reflect deposition in a low-energy marginal marine environment. The upward increase in sandy glauconitic sediments suggests an increasing marine influence, although the persistent lignite content suggests close proximity to a vegetated landmass.
Lithologically similar sequences were intersected by a borehole at Tonfannau, sited on marine alluvium between the Mawddach and Dyfi estuaries, and in two offshore boreholes close to the northern edge of Cardigan Bay. Farther offshore, in St George’s Channel Basin [[Media:P916208.jpg|(P916208)]], the thickest Cainozoic sequence is intersected by the St George’s Fault along part of which a salt wall has been emplaced. North-west of the fault, in a faulted syncline, over 1500 m of Palaeogene (Eocene to Oligocene) strata lie unconformably on Cretaceous strata, and four boreholes have shown that the overall characteristics of the sediments are closely comparable with the Mochras sequence. A similar environment is envisaged — a floodplain with fluvial channel sands and overbank deposits. Between St George’s Channel Basin and the Welsh mainland, a small outlier of Eocene to mid-Oligocene sediments has been preserved in the Teifi basin, where 10 m of dark green and pale brown laminated clay, with plant fragments and glauconite, indicate possibly the most northerly marine incursion about Wales. The general disposition of Palaeogene facies across Cardigan Bay reflects a widely meandering river encroaching a swamp-dominated flood plain [[Media:P916209.jpg|(P916209)]].


During Cainozoic times, the Bristol Channel Basin was no longer extant, but farther south, to the west of Lundy Island, a new basin, the Bristol Channel Marginal Basin, developed. This basin has a downfaulted northern edge, and 300 m of Cainozoic strata are preserved within it. Seismic records indicate that a lower sequence, possibly sandstone, is overlain by clay and lignite, which have been recovered from the sea floor. East of Lundy Island, 335 m of braided channel and flood plain deposits are preserved in the Stanley Bank Basin on the east side of an extension of the Sticklepath–Lustleigh Fault Zone. Lundy Island is the site of an early Eocene igneous complex — a coarse- to fine-grained granite is itself intruded by numerous south-east-trending dykes of olivine and quartz dolerite, trachyte and trachyandesite. Dolerite dykes of Eocene age have been distinguished in a swarm, offshore, north­east of Anglesey, and many have been determined across Anglesey, Llŷn and Arfon with some extending into Snowdonia and beyond. These Paleocene–early Eocene igneous rocks lie at the edge of a large igneous province (Thulean), which straddled the eastern edge of Greenland, Iceland and extended to the Hebrides and north-east Ireland. The dykes in north Wales are the fringe to a volcanic centre at Mourne in Northern Ireland, although the currently exposed dykes imply a significant cover at the time of their intrusion and, consequently, considerable erosion since Eocene times.
In the South Celtic Sea Basin, over 500 m of Palaeogene sediments have been dated as mid-Eocene to Oligocene on the basis of a sparse microfloral assemblage. One borehole penetrated finely laminated, green, micaceous, glauconitic and lignitic sandstone with interbeds of silty clay and dark brown to black lignite. Lignite is more abundant near the base of the sequence and is thought to reflect deposition in a low-energy marginal marine environment. The upward increase in sandy glauconitic sediments suggests an increasing marine influence, although the persistent lignite content suggests close proximity to a vegetated landmass.
 
During Cainozoic times, the Bristol Channel Basin was no longer extant, but farther south, to the west of Lundy Island, a new basin, the Bristol Channel Marginal Basin, developed. This basin has a downfaulted northern edge, and 300 m of Cainozoic strata are preserved within it. Seismic records indicate that a lower sequence, possibly sandstone, is overlain by clay and lignite, which have been recovered from the sea floor. East of Lundy Island, 335 m of braided channel and flood plain deposits are preserved in the Stanley Bank Basin on the east side of an extension of the Sticklepath–Lustleigh Fault Zone. Lundy Island is the site of an early Eocene igneous complex  a coarse- to fine-grained granite is itself intruded by numerous south-east-trending dykes of olivine and quartz dolerite, trachyte and trachyandesite. Dolerite dykes of Eocene age have been distinguished in a swarm, offshore, north­east of Anglesey, and many have been determined across Anglesey, Llŷn and Arfon with some extending into Snowdonia and beyond. These Paleocene–early Eocene igneous rocks lie at the edge of a large igneous province (Thulean), which straddled the eastern edge of Greenland, Iceland and extended to the Hebrides and north-east Ireland. The dykes in north Wales are the fringe to a volcanic centre at Mourne in Northern Ireland, although the currently exposed dykes imply a significant cover at the time of their intrusion and, consequently, considerable erosion since Eocene times.


