Content and licensingview original scan buy a printed copy

Geology of the Irvine district. Memoir for 1:50 000 geological sheet 22W and part of Sheet 21E (Scotland)

by S K Monro

Bibliographical reference: Monro, S K. 1999. Geology of the Irvine district. Memoir of the British Geological Survey, Sheet 22W and part of Sheet 21E (Scotland)

Contributors

• Geophysics K E Rollin
• Hydrogeology N S Robins
• Igneous rocks D Stephenson
• Palaeontology P J Brand, M T Dean, D K Graham

London: The Stationery Office 1999. © NEW; copyright 1999 First published 1999. ISBN 11 884550 0. Printed in the UK for The Stationery Office J74212 C6 2/99

The grid used in the figures is the National Grid taken from the Ordnance Survey map. (Figure 1) is based on material from Ordnance Survey 1:50 000 scale maps, numbers 63, 64 and 70. © Crown copyright reserved.  Ordnance Survey Licence No. GD272191/1999

• Author
• S K Monro, BSc, PhD, CGeol British Geological Survey, Edinburgh
• Contributors
• M T Dean, BSc, MPhil., D Stephenson, BSc, PhD British Geological Survey, Edinburgh
• N S Robins, MSc, CGeol British Geological Survey, Wallingford
• K E Rollin, BSc British Geological Survey; Keyworth
• P J Brand, BSc., D K Graham, BA formerly British Geological Survey

(Front cover) Cover photograph: Main rock platform south of Seamill [NS 205 460] with prominent cliff feature at 15 m OD. Rocks of the Upper Devonian, Seamill Sandstone Formation dip gently to the south-west (right). (MNS1892) (Photographer: A Christie)

(Rear cover)

Acknowledgements

Cores from boreholes drilled by civil engineering contractors to investigate foundation conditions for roads, buildings and other works have contributed substantially to the understanding of the geology of this district. The assistance of the Irvine Development Corporation in the assembly of the database of site investigation information in the area of Irvine New Town is gratefully acknowledged.

R B Wilson, P J Brand and D K Graham were responsible for the palaeontological identifications. Chapter 10 was written by P J Brand, with the assistance of M T Dean, and D K Graham contributed to Chapter 15. D Stephenson wrote the section on volcanic rocks in chapters 7 and 11, and the account of intrusive igneous rocks (Chapter 12). K E Rollin contributed the geophysical account (Chapter 13) and N S Robins wrote the section on water resources and landfill in Chapter 2. The photographs were mainly taken by T Bain. The memoir was compiled by S K Monro and edited by I B Cameron, D J Fettes, A D McAdam, AA Jackson and J I Chisholm.

Notes

Throughout this memoir the word 'district' refers to the area covered by the 1:50 000 scale map, Irvine (Sheet 22W and part of 21E).

National Grid references are given in square brackets; unless otherwise stated, all lie within the 100 km square NS.

Preface

An understanding of geology is essential to the sustainable development of the United Kingdom. Geology is particularly important in relation to the exploration for and development of mineral resources, the recognition of natural and man-made geological hazards, and the implication of these factors for land-use planning by national and local government. In recognition of this, the British Geological Survey is funded by central government to improve the understanding of the three-dimensional geology of the United Kingdom through a programme of data collection, interpretation, publication and archiving. One aim of this programme is to ensure coverage of the United Kingdom land mass with modern 1:50 000 geological maps and explanatory memoirs.

This memoir on the Irvine district, with the accompanying 1:50 000 maps of the Solid and Drift geology, provides an up-to-date account of the geology of an area adjacent to the Firth of Clyde, including a part of the island of Great Cumbrae and the Hunterston peninsula. The mainland area extends from Largs and Lochwinnoch in the north to Troon in the south.

The mainland coastal strip from Ardrossan north to Largs is rural, and the sandy beaches of Seamill and West Kilbride, the sheltered yacht moorings at. Largs and the attractive scenery of the hinterland are important tourist attractions. Within this stretch of coastline, however, are the Hunterston Power Station and the former Hunterston Ore Terminal, which were sited at the northern end of the Hunterston peninsula to take advantage of the deep water in the Firth of Clyde. South of Ardrossan the coastal zone is heavily industrialised with a mix of large-scale industrial developments and many new smaller-scale enterprises associated with Irvine New Town.

The inland area likewise has its contrasts. The industrialised towns such as Dairy and Kilbirnie in the north of the district are flanked by sparsely populated hills. In the south of the district the strip from Irvine and Troon eastward towards Kilmarnock is now a conurbation of interconnected small towns. The population distribution reflects the underlying geology, because during the 19th and early part of the 20th centuries communities sprang up in areas where coal, ironstone, fireclay, sandstone and limestone were being mined. These minerals were also the feedstock of the smelting and steelmaking industries that were prominent in the district until recently.