The extensive offshore record of Palaeogene–Neogene sedimentation contrasts with the patchy evidence onshore. Evidence of Palaeogene sediments on mainland Wales is restricted to deeply weathered regoliths, such as the ‘pocket deposits’ on the Carboniferous limestone outcrops between Llandudno and Mold in north Wales. The deposits include both alluvial sediments, introduced by subsidence into solution cavities, stratified sand and clay (products of subtropical, subaerial weathering, with gibbsite and kaolinite) and interbedded lignite, introduced by underground streams. Similar mottled clays, some 15 m thick, occur within limestone cavities at Flimston, near Castlemartin, in south Pembrokeshire, and others are exposed in the quarry at Trefil north of Merthyr Tydfil [[Media:P662436.jpg|(P662436)]]. Elsewhere in south Wales, in situ weathered profiles in Namurian sandstone along the north crop and in Old Red Sandstone orthoquartzite in the Black Mountains have also been regarded as Cainozoic in age.
The extensive offshore record of Palaeogene–Neogene sedimentation contrasts with the patchy evidence onshore. Evidence of Palaeogene sediments on mainland Wales is restricted to deeply weathered regoliths, such as the ‘pocket deposits’ on the Carboniferous limestone outcrops between Llandudno and Mold in north Wales. The deposits include both alluvial sediments, introduced by subsidence into solution cavities, stratified sand and clay (products of subtropical, subaerial weathering, with gibbsite and kaolinite) and interbedded lignite, introduced by underground streams. Similar mottled clays, some 15 m thick, occur within limestone cavities at Flimston, near Castlemartin, in south Pembrokeshire, and others are exposed in the quarry at Trefil north of Merthyr Tydfil [[Media:P662436.jpg|(P662436)]]. Elsewhere in south Wales, in situ weathered profiles in Namurian sandstone along the north crop and in Old Red Sandstone orthoquartzite in the Black Mountains have also been regarded as Cainozoic in age.


The patchy onshore Palaeogene–Neogene record makes it difficult to determine the geological evolution of this period in Wales. The task is further complicated by the variety of processes — continental and regional tectonism with concomitant movement of magma, climate changes and sea level fluctuations — that were particularly active throughout the period. The most clearly defined morphological feature in Wales is the ancient Lower Palaeozoic Caledonian massif, which developed from uplift that culminated in Mid Devonian times. Later, uplift was reactivated through Carboniferous times, and then the massif was enlarged by accretion in the late stages of the Variscan orogeny. However, interpretation of subsequent Mesozoic events and the final emergence in Cainozoic times is inhibited by the extremely restricted outcrops. In early Mesozoic times, the Palaeozoic landmass was probably reduced to a relatively low peneplain on which Triassic and Lias Group sediments were probably deposited and subsequently eroded. Later, with the exception of a restricted area in north-west Wales, the area was submerged beneath late Cretaceous seas. However, no Cretaceous rocks have been found on land, in solution cavities in Carboniferous limestone or in the Cainozoic of the Mochras Borehole. It is possible that the massif was uplifted prior to the transgression and, consequently, the preglacial evolution of Wales was entirely controlled by Cainozoic tectonic events and sea-level changes.
The patchy onshore Palaeogene–Neogene record makes it difficult to determine the geological evolution of this period in Wales. The task is further complicated by the variety of processes  continental and regional tectonism with concomitant movement of magma, climate changes and sea level fluctuations  that were particularly active throughout the period. The most clearly defined morphological feature in Wales is the ancient Lower Palaeozoic Caledonian massif, which developed from uplift that culminated in Mid Devonian times. Later, uplift was reactivated through Carboniferous times, and then the massif was enlarged by accretion in the late stages of the Variscan orogeny. However, interpretation of subsequent Mesozoic events and the final emergence in Cainozoic times is inhibited by the extremely restricted outcrops. In early Mesozoic times, the Palaeozoic landmass was probably reduced to a relatively low peneplain on which Triassic and Lias Group sediments were probably deposited and subsequently eroded. Later, with the exception of a restricted area in north-west Wales, the area was submerged beneath late Cretaceous seas. However, no Cretaceous rocks have been found on land, in solution cavities in Carboniferous limestone or in the Cainozoic of the Mochras Borehole. It is possible that the massif was uplifted prior to the transgression and, consequently, the preglacial evolution of Wales was entirely controlled by Cainozoic tectonic events and sea-level changes.


The relationship of the Welsh massif to the Mesozoic — Cainozoic offshore basins has long been a controversial subject, but the debate was intensified during the sinking of the Mochras Borehole and its completion in 1972. The juxtaposition of the thick Mesozoic and Palaeogene sequences in the borehole with Cambrian rocks that are exposed on the flank of the Harlech Dome just 2.5 km to the east, emphasises the influence of the intervening contemporaneous, coast-aligned Llanbedr Fault on sedimentation. This fault has an estimated downthrow of not less than 3750 m on the postulated base of the Triassic, and not less than 1350 m on the base of the Palaeogene.
The relationship of the Welsh massif to the Mesozoic  Cainozoic offshore basins has long been a controversial subject, but the debate was intensified during the sinking of the Mochras Borehole and its completion in 1972. The juxtaposition of the thick Mesozoic and Palaeogene sequences in the borehole with Cambrian rocks that are exposed on the flank of the Harlech Dome just 2.5 km to the east, emphasises the influence of the intervening contemporaneous, coast-aligned Llanbedr Fault on sedimentation. This fault has an estimated downthrow of not less than 3750 m on the postulated base of the Triassic, and not less than 1350 m on the base of the Palaeogene.


The contrast in the preserved Mesozoic–Cainozoic sequences in the offshore basins to the north and south of the Variscan Front reflects a marked contrast in deformation. To the north, in St George’s Channel and Cardigan Bay basins, no Cretaceous strata are preserved and the Palaeogene sequence, of mainly Eocene to Oligocene terrestrial rocks, rests on Upper Jurassic rocks. To the south of the Variscan Front, in the Celtic Sea Basin, variable Lower Cretaceous and thick Chalk sequences overlie Jurassic strata, and in the main Bristol Channel Basin the youngest preserved sediments are usually of late Jurassic–early Cretaceous age, although terrestrial Oligocene sediments are preserved in faulted sub-basins. For much of the Mesozoic and Cainozoic, the Welsh massif was not a significant feature and there is evidence of Neogene uplift offshore.
The contrast in the preserved Mesozoic–Cainozoic sequences in the offshore basins to the north and south of the Variscan Front reflects a marked contrast in deformation. To the north, in St George’s Channel and Cardigan Bay basins, no Cretaceous strata are preserved and the Palaeogene sequence, of mainly Eocene to Oligocene terrestrial rocks, rests on Upper Jurassic rocks. To the south of the Variscan Front, in the Celtic Sea Basin, variable Lower Cretaceous and thick Chalk sequences overlie Jurassic strata, and in the main Bristol Channel Basin the youngest preserved sediments are usually of late Jurassic–early Cretaceous age, although terrestrial Oligocene sediments are preserved in faulted sub-basins. For much of the Mesozoic and Cainozoic, the Welsh massif was not a significant feature and there is evidence of Neogene uplift offshore.
Line 26: Line 28:
The geomorphological detail of events in Neogene times across the Welsh massif is difficult to determine, but the landforms suggest they were largely erosional. Most probably the evidence of any significant deposits would have been removed by the subsequent glaciation; the numerous quartz pebbles on the Carboniferous limestone surface across Castlemartin peninsula [[Media:P662437.jpg|(P662437)]], in south Pembrokeshire, continue to be quoted as being of possible Pliocene age. It was only in late Miocene–early Pliocene times that sea encroached the landmass, possibly along synclinal sags in the Celtic Sea and Bristol Channel basins. Simultaneous uplift of the land surface was probably dome-like and sporadic and, consequently, a stepped profile developed.
The geomorphological detail of events in Neogene times across the Welsh massif is difficult to determine, but the landforms suggest they were largely erosional. Most probably the evidence of any significant deposits would have been removed by the subsequent glaciation; the numerous quartz pebbles on the Carboniferous limestone surface across Castlemartin peninsula [[Media:P662437.jpg|(P662437)]], in south Pembrokeshire, continue to be quoted as being of possible Pliocene age. It was only in late Miocene–early Pliocene times that sea encroached the landmass, possibly along synclinal sags in the Celtic Sea and Bristol Channel basins. Simultaneous uplift of the land surface was probably dome-like and sporadic and, consequently, a stepped profile developed.


Subsequent glacial erosion has obscured many of the geomorphological features that were sculpted through Cainozoic times, and there has been much conflicting interpretation of the features. For example, a possible concordance between north and south Wales in summit heights below 600 m, has been attributed to eustatic uplift, in stepped pulses, and a falling sea level. However, O T Jones saw an inclination from north Wales southwards, which he attributed to subaerial (Triassic) erosion and Cainozoic tilting; such extrapolation is more difficult with the possible residual surfaces over 335 m (1100 feet). Later E H Brown recognised four major 'peneplains', each ranging through 100 m or more, and ascribed them to subaerial erosion and eustatic uplift. When afforded with extended views from the numerous vantage points about Wales the existence of such ‘levels’ or ‘surfaces’, in general terms, is most persuasive but their detail is much more difficult to record. In the Brown analysis, the ‘Low Peneplain’, 215 to 335 m above OD, and probably late Pliocene, is most clearly distinguished on the sides of the Tywi and Teifi valleys and in the inner reaches of the Monmouthshire valleys; in the mountains of north-west Wales it cannot be distinguished. The ‘Middle Peneplain’, 460 to 490 m above OD, and probably an intermediate Neogene stage, is most clearly developed in the scarp and dip slopes of Mynydd Eppynt and the hills to the south-west. However, it is possible that these features, and similar notched features at an equivalent level in the Old Red Sandstone of the Brecon Beacons and in the Pennant Sandstone in the coalfield, may be due to the underlying geology. The ‘High Plateau’, 520 to 580 m above OD, possibly of mid-Cainozoic uplift, was identified mainly by its height rather than its topography as it is totally fragmented and widely distributed between the escarpments in the Brecon Beacons and Black Mountains, the complex features between Plynlimon and Drygarn, and about Llyn Vyrnwy. The uppermost, ‘summit Plain’, 615 to 1100 m above OD, and possibly the sub-Mesozoic surface exhumed, was clearly confined to the summits but their isolation makes their interconnection difficult; just to link the Snowdonia summits in such a way would necessitate a domed warping with an amplitude of some 300 m in a distance of 8 km. To reconstruct a wider surface, between the Snowdonia summits and Pen y Fan (886 m OD) in south Wales, with Arenig Fawr (855 m OD), Cadair Idris (890 m OD) and Radnor Forest (660 m OD), in between, is much more difficult.
Subsequent glacial erosion has obscured many of the geomorphological features that were sculpted through Cainozoic times, and there has been much conflicting interpretation of the features. For example, a possible concordance between north and south Wales in summit heights below 600 m, has been attributed to eustatic uplift, in stepped pulses, and a falling sea level. However, O T Jones saw an inclination from north Wales southwards, which he attributed to subaerial (Triassic) erosion and Cainozoic tilting; such extrapolation is more difficult with the possible residual surfaces over 335 m (1100 feet). Later E H Brown recognised four major 'peneplains', each ranging through 100 m or more, and ascribed them to subaerial erosion and eustatic uplift. When afforded with extended views from the numerous vantage points about Wales the existence of such ‘levels’ or ‘surfaces’, in general terms, is most persuasive but their detail is much more difficult to record. In the Brown analysis, the ‘Low Peneplain’, 215 to 335 m above OD, and probably late Pliocene, is most clearly distinguished on the sides of the Tywi and Teifi valleys and in the inner reaches of the Monmouthshire valleys; in the mountains of north-west Wales it cannot be distinguished. The ‘Middle Peneplain’, 460 to 490 m above OD, and probably an intermediate Neogene stage, is most clearly developed in the scarp and dip slopes of Mynydd Eppynt and the hills to the south-west. However, it is possible that these features, and similar notched features at an equivalent level in the Old Red Sandstone of the Brecon Beacons and in the Pennant Sandstone in the coalfield, may be due to the underlying geology. The ‘High Plateau’, 520 to 580 m above OD, possibly of mid-Cainozoic uplift, was identified mainly by its height rather than its topography as it is totally fragmented and widely distributed between the escarpments in the Brecon Beacons and Black Mountains, the complex features between Plynlimon and Drygarn, and about Llyn Vyrnwy. The uppermost, ‘summit Plain’, 615 to 1100 m above OD, and possibly the sub-Mesozoic surface exhumed, was clearly confined to the summits but their isolation makes their interconnection difficult; just to link the Snowdonia summits in such a way would necessitate a domed warping with an amplitude of some 300 m in a distance of 8 km. To reconstruct a wider surface, between the Snowdonia summits and Pen y Fan (886 m OD) in south Wales, with Arenig Fawr (855 m OD), Cadair Idris (890 m OD) and Radnor Forest (660 m OD), in between, is much more difficult.