Revision of the geology of the Irvine district took place during the redevelopment of Irvine New Town. At an early stage it was recognised that effective planning of the redevelopment required information about the underlying geology, partly to highlight existing resources, but more particularly to identify the distribution of old mine workings which are a constraint to land use. A partnership developed between the British Geological Survey and Irvine Development Corporation, which facilitated redevelopment and added considerably to the database of boreholes and so to the mapping of this district. The reassessment of the geology of the Irvine district, and particularly the expansion of the database of borehole information, has also increased understanding of the controls which were being exercised on sedimentation during late Devonian and Carboniferous times.

The interpretation contained in this memoir, together with the accompanying 1:50 000 maps of the Solid and Drift geology, provide a basis for developing the Irvine district in a sensible and sustainable manner.

David A Falvey, PhD Director,  Kingsley Dunham Centre,  British Geological Survey,  Keyworth Nottingham NG12 5GG

Geology of the Irvine district—summary

This memoir, to accompany the remapped Irvine (22W) sheet, describes the district's geology in a format that relates the rich and diverse geology of the Irvine district to a story of environmental change over the last 400 million years of Earth's history. The materials deposited in these environments were the feedstock for the economic development of this area and their extraction has left a legacy of undermining. The key to future prosperity lies in understanding the underlying geology and how it will impact on surface development.

An introductory section (Chapter 1) puts the geology in context and discusses the relationship of the scenery to the underlying rocks. The Applied Geology section (Chapter 2) looks at the geology from the perspective of the planner and engineer. Here, the major issues of importance to land-use planning are discussed: the distribution of mineral resources and metalliferous minerals, water resources, landfill and made ground. To the engineer involved with the design and implementation of major infrastructure, and industrial and housing developments, these are key issues. This chapter illustrates the relevance of the new survey to these issues, and highlights the extent and depth of the data available from BGS to the applied geology professional.

The following sections describe and interpret scientifically all aspects of the geology of the Irvine district. The nature of the oldest rocks of the district, of Lower Devonian age, are considered in Chapter 3. The continental alluvial deposits of Upper Devonian age are described in Chapter 4 and fitted into the conditions of deposition and palaeogeography that existed at that time.

The Carboniferous strata are introduced in Chapter 5 with a discussion of the classification used in the variable succession of the Irvine district. The succeeding chapters 6 to 9 describe the Carboniferous succession from the basal Inverclyde Group through to the Coal Measures. Each of these chapters draws the sedimentary succession together with an interpretation of the environment of deposition. Volcanic episodes, which relate closely to the tectonic evolution of the district, are discussed in their appropriate time period from the Clyde Plateau Volcanic Formation through to the Troon Volcanic Member.

Biostratigraphy is an important tool both in correlation and in the interpretation of environmental conditions during the Carboniferous. Chapter 10 gives a detailed account of the variation in fauna and flora throughout this period of time. The Permian rocks of the Irvine district are represented by the Mauchline Volcanic Formation and these are described in Chapter 11 and related to the earlier pattern of volcanic activity. The diverse suites of intrusive igneous rocks are described in Chapter 12, including the exotic lithologies found in some vents which give clues to the nature of crust and upper mantle beneath the Irvine district.

Investigation and interpretation of the deeper structure also involves the use of geophysical data and Chapter 13 describes these data and presents possible models of the deeper structure. Chapter 14 discusses the main structural elements found at the surface and relates these to the various plate tectonic models, which have been proposed for the evolution of the Midland Valley.

The most recent period of geological history is discussed in Chapter 15. The last glacial episode did much to shape the landscape of the Irvine district, eroding the softer rocks and depositing debris from the glaciers and from the meltwater.

The appendix dealing with information sources will be a useful starting point in seeking more data about the nature of the diverse rock-types which occur in the Irvine district.

Geological sequence

Chapter 1 Introduction

The district described in this memoir falls within the Irvine Sheet, which incorporates Sheet 22W and the eastern part of Sheet 21E of the 1:50 000 geological map of Scotland. The district is flanked to the west by the Firth of Clyde and includes a part of the island of Great Cumbrae. The mainland area extends from Largs and Lochwinnoch in the north to Troon in the south (Figure 2).

In the northern part of the district, strata of Devonian and early Carboniferous age occur on the coastal strip of low-lying ground between Saltcoats and Largs (Figure 1). This is a fertile area of farm land and deep sandy soils. Inland the topography is dominated by high ground, with peaty soils underlain by the Clyde Plateau Volcanic Formation. The volcanic rocks also crop out on the eastern side of the district between Beith and Dunlop where the expression of this underlying rock type is again an upland with thin peaty soils. Between the two upland areas lies a broad lowland belt in which lie the towns of Dairy and Kilbirnie. The lowland belt corresponds to an outcrop of younger, mostly sedimentary rocks ranging in age from Visean to Namurian, and extends southwards as far as Saltcoats in the west and Kilmaurs in the east.

The southern part of the district is underlain by Westphalian Coal Measures, which were heavily exploited in the past. The strata here consist mostly of sedimentary rocks and these form the low ground from Stevenston south to Troon and eastwards to Kilmarnock in the adjacent district. A dolerite sill intruded into these strata forms a positive topographical feature in the Dundonald area. In the extreme south-east of the district, south of the Inchgotrick Fault, basaltic lavas of late Carboniferous to early Permian age occur but, unlike the earlier Clyde Plateau lavas, these do not form prominent uplands.