The ‘coastal plateau’ is probably the best developed erosion surface [[Media:P662437.jpg|(P662437)]], although it is a composite feature in which erosion on the lower platforms excavate and locally obliterate the higher, and older, platforms. The recognition of outliers of higher platforms above lower platforms reflect pulses of emergence, and support the broad subdivision into the ‘200 feet’, ‘400 feet’ and ‘600 feet’ platforms in Gower, Pembrokeshire, Anglesey and Llyn. However, there may be a vertical range up to 60 m due to variations in the rate of uplift. The ‘600 feet’ platform has been recognised around much of the Welsh coast, apart from in the vicinity of the Llanbedr Fault in Merionethshire where, as on Llyˆn, Anglesey and Arfon, it has been removed by erosion of the lower platforms. Its apparent uniformity is more easily reconciled by eustatic uplift and a beach rather than a subaerial peneplain. The ‘400 feet’ and the ‘200 feet’ platforms are more clearly defined, but again they show a wide variation in altitude. For example, the ‘200 feet’ platform has been ascribed to features at 50 m above OD in Gower and Pembrokeshire and at 30 m above OD in the Vale of Glamorgan. The only evidence of a possible Neogene or later age is the truncation of the pipe clays (Oligocene) by the ‘200 feet’ platform at Flimston, south Pembrokeshire.
The ‘coastal plateau’ is probably the best developed erosion surface [[Media:P662437.jpg|(P662437)]], although it is a composite feature in which erosion on the lower platforms excavate and locally obliterate the higher, and older, platforms. The recognition of outliers of higher platforms above lower platforms reflect pulses of emergence, and support the broad subdivision into the ‘200 feet’, ‘400 feet’ and ‘600 feet’ platforms in Gower, Pembrokeshire, Anglesey and Llyn. However, there may be a vertical range up to 60 m due to variations in the rate of uplift. The ‘600 feet’ platform has been recognised around much of the Welsh coast, apart from in the vicinity of the Llanbedr Fault in Merionethshire where, as on Llyˆn, Anglesey and Arfon, it has been removed by erosion of the lower platforms. Its apparent uniformity is more easily reconciled by eustatic uplift and a beach rather than a subaerial peneplain. The ‘400 feet’ and the ‘200 feet’ platforms are more clearly defined, but again they show a wide variation in altitude. For example, the ‘200 feet’ platform has been ascribed to features at 50 m above OD in Gower and Pembrokeshire and at 30 m above OD in the Vale of Glamorgan. The only evidence of a possible Neogene or later age is the truncation of the pipe clays (Oligocene) by the ‘200 feet’ platform at Flimston, south Pembrokeshire.


The controlling influence on the drainage patterns across Wales has long been a major topic of geomorphological discussion, and even now that there is little evidence to suggest that the patterns were superimposed from a Chalk cover, there is still much disagreement. Clearly, through the Cainozoic, the intermittent development of the plateau surfaces was a significant influence. The knick points in the long profiles of most river valleys indicate repeated phases of uplift. Of particular interest are the south Wales cave systems in the Dinantian limestones along the north crop. Entries into the Agen Allwedd and Craig y Ffynnon systems, between Mynydd Llangynidr and Llangattwg, lie some 200 m above the current level of the River Usk and this possibly indicates the minimum uplift since Neogene times. In addition, there is much evidence of river capture as the early radial drainage patterns, of the high plateau stage, diminishes and the stream network becomes increasingly constrained by the geology. The drainage patterns across the south Wales coalfield, from the edge of the Old Red Sandstone escarpment, show no evidence of having been influenced by the Variscan fold and fault patterns. Similarly, the drainage of the Lower Palaeozoic terrain in the Tywi, Wye and Severn river systems provides equally striking disregard of the Caledonian structures. In north Wales, the distribution of the drainage patterns about the major structures are broadly radial. The possibility of the removal of most or all of the Mesozoic cover prior to the Neogene would necessitate the development of a mosaic of separate drainage centres about the summits of Snowdonia, the Harlech Dome, Cadair Idris and the Arans, Plynlimon and Drygarn and the Old Red Sandstone escarpment. The classic asymmetry of the drainage, with long gentle profiles of rivers such as the Severn and Wye to the east and the short steep profiles of the westward-flowing rivers (e.g. the Rheidol and Ystwyth), was enhanced by contemporaneous movement along the Llanbedr Fault.
The controlling influence on the drainage patterns across Wales has long been a major topic of geomorphological discussion, and even now that there is little evidence to suggest that the patterns were superimposed from a Chalk cover, there is still much disagreement. Clearly, through the Cainozoic, the intermittent development of the plateau surfaces was a significant influence. The knick points in the long profiles of most river valleys indicate repeated phases of uplift. Of particular interest are the south Wales cave systems in the Dinantian limestones along the north crop. Entries into the Agen Allwedd and Craig y Ffynnon systems, between Mynydd Llangynidr and Llangattwg, lie some 200 m above the current level of the River Usk and this possibly indicates the minimum uplift since Neogene times. In addition, there is much evidence of river capture as the early radial drainage patterns, of the high plateau stage, diminishes and the stream network becomes increasingly constrained by the geology. The drainage patterns across the south Wales coalfield, from the edge of the Old Red Sandstone escarpment, show no evidence of having been influenced by the Variscan fold and fault patterns. Similarly, the drainage of the Lower Palaeozoic terrain in the Tywi, Wye and Severn river systems provides equally striking disregard of the Caledonian structures. In north Wales, the distribution of the drainage patterns about the major structures are broadly radial. The possibility of the removal of most or all of the Mesozoic cover prior to the Neogene would necessitate the development of a mosaic of separate drainage centres about the summits of Snowdonia, the Harlech Dome, Cadair Idris and the Arans, Plynlimon and Drygarn and the Old Red Sandstone escarpment. The classic asymmetry of the drainage, with long gentle profiles of rivers such as the Severn and Wye to the east and the short steep profiles of the westward-flowing rivers (e.g. the Rheidol and Ystwyth), was enhanced by contemporaneous movement along the Llanbedr Fault.
 
== Bibliography ==
== Bibliography ==
BEVINS, R E, HORAK, J M, EVANS, A D, and MORGAN, R. 1996. Palaeogene dyke swarm, north-west Wales: evidence for Cenozoic sinistral fault movement. ''Journal of the'' ''Geological Society of London'', Vol. 153, 177–180.
BEVINS, R E, HORAK, J M, EVANS, A D, and MORGAN, R. 1996. Palaeogene dyke swarm, north-west Wales: evidence for Cenozoic sinistral fault movement. ''Journal of the'' ''Geological Society of London'', Vol. 153, 177–180.
Line 43: Line 46:


JOHN, B S. 1970. Pembrokeshire. 267–314 in ''The glaciations of Wales and adjoining'' ''regions''. LEWIS, C A (editor). (London: Longman.)
JOHN, B S. 1970. Pembrokeshire. 267–314 in ''The glaciations of Wales and adjoining'' ''regions''. LEWIS, C A (editor). (London: Longman.)
[[Category:Wales]]
{{British regional geology of Wales - contents}}
[[Category:Regional Geology of Wales]]

Latest revision as of 09:53, 5 May 2016

From: Howells, M F. 2007. British regional geology: Wales. Keyworth, Nottingham: British Geological Survey.

Palaeogene strata in offshore basins (after Tappin et al., 1994). P916208.
Depositional model for Cardigan Bay, early Palaeogene (after Tappin et al., 1994). P916209.
Palaeokarstic surface at the top of the Dinantian Dowlais Limestone overlain unconformably by pebbly grit at the base of the Namurian; the grit also infills large solution cavities in the limestones, Trefil Quarry (MFH P662436).
Platform across truncated Dinantian limestones near St Govan’s Head, south Pembrokeshire (MFH P662437).

Sea level rose progressively throughout most of Cretaceous time, although in the final stages, Campanian to Maastrichtian, there was a reversal of this trend and most of the basins close to the Welsh coast were inverted and, by early Palaeogene times, were subaerially exposed. The uplift of the basins led to erosion and, for the most part, succeeding deposition was mainly terrestrial. During this period, the climate was tropical or subtropical and it was only later, in Neogene times, that there was a gradual reduction in temperature. Palaeogene sediments have been determined in the St George’s Channel, South Celtic Sea and North Celtic Sea basins. Smaller outliers are preserved in half-grabens in the Cardigan Bay and Bristol Channel basins. The only Neogene sediments recorded are the Lower Miocene sediments in the Mochras Borehole.