The landscape of the Irvine district has also been fashioned by the processes of glaciation that took place during the Quaternary Period, when the district was covered by ice on several occasions. The movement of the ice sheets eroded the softer rocks, leaving the harder rocks as positive features of the landscape. The landscape was also modified by the deposition of materials either directly from the ice or via the meltwaters as the ice retreated. Drumlins, which are cigar-shaped mounds of till, commonly rock-cored, are characteristic of many of the lowland parts of the Irvine district. In the immediate postglacial period when sea level was rather higher than it is at present, marine erosion caused the formation of the raised beaches which are a prominent feature of the coastal scenery to the north of Ardrossan (front cover photograph).

Deposition of the rocks that now crop out in the Irvine district began in a palacogeographical setting where the Lower Palaeozoic ocean (Iapetus) had closed and the Caledonian mountain chain had been thrown up. The mountains eroded rapidly and deposition during the early Devonian consisted mainly of alluvial fans building out from the north. Fluvial conditions were also widespread in the late Devonian, with deposition from braided streams and the limited formation of aeolian deposits. Later, in the earliest Carboniferous, evaporation led to the development of cornstones in fossil soils and cement-stones in coastal sabkhas. A period of erosion followed, and in Visean times the Clyde Plateau lavas were erupted on to an eroded land surface of different stratigraphical units. The lava pile in the north of the district is about 1 km thick, thinning towards the Dusk Water Fault. It is again thicker south of the Dusk Water Fault, but thins to nothing southward to the Inchgotrick Fault. Erosion that followed, together with variation in the original lava thickness, produced an uneven surface which exerted an important control on the pattern of subsequent sedimentation. These late Visean and Namurian rocks consist of cyclical fluviodeltaic sequences of limestones, mudstones, siltstones, sandstones, seatbeds and coals. The character of the qclicity varies in each formation, with a greater or lesser marine influence and shifts in the importance of fluvial processes.

Tectonism, and in particular movement on the major north-east-trending faults, had a significant effect on the pattern of sedimentation up until the late Namurian but was less important during deposition of the Westphalian strata. Volcanic processes also continued, with eruption of the basaltic lavas of the Troon Volcanic Formation in the late Namurian and the Mauchline Volcanic Formation in late Carboniferous and early Permian times.

Igneous intrusions were emplaced during the Carboniferous and Permian volcanic episodes, and also during the Palaeogene (Tertiary) volcanic activity which was centred outwith the district.

History of survey

The first edition of the Kilmarnock map (Sheet 22) was published in 1870, with a sheet explanation by A Geikie, J Geikie, R L Jack and R Etheridge jun. following in 1872. A Geikie continued mapping of the mainland part of the adjacent Sheet 21, to the west, and the memoir for this district was published in 1903 (Gunn et al., 1903). Between 1911 and 1924 the Kilmarnock district was resurveyed by E M Anderson, C T Clough, C H Dinham, V A Eylcs, J E Richey and G V Wilson and published in 1928. A memoir on the Ayrshire bauxitic clay by G V Wilson was published in 1922, and between 1925 and 1930 three memoirs relating to the economic geology of the Ayrshire Coalfield were also produced. The comprehensive memoir for the Kilmarnock district by J E Richey, E M Anderson and A G MacGregor followed in 1930. The present resurvey was carried out between 1968 and 1982 by I B Cameron, A Davies, J Dawson, S K Monro and D Stephenson.

History of research

The earliest accounts of the geology of the Irvine district focused on the stratigraphy and the economic deposits. In 1839 Montgomery gave 'Outlines of the geology of Renfrewshire and the north of Ayrshire' to the Highland and Agricultural Society, and in the following year J Craig presented a stratigraphical subdivision of the Carboniferous strata of the Glasgow area to the British Association. The New Statistical Account of Scotland, published in 1845, gave details of the economic deposits of coal, limestone etc. which were worked, and included detailed sections of the coal-bearing strata of the Kilbirnie area.

Between 1872 and 1911 many papers on the strati-graphical and palaeontological problems of the district were published, laying the foundations for many of the correlations used today. James Neilson (1874) recognised a richly fossiliferous mudstone (now called the Neilson Shell Bed) in the roof of the Blackball Limestone. Robert Craig contributed papers on the sedimentary rocks around Dalry and the significance of their faunal assemblages (1875, 1883, 1885, 1886, 1891). John Smith, a native of Dairy and an enthusiastic amateur geologist, made a very valuable contribution by carefully observing, collecting and recording sections in the district, many of which are no longer available. A comprehensive account of the geology of the Dairy area was produced (Smith, 1882) and he later (1895a) recorded for the first time the presence of bauxitic clays in the district.