The morphology of the main Cainozoic basins was controlled by pre-existing structures; the north-easterly trend of the St George’s Channel Basin is Caledonian, and the more easterly trend of the South Celtic Sea and the Bristol Channel basins is Variscan. In the St George’s Channel Basin, the thickest Cainozoic accumulations coincide with those of the Mesozoic, whereas in the east–west-trending basins they are offset laterally. North of the Variscan Front, the absence of the upper part of the Cretaceous sequence in the St George’s Channel and Cardigan Bay basins may be due to non-deposition or, more probably, uplift and erosion, which would also account for the unconformity between the Jurassic and Cainozoic sequences. However, in St George’s Channel Basin, the unconformity has been ascribed to basin inversion with movement along St George’s and associated faults. To the south of the Variscan Front, the South Celtic Sea Basin was also inverted, although uplift was generally more regional.

The sedimentary character of the Cainozoic sequence is best known from the Mochras Borehole, sited close to the faulted margin of the Cardigan Bay Basin, where 524 m of Middle Oligocene to Lower Miocene strata were proved. The eastward-dipping sequence is contained in a half-graben (P916208), and rests unconformably on Lower Jurassic strata. The sequence has been divided into three lithostratigraphical units. The Basal Red Unit comprises coarse-grained, cobble conglomerates, up to 14 m thick, with interbedded fining-upward sand to clay sequences, 1 to 2 m thick. The conglomerates are intepreted as alluvial debris is marked by a gradual change in colour from red to grey, and is characterised by upward-fining cycles of silt and clay with a carbonaceous and lignite component. The facies associations are characterisitic of a floodplain environment. The contact with the overlying, uppermost Lignite and Clay Unit is marked by a gradual increase in the proportion of lignite. The olive-green sediments form sharp-based, upward-fining sand to clay sequences laid down in brackish water or on well drained, vegetated surfaces. In many places, internal structures are obliterated by bioturbation, organic debris is scattered throughout the upper parts of the cycles and spherulitic siderite is common. The sequence suggests that swamp conditions, with possible seasonal variations, were established at the height of the water table.

Lithologically similar sequences were intersected by a borehole at Tonfannau, sited on marine alluvium between the Mawddach and Dyfi estuaries, and in two offshore boreholes close to the northern edge of Cardigan Bay. Farther offshore, in St George’s Channel Basin (P916208), the thickest Cainozoic sequence is intersected by the St George’s Fault along part of which a salt wall has been emplaced. North-west of the fault, in a faulted syncline, over 1500 m of Palaeogene (Eocene to Oligocene) strata lie unconformably on Cretaceous strata, and four boreholes have shown that the overall characteristics of the sediments are closely comparable with the Mochras sequence. A similar environment is envisaged — a floodplain with fluvial channel sands and overbank deposits. Between St George’s Channel Basin and the Welsh mainland, a small outlier of Eocene to mid-Oligocene sediments has been preserved in the Teifi basin, where 10 m of dark green and pale brown laminated clay, with plant fragments and glauconite, indicate possibly the most northerly marine incursion about Wales. The general disposition of Palaeogene facies across Cardigan Bay reflects a widely meandering river encroaching a swamp-dominated flood plain (P916209).

In the South Celtic Sea Basin, over 500 m of Palaeogene sediments have been dated as mid-Eocene to Oligocene on the basis of a sparse microfloral assemblage. One borehole penetrated finely laminated, green, micaceous, glauconitic and lignitic sandstone with interbeds of silty clay and dark brown to black lignite. Lignite is more abundant near the base of the sequence and is thought to reflect deposition in a low-energy marginal marine environment. The upward increase in sandy glauconitic sediments suggests an increasing marine influence, although the persistent lignite content suggests close proximity to a vegetated landmass.

During Cainozoic times, the Bristol Channel Basin was no longer extant, but farther south, to the west of Lundy Island, a new basin, the Bristol Channel Marginal Basin, developed. This basin has a downfaulted northern edge, and 300 m of Cainozoic strata are preserved within it. Seismic records indicate that a lower sequence, possibly sandstone, is overlain by clay and lignite, which have been recovered from the sea floor. East of Lundy Island, 335 m of braided channel and flood plain deposits are preserved in the Stanley Bank Basin on the east side of an extension of the Sticklepath–Lustleigh Fault Zone. Lundy Island is the site of an early Eocene igneous complex — a coarse- to fine-grained granite is itself intruded by numerous south-east-trending dykes of olivine and quartz dolerite, trachyte and trachyandesite. Dolerite dykes of Eocene age have been distinguished in a swarm, offshore, north­east of Anglesey, and many have been determined across Anglesey, Llŷn and Arfon with some extending into Snowdonia and beyond. These Paleocene–early Eocene igneous rocks lie at the edge of a large igneous province (Thulean), which straddled the eastern edge of Greenland, Iceland and extended to the Hebrides and north-east Ireland. The dykes in north Wales are the fringe to a volcanic centre at Mourne in Northern Ireland, although the currently exposed dykes imply a significant cover at the time of their intrusion and, consequently, considerable erosion since Eocene times.

The extensive offshore record of Palaeogene–Neogene sedimentation contrasts with the patchy evidence onshore. Evidence of Palaeogene sediments on mainland Wales is restricted to deeply weathered regoliths, such as the ‘pocket deposits’ on the Carboniferous limestone outcrops between Llandudno and Mold in north Wales. The deposits include both alluvial sediments, introduced by subsidence into solution cavities, stratified sand and clay (products of subtropical, subaerial weathering, with gibbsite and kaolinite) and interbedded lignite, introduced by underground streams. Similar mottled clays, some 15 m thick, occur within limestone cavities at Flimston, near Castlemartin, in south Pembrokeshire, and others are exposed in the quarry at Trefil north of Merthyr Tydfil (P662436). Elsewhere in south Wales, in situ weathered profiles in Namurian sandstone along the north crop and in Old Red Sandstone orthoquartzite in the Black Mountains have also been regarded as Cainozoic in age.

The patchy onshore Palaeogene–Neogene record makes it difficult to determine the geological evolution of this period in Wales. The task is further complicated by the variety of processes — continental and regional tectonism with concomitant movement of magma, climate changes and sea level fluctuations — that were particularly active throughout the period. The most clearly defined morphological feature in Wales is the ancient Lower Palaeozoic Caledonian massif, which developed from uplift that culminated in Mid Devonian times. Later, uplift was reactivated through Carboniferous times, and then the massif was enlarged by accretion in the late stages of the Variscan orogeny. However, interpretation of subsequent Mesozoic events and the final emergence in Cainozoic times is inhibited by the extremely restricted outcrops. In early Mesozoic times, the Palaeozoic landmass was probably reduced to a relatively low peneplain on which Triassic and Lias Group sediments were probably deposited and subsequently eroded. Later, with the exception of a restricted area in north-west Wales, the area was submerged beneath late Cretaceous seas. However, no Cretaceous rocks have been found on land, in solution cavities in Carboniferous limestone or in the Cainozoic of the Mochras Borehole. It is possible that the massif was uplifted prior to the transgression and, consequently, the preglacial evolution of Wales was entirely controlled by Cainozoic tectonic events and sea-level changes.

The relationship of the Welsh massif to the Mesozoic — Cainozoic offshore basins has long been a controversial subject, but the debate was intensified during the sinking of the Mochras Borehole and its completion in 1972. The juxtaposition of the thick Mesozoic and Palaeogene sequences in the borehole with Cambrian rocks that are exposed on the flank of the Harlech Dome just 2.5 km to the east, emphasises the influence of the intervening contemporaneous, coast-aligned Llanbedr Fault on sedimentation. This fault has an estimated downthrow of not less than 3750 m on the postulated base of the Triassic, and not less than 1350 m on the base of the Palaeogene.

The contrast in the preserved Mesozoic–Cainozoic sequences in the offshore basins to the north and south of the Variscan Front reflects a marked contrast in deformation. To the north, in St George’s Channel and Cardigan Bay basins, no Cretaceous strata are preserved and the Palaeogene sequence, of mainly Eocene to Oligocene terrestrial rocks, rests on Upper Jurassic rocks. To the south of the Variscan Front, in the Celtic Sea Basin, variable Lower Cretaceous and thick Chalk sequences overlie Jurassic strata, and in the main Bristol Channel Basin the youngest preserved sediments are usually of late Jurassic–early Cretaceous age, although terrestrial Oligocene sediments are preserved in faulted sub-basins. For much of the Mesozoic and Cainozoic, the Welsh massif was not a significant feature and there is evidence of Neogene uplift offshore.

The geomorphological detail of events in Neogene times across the Welsh massif is difficult to determine, but the landforms suggest they were largely erosional. Most probably the evidence of any significant deposits would have been removed by the subsequent glaciation; the numerous quartz pebbles on the Carboniferous limestone surface across Castlemartin peninsula (P662437), in south Pembrokeshire, continue to be quoted as being of possible Pliocene age. It was only in late Miocene–early Pliocene times that sea encroached the landmass, possibly along synclinal sags in the Celtic Sea and Bristol Channel basins. Simultaneous uplift of the land surface was probably dome-like and sporadic and, consequently, a stepped profile developed.