Further work on the bauxitic clays was done by Wilson (1922) and in 1936 De Lapparent identified the presence of boehmite and diaspore in some samples. Monro, Loughnan and Walker (1983) compared these rocks with similar deposits elsewhere in the world. The relationship between faulting and the distribution of the sedimentary rock types was investigated by Richey (1925) and Anderson (1925). Regional studies on the distribution and thickness variation of sedimentary sequences including those developed in the Irvine district have been made by Macgregor and Manson (1935), Richey (1937) and Goodlet (1957). Different aspects of the fossil record of the district have been examined by Burgess (1965), Shiells (1966), Shiells and Penn (1971) Wilson (1979) and Whyte (1981).

Major studies of all the igneous rocks of the district were made as part of the first two surveys, mainly by A Geikie, J E Richey and A G MacGregor, and detailed descriptions were included in the 1930 edition of the Kilmarnock memoir. This pioneering work undoubtedly contributed greatly to seminal works on Carboniferous and Permian igneous activity of the Midland Valley as a whole (Richey, 1928; MacGregor, 1928, 1937, 1948). The Geological Survey investigations were supplemented by independent work in the district by Tyrrell (1912, 1917a, 1928a), who made particular advances in the petrology of alkaline basic intrusions (Tyrrell, 1909, 1923, 1928b). The basic sills also attracted attention from other workers, such as Falconer (1907) and Patterson (1945, 1946) on the Saltcoats Main Sill, and Mykura (1965), who studied the effects of volatiles released from coal seams during intrusion. More recent geochemical studies and interpretations of magma genesis in the Midland Valley have incorporated data from the Irvine district, notably in the work of Smedley (1986a, b, 1988) on the Visean rocks, and Macdonald et al. (1977) and Wallis (1989) on those of the Silesian and Permian. The district includes part of the only Palaeogene sill in the mainland of southern Scotland (Mykura, 1967) and some of the laterally extensive Palaeogene dykes have formed part of a regional geochemical study (Macdonald et al., 1988). Samples from some of the igneous suites have been dated by K-Ar methods and constitute key elements in the timing of igneous activity in the Midland Valley (De Souza, 1979, 1982). A feature of many of the volcanic vents is the presence of inclusions of middle crustal, lower crustal and upper mantle material, which have yielded vital information on the deep geology of the Midland Valley through the studies of several workers (Graham and Upton, 1978; Upton et al., 1983, 1984; Hunter and Upton, 1987; Halliday et al., 1993).

Chapter 2 Applied geology

Coal, limestone and ironstone provided the foundation for the industrial development of the district, and extraction goes back to the time before systematic records were kept. Ecclesiastical communities, like the one at Kilwinning, played an important part in the early extraction of these materials, and exploitation of the mineral resources of the district was formerly much more extensive than at present. Metalliferous minerals are present in small quantities and have been worked in the past. Bulk minerals presently worked include crushed rock aggregate, limestone, and mudstone for brickmaking. Small deposits of sand and gravel are also present.

Applied geological issues

In this chapter the geological factors relevant to land-use planning and development within the district are reviewed. The key issues are identified and some are considered in more detail. The reader is referred to sources where more information can be obtained.

Key issues

mineral resources: bulk minerals and metalliferous minerals water resources landfall made ground: extent and compositional variation

Mineral resources: bulk minerals

Crushed rock aggregate

The extensive and mainly upland area in the northern part of the district includes important resources of igneous rock. The extrusive rocks (the lavas) tend to be variable in their physical characteristics, thus limiting their potential as a source of crushed rock aggregate; some are highly amygdaloidal and extensively weathered to clay minerals whereas others are fresh and massive, breaking into cuboidal blocks. The intrusive rocks (plugs, sills and dykes) are more consistent in physical properties and therefore more useful.

In the southern part of the district there are strategically important resources of intrusive rocks as well as many marginal resources in the form of lavas and badly altered intrusions. Notes on many of the old quarries are given in the memoirs on the economic geology of the Ayrshire Coalfield: Richey et al. (1925), Anderson (1925), Eyles et al. (1930), Simpson and MacGregor (1932).

Extrusive rocks

Clyde Plateau Volcanic Formation

This sequence of lavas is generally over 300 in thick, but thins to a few metres at Ardrossan. The lavas are predominantly basaltic, but more acid types ranging to trachyte and rhyolite commonly occur towards the top of the pile. Individual flows are generally between 6 and 9 m thick but may reach 20 m or more locally; basaltic flows commonly have deeply weathered tops capped by a layer of red earth (bole). Layers of tuff or other sedimentary rock are intercalated in places. Adjacent flows are typically of slightly differing composition, and some are almost completely decomposed. A 20 m-high quarry face will almost inevitably include two or three lava flows, generally of contrasting lithology. The fresh cores of adjacent flows are typically separated by a few metres of weathered, slaggy rock and bole which have to be carefully removed as waste.

These lavas generally do not constitute a very attractive resource because of the inherent difficulties of maintaining a consistent and saleable product, but they have been sought at numerous localities, mainly for road bottoming and fill. With judicious processing they are presently used for granular sub-bases and bottoming, with some single-sized aggregate for general concreting purposes. The rocks are generally too weak or inconsistent for road surfacing although polished stone values may be quite reasonable.

Basaltic lavas, both massive and amygdaloidal, are presently worked, together with a basalt dyke at Loanhead Quarry [NS 365 554] near Beith. A very hard, durable rhyolite lava is worked intermittently at Swinlees Quarry [NS 288 529] near Kilbirnie; this rock has been used extensively for armourstone and imported fill.

Troon Volcanic Member

Potential resources within this unit are restricted to the extreme north of the outcrop, between Saltcoats and Kilmaurs. Smaller outcrops occur in the vicinity of Symington. The lavas, when fresh, are hard basaltic rocks, but they are very commonly decomposed and generally have a cover of glacial till. The uppermost part of the sequence has been altered and kaolinised. The volcanic sequence varies from 9 to 170 m in thickness and comprises lavas interbedded with sandstones, fireclay and coal. Few of the individual flows will yield as much as 10 in of massive, hard, unweathered rock. There are no active quarries in the Troon Volcanic Member but they were formerly worked, most notably at Auchenharvie [NS 363 443] and Bowertrapping [NS 326 495].

Mauchline Volcanic Formation

These lavas only crop out in a very small area in the southeast corner of the district, but regionally they form a more or less continuous outcrop surrounding the Permian sandstones of the Mauchline Basin, to the south. The lavas are mainly highly vesicular olivine-basalts with amygdales of calcite and zeolites. The flows are generally thin, badly weathered, and interbedded with sedimentary rocks and tuff The compact heart-rock of the flows is typically purplish red to grey, with a red speckling caused by the alteration of olivine crystals. As much as 20 per cent of the 90 to 170 m-thick lava sequence comprises tuff or sedimentary rock. The lavas have a foreseeable use only as granular sub-base or fill.

Intrusive rocks

Plugs

Most of the plugs in the district form conspicuous hills. They are variable in composition; some are basaltic but others are of intermediate composition, such as trachyte. Many of the less remote plugs have been quarried in the past, for example Castle Hill, but none supports active quarries. Nevertheless they contain important resources, especially the larger bodies of trachytic composition.

In the Beith area there are trachytic plugs, the most notable being Lochlands Hill [NS 374 553]. Numerous neck intrusions of trachyte also occur in the Kilbirnie Hills in the district to the north, where they are associated with trachytic lavas, agglomerates and tuffs. Only one of this group, at Knockside Hill [NS 256 582], is in the Irvine district.

Plugs of basalt are less numerous and are generally smaller than those of trachyte. However, many do occur, especially in the Kilbirnie Hills; Castle Hill [NS 287 536] is the most notable. A plug of Permian age was worked until recently, together with the surrounding basaltic agglomerate, at Helenton Hill Quarry [NS 391 305], near Craigie; the material was mostly used for bottoming and imported fill but much of the rock is badly decomposed.

Sills

Sills of trachytic or rhyolitic composition crop out at intervals along the eastern shore of the Firth of Clyde from Inverkip to Ardrossan; most are intruded into the sandstones which underlie the Clyde Plateau Volcanic Formation. Dolerite sills are less common.

A lenticular intrusion of trachyte is presently wrought for armourstone and imported fill at Biglees Quarry [NS 210 515], near West Kilbride. The rock is typically medium grained and massive, pale reddish grey in colour but weathering reddish brown. The trachytes are potentially suitable for most roadstone products, although the fresher varieties may be too prone to polishing to be used in the wearing courses of roads where traffic is heavy. The fresh rocks are strong and durable and may be suitable for relatively low-shrinkage concrete and probably also for rail ballast.

The alkali dolerite and related sills are of considerable economic importance as they are generally exceptionally fresh, massive, compact, tough rocks with good wide-set blocky jointing. There is generally some vertical variation in grain size through each sill, and compositional variation is also common in some teschenite sills. The doleritic sills are thick and characteristically form near-horizontal cappings to hills which are encircled by high craggy escarpments. The larger intrusions form the Dundonald Hills [NS 359 327] and the nearby Craigie Hill [NS 423 327], both to the east of Troon.

Sills have been quarried extensively, especially at Dundonald and Craigie, and large quantities of rock were formerly shipped out from Troon to England and Wales. The Hillhouse Sill at Dundonald is at least 45 to 60 m thick and supports two active workings, Hillhouse Quarry [NS 345 342] and Hallyards Quarry [NS 359 335]. Hillhouse is perhaps the largest hard rock quarry in central Scotland and produces almost the complete range of crushed rock products, including precoated chips, all categories of roadstone, concreting aggregate and rail ballast. Hallyards specialises more in concreting aggregate. The freshness of the Hillhouse Sill, in particular the fresh condition of the constituent olivine crystals, results in the rock being especially well suited for dense low-shrinkage' concreting aggregate. The rock has been used in the manufacture of concrete oil platforms. The Craigie Sill is worked at Craigie Hill Quarry [NS 423 328], mainly for concreting aggregate, granular sub-base and armourstone. None of the other sills is worked at present, probably because they are relatively remote.

Dykes

Numerous dykes cross the area and many have formerly supported small quarries for making setts, kerbs and road metal, especially the more extensive, north-west-trending basalt dykes. Many old quarries can be found around Barrmill and Lugton, for example. A north-west-trending basalt dyke is presently worked at Loanhead Quarry [NS 365 554], near Beith; however, the quarry has expanded into the adjacent lavas and the dyke-rock now contributes little to the total output.

Limestone

The Irvine district contains the only major limestone resource in the west of Scotland. Three limestone beds are of economic significance: these are the Broadstone and Dockra limestones in the Lower Limestone Formation and the Upper Linn Limestone in the Upper Limestone Formation. Other limestones may be of local significance.

Lower Limestone Formation

Broadstone Limestone

This not developed throughout the whole of the district, its distribution being influenced by the topography of the underlying lavas (Chapter 7). The Dusk Water Fault forms the eastern limit of its development as a low-dipping bed: to the north-west of the fault it is up to 9 m thick and has a CaCO₃ content of approximately 90 per cent but to the south-east it thins rapidly and is represented by a much reduced sequence of limy mudstones. However, it is present again south of Barrmill in dimensions similar to its north-western occurrence. The thin magnesium-rich Wee Post Limestone is also present north-west of the Dusk Water Fault, but south of the fault it is impersistent.

Dockra Limestone

This limestone occurs stratigraphically 10 to 20 m above the Broadstone Limestone. It may be up to 18 m thick but over a third of this may be calcareous mudstone. The Dockra Limestone has a CaCO₃ content of approximately 90 per cent, and extensive low-dip areas lying at shallow depths occur in the northern part of the district, on both sides of the Dusk Water Fault. Most of the main deposits are truncated by faults, but between the faults extensive low-dip areas with thin drift cover are present. The proximity of Broadstone Limestone at shallow depth may allow both beds to be worked within the same quarry.

Extraction of these limestones was very extensive and they are recorded as having been worked at Auchenmade Quarry [NS 339 485], Auchenskeith Quarry [NS 392 525], Dockra Quarry [NS 364 525], Lugton area [NS 41 52], North and South Biggart [NS 40 53], Middleton [NS 402 526], Hessilhead [NS 378 531], Broadstone Quarry [NS 36 53], Trearne Quarry [NS 37 53], Langside [NS 369 537] and Lyonshield [NS 373 538]. Only Trearne is presently active.

Upper Limestone Formation

Upper Linn Limestone

This limestone occurs in the Dairy area where it is up to 9 m thick, with a CaCO₃ content up to 89 per cent. Although the limestone is of reasonable quality, the thick cover reduces its value. It is unlikely that it would be exploited on any large scale due to the proximity of more extensive and better quality limestones in the Lower Limestone Formation. The Index Limestone, although it. was formerly sought, is not now a viable resource because it is only about 2 in thick.

These two limestones were formerly sought at Monkredding [NS 326 454], Goldcraig [NS 320 499], Clonbeith [NS 341 455], Lylestone [NS 332 466], Lynn Quarry [NS 284 485], Dairy, Loans Quarry [NS 311 486] and Highfield Mine [NS 323 505], Dairy.

Further potential

The limestones of the district are important on a national basis as they represent a resource which could support a cement-making industry similar to that at Dunbar where similar beds in the Lower Limestone Formation are worked. When the Dunbar resources are eventually depleted, the Irvine district may provide an alternative source of both limestone and mudstonc.

Sand and gravel

Largs–Fairlie

In the area north of the Hunterston peninsula there is a coastal strip of raised beach deposits resting on a wave-cut rock platform. The strip is up to 200 m wide in places and is covered mainly with sand and a little gravel. The thickness of the deposits is unlikely to average more than 2 m. They are thicker and more extensive at Largs and west of Kelburn [NS 216 566] where they are intercalated with and overlain by river alluvium. A thickness of 18 m of sand and gravel was recorded in a borehole at Largs.

Hunterston Peninsula

Raised beach deposits form a strip about 150 to 200 m wide at the south end of the peninsula between Portencross [NS 178 488] and Seamill. The deposit consists of fine-and medium-grained sand resting on a wave-cut platform, partly of rock, partly of glacial debris. The sand is unlikely to average more than 2 m in thickness. Similar deposits occur in the north of the peninsula also [NS 20 511.

A flattish area, about 320 ha in extent [NS 19 48] occurs in the southern half of the peninsula. Recent boreholes have shown that the deposits are up to 22 m thick and consist of sand with some small gravel. The pebbles in the gravel consist of quartz and red sandstone. This deposit may be of marine or glacial origin.

Stevenston–Irvine

The deposits of this area consist of blown sand overlying raised marine deposits. The blown sand occurs immediately behind the foreshore, in a zone about 200 m wide in the south of the area but increasing to 2 km near Stevenston. The sand is a fine-grained deposit in the form of dunes and quantities are difficult to estimate.

The raised marine deposits extend inland behind the sand dunes for 1 to 2.5 km. They consist mainly of fine-and medium-grained sand with a little gravel, and are mainly flat bedded with some cross-bedding. The gravel is largely of local origin consisting of pebbles of basalt, sandstone, and some quartz. In places beds of peat occur in the upper part of the deposit, and there are some coal fragments in the sand. The two deposits together cover an area of about 2200 ha, and boreholes indicate that over most of the area their combined thickness is greater than 10 m. Gravel forms a relatively small proportion of the total. Large parts of the deposit are sterilised by the built-up areas of Irvine, Stevenston and Ardeer.

There are two former sand pits [NS 281 415], [NS 296 387] in the blown sand in the Ardeer area and one [NS 334 364] in the raised beach deposits south of Irvine. The Ardeer pits supplied sand for glass-making and foundry work.

Mudstone for brickmaking

The main material used by the brickmaking industry in Scotland is mudstone of Carboniferous age. The mudstone may be extracted directly or may be obtained as a byproduct of other extraction such as coal mining. In the Irvine district mudstone for brickmaking has been obtained both by opencast methods and by reworking of tips on the site of former collieries or ironstone mines. Most parts of the stratigraphical column can produce mudstone which could be used for brickmaking but, if criteria such as thickness and quality are applied, then target areas are very limited.

Ballagan Formation

South of the Inchgotrick Fault, the Ballagan Formation consists of dark grey blocky mudstone with beds of cementstone and local sandstone beds. There is no evidence that these strata in the Irvine district have been used for brickmaking, but the so-called marls (massive mudstones) and marly detrital beds may be potential sources of material for facing bricks. However, localities where there are concretionary limestone nodules or beds of cementstone in the mudstones may not be suitable.

Lower Limestone Formation

Beds of fissile mudstone occur above many of the limestones but tend to be thin and calcareous. Among the Hosie limestones the main rock type is fissile mudstone in three beds each about 2 to 7 m thick with thinner beds of limy mudstone.

Limestone Coal Formation

The Kilbirnie Mudstone Member in the Dairy area includes thick sequences of fissile mudstone with ironstone ribs. Individual beds of mudstone range from 3 to 5 m thick north of Lugton to 14.6 m near Kilbirnie. There is a thicker sequence (22 m) of mudstone from the Dalry Clayband Ironstone upwards and the sequence incorporates, at higher levels, the Johnstone Shell Bed and a higher ironstone. The total thickness of mudstone can be as much as 36 in.

In the higher part of the Limestone Coal Formation, mudstones about 6 m thick, some with thin ironstone ribs, are present above the Dairy Blackband Ironstone. Logan's Bands, a mass of fissile mudstone containing thin clayband ironstones, occurs at a higher level. Thick mudstones occur at these horizons at Kilbirnie, Glengarnock, Caddell [NS 262 472], Girthill [NS 260 468] and Monkredding [NS 326 454]. However, in the Caddell–Girthill area there are no mudstones above the Dalry Blackband Ironstone, and at Monkredding the mudstone sequences are thinner. Fissile mudstones are thinner and less common in the upper part of the Limestone Coal Formation and are unlikely to be economically worked except with coals in opencast operations.

Potentially this is a highly promising area for resources of mudstone for brickmaking, since the mudstones are present in considerable thickness and it is known that bricks have been made in the past from the waste mudstone of ironstone and coal workings. North-east of Dalry, old mudstone tips (waste from ironstone workings) supplied the Kersland and Carsehead brickworks. The bricks made were largely used for outside work. Brickworks between Lugton and Caldwell Railway Station at one time used waste mudstones from an old pit which worked the Dairy Clayband Ironstone.

Upper Limestone Formation

The Upper Limestone Formation is most fully developed in the Dairy area, north-west of the Dusk Water Fault. The Index Limestone is usually overlain by calcareous fissile mudstone with thin beds of limestone. Higher in the succession a sequence of mudstones encloses a thin coal and the Third Post Limestone, while the Lower Linn Limestone is overlain by a thick mass of siltstone and mudstone. Still higher the Upper Linn Limestone is underlain by a bed of fossiliferous mudstone. The calcareous nature of most of the mudstones probably renders them unsuitable for brick manufacture though they might he suitable for the production of bloated lightweight aggregate.

South of the Dusk Water Fault only the Index Limestone is identifiable and it may be overlain by mudstone or by sandstone. In most of this area there are no thick sequences of mudstone.

Coal Measures

Beds of mudstone are common in some areas. A thick pale mudstone in the roof of the Five-Quarter Coal appears to be a widespread feature; at Auchenharvic Colliery it is 5.5 m thick. No Coal Measures mudstones have been quarried directly for brickmaking but waste material from the collieries has been used in some places.

Refractory clays

A range of clays suitable for the manufacture of refractories exists in the Irvine district. All are found as thin beds in the Carboniferous sequence. Lower-grade fireclays occur as seatearths in the Coal Measures, and are also known in the Lower Limestone and Limestone Coal formations. Higher-grade materials are represented in the Passage Formation: high-alumina fireclays occur in the basal sandstone-dominated member and bauxitic clays in the Ayrshire Bauxitic Clay Member. Both of these developed by lateritic weathering, the former of terrigenous sediments, the latter of basaltic lavas.

Fireclay

Lower Limestone Formation

In the northern part of the district the succession between the Broadstone and Dockra limestones, exposed east of Glengarnock, contains several thin seatclays individually up to about 1 m thick. Near Lugton Water a bed of seat-clay and mudstone about 1.6 in thick has been recorded at about the horizon of the Wee Post Limestone. A fire-clay of the same thickness was exposed over calcareous mudstone above the Broadstone limestone at Auchenskeith [NS 392 525]. The fireclay was very plastic but because of its poor refractory quality was not worked. Near Byrehill [NS 423 441], south of Stewarton, a small trial was sunk to the fireclay but it was ferruginous and considered valueless. At Low Gainford [NS 445 440], east of Byrehill, a pure white clay or mudstone (1.5 m thick) rests on what was thought to be the Broadstone Limestone. At the top of the succession are the Hosie limestones, with the Hosie Fireclay, 1.3 m thick, near the middle.

Limestone Coal Formation

A single example is known, at Lylestone Quarry [NS 332 466] where the ganister-like fireclay underlying the Highfield (Index) Limestone was worked as a refractory.

Coal Measures

Various fireclays have been mined in the district, usually in combination with coal. In Saltcoats, a bore in Warner's Clay Quarry near Broom [NS 282 421] encountered some 6.4 m of 'fireclay' overlying the Raise Coal at the base of the Coal Measures. At Reden Pit near Stevenston the Lower Little Parrot or Lower Wee Coal is said to have been worked mainly because of an associated fireclay. At Auchenharvie [NS 257 416] the Wee Coal and Davis or Ladyha' Coal are each underlain by almost a metre of fireclay, both said to he of good quality and used for firebricks. At Cunninghamhead [NS 377 418] almost 1 m of fireclay underlies the Ladyha' Coal. The fireclay beneath the Ell Coal was worked with the coal near Law Farm south of Perceton [NS 340 405], and the same fireclay was worked at Caprington Colliery [NS 40 36]. Fireclays above and below the Ladyha' Coal were worked at Ladyha' and other collieries. The fireclay (0.15 to 0.45 m) below the Wee Coal was worked in pits north of Dreghorn and was used in the Bourtreehill Fireclay Works. The fireclay was also important in the Perceton pits.

High-alumina fireclay

A high-quality refractory clay occurs in the basal member of the Passage Formation in the area south-west of Dairy. The deposit occurs in two leaves, the upper Monkcastle Douglas Fireclay (1.2 to 2.4 m-thick) and the lower Douglas Pink Fireclay (0.9 to 2.7 m-thick), separated by about 4.3 m of strata. Despite the thickness of the deposit, its lateral extent is limited to an area north of the Dusk Water Fault.

These fireclays have been mined until recently at Monkcastle and analysis indicates alumina contents of 40 to 42 per cent (calcined) (Highley, 1982), a range very similar to that of the bauxitic clays. The Monkcastle fireclays were used locally in the production of high quality refractory bricks.

Bauxitic clay

Ayrshire Bauxitic Clay Member

This member crops out along the margin of the Westphalian (Coal Measures) coalfield of Ayrshire. It is best developed along the northern side, extending from Saltcoats [NS 24 41] on the coast eastward to Cunningham-head [NS 38 41], with only small isolated outcrops elsewhere. It rests on basaltic lavas of the Troon Volcanic Member.

In the High Smithstone area [NS 283 457], north of the Dusk Water Fault, interbedded bauxitic clay and coal form sequences up to around 20 m thick, while at Dubbs [NS 422 277], south of the Dusk Water Fault, a single bauxitic clay unit is present, around 4 m thick. The upward transition from basaltic lava to bauxitic clay is never sharp and is commonly marked by a transitional clayrock lithology rich in sphaerosiderite. South of the Dusk Water Fault, interbedding with other lithologies occurs locally, as at Annick Water where a black laminated mudstone with plant remains and a seatclay occur within the bauxitic clay.

The bauxitic clay is noted for the occurrence of boehmite and diaspore, although these minerals have not been detected anywhere other than at the original localities examined by De Lapparent. (1936) at Saltcoats and at Dubbs.

The northern outcrop of the Ayrshire Bauxitic Clay from Saltcoats eastward to Kilmarnock contains some of the highest quality fireclay in the UK, with 42 per cent Al₂O₃ (uncalcined). The quality of the deposit is unpredictably variable laterally, with the iron content fluctuating considerably. The deposit has been worked in the past both by opencast pits and in drift mines. Despite the high quality of the material and its association with thin coal seams which were often also extracted, working of bauxitic clay has declined dramatically in the last decade