Subsequent glacial erosion has obscured many of the geomorphological features that were sculpted through Cainozoic times, and there has been much conflicting interpretation of the features. For example, a possible concordance between north and south Wales in summit heights below 600 m, has been attributed to eustatic uplift, in stepped pulses, and a falling sea level. However, O T Jones saw an inclination from north Wales southwards, which he attributed to subaerial (Triassic) erosion and Cainozoic tilting; such extrapolation is more difficult with the possible residual surfaces over 335 m (1100 feet). Later E H Brown recognised four major 'peneplains', each ranging through 100 m or more, and ascribed them to subaerial erosion and eustatic uplift. When afforded with extended views from the numerous vantage points about Wales the existence of such ‘levels’ or ‘surfaces’, in general terms, is most persuasive but their detail is much more difficult to record. In the Brown analysis, the ‘Low Peneplain’, 215 to 335 m above OD, and probably late Pliocene, is most clearly distinguished on the sides of the Tywi and Teifi valleys and in the inner reaches of the Monmouthshire valleys; in the mountains of north-west Wales it cannot be distinguished. The ‘Middle Peneplain’, 460 to 490 m above OD, and probably an intermediate Neogene stage, is most clearly developed in the scarp and dip slopes of Mynydd Eppynt and the hills to the south-west. However, it is possible that these features, and similar notched features at an equivalent level in the Old Red Sandstone of the Brecon Beacons and in the Pennant Sandstone in the coalfield, may be due to the underlying geology. The ‘High Plateau’, 520 to 580 m above OD, possibly of mid-Cainozoic uplift, was identified mainly by its height rather than its topography as it is totally fragmented and widely distributed between the escarpments in the Brecon Beacons and Black Mountains, the complex features between Plynlimon and Drygarn, and about Llyn Vyrnwy. The uppermost, ‘summit Plain’, 615 to 1100 m above OD, and possibly the sub-Mesozoic surface exhumed, was clearly confined to the summits but their isolation makes their interconnection difficult; just to link the Snowdonia summits in such a way would necessitate a domed warping with an amplitude of some 300 m in a distance of 8 km. To reconstruct a wider surface, between the Snowdonia summits and Pen y Fan (886 m OD) in south Wales, with Arenig Fawr (855 m OD), Cadair Idris (890 m OD) and Radnor Forest (660 m OD), in between, is much more difficult.

The ‘coastal plateau’ is probably the best developed erosion surface (P662437), although it is a composite feature in which erosion on the lower platforms excavate and locally obliterate the higher, and older, platforms. The recognition of outliers of higher platforms above lower platforms reflect pulses of emergence, and support the broad subdivision into the ‘200 feet’, ‘400 feet’ and ‘600 feet’ platforms in Gower, Pembrokeshire, Anglesey and Llyn. However, there may be a vertical range up to 60 m due to variations in the rate of uplift. The ‘600 feet’ platform has been recognised around much of the Welsh coast, apart from in the vicinity of the Llanbedr Fault in Merionethshire where, as on Llyˆn, Anglesey and Arfon, it has been removed by erosion of the lower platforms. Its apparent uniformity is more easily reconciled by eustatic uplift and a beach rather than a subaerial peneplain. The ‘400 feet’ and the ‘200 feet’ platforms are more clearly defined, but again they show a wide variation in altitude. For example, the ‘200 feet’ platform has been ascribed to features at 50 m above OD in Gower and Pembrokeshire and at 30 m above OD in the Vale of Glamorgan. The only evidence of a possible Neogene or later age is the truncation of the pipe clays (Oligocene) by the ‘200 feet’ platform at Flimston, south Pembrokeshire.

The controlling influence on the drainage patterns across Wales has long been a major topic of geomorphological discussion, and even now that there is little evidence to suggest that the patterns were superimposed from a Chalk cover, there is still much disagreement. Clearly, through the Cainozoic, the intermittent development of the plateau surfaces was a significant influence. The knick points in the long profiles of most river valleys indicate repeated phases of uplift. Of particular interest are the south Wales cave systems in the Dinantian limestones along the north crop. Entries into the Agen Allwedd and Craig y Ffynnon systems, between Mynydd Llangynidr and Llangattwg, lie some 200 m above the current level of the River Usk and this possibly indicates the minimum uplift since Neogene times. In addition, there is much evidence of river capture as the early radial drainage patterns, of the high plateau stage, diminishes and the stream network becomes increasingly constrained by the geology. The drainage patterns across the south Wales coalfield, from the edge of the Old Red Sandstone escarpment, show no evidence of having been influenced by the Variscan fold and fault patterns. Similarly, the drainage of the Lower Palaeozoic terrain in the Tywi, Wye and Severn river systems provides equally striking disregard of the Caledonian structures. In north Wales, the distribution of the drainage patterns about the major structures are broadly radial. The possibility of the removal of most or all of the Mesozoic cover prior to the Neogene would necessitate the development of a mosaic of separate drainage centres about the summits of Snowdonia, the Harlech Dome, Cadair Idris and the Arans, Plynlimon and Drygarn and the Old Red Sandstone escarpment. The classic asymmetry of the drainage, with long gentle profiles of rivers such as the Severn and Wye to the east and the short steep profiles of the westward-flowing rivers (e.g. the Rheidol and Ystwyth), was enhanced by contemporaneous movement along the Llanbedr Fault.

Bibliography

BEVINS, R E, HORAK, J M, EVANS, A D, and MORGAN, R. 1996. Palaeogene dyke swarm, north-west Wales: evidence for Cenozoic sinistral fault movement. Journal of the Geological Society of London, Vol. 153, 177–180.

BROWN, E H. 1960. The relief and drainage of Wales. (Cardiff: University of Wales.)

CAMPBELL, S, and BOWEN, D Q (editors). 1989. Quaternary of Wales. Geological Conservation Review Series, No. 2. (Peterborough: Nature Conservancy Council.)

GEORGE, T N. 1974. The Cenozoic evolution of Wales. 341–371 in The Upper Palaeozoic and post-Palaeozoic rocks of Wales. OWEN, T R (editor). (Cardiff: University of Wales Press.)

GODWIN, H. 1943. Coastal peat beds of the British Isles and North Sea. Journal of Ecology, Vol. 31, 199–247.

JOHN, B S. 1970. Pembrokeshire. 267–314 in The glaciations of Wales and adjoining regions. LEWIS, C A (editor). (London: Longman.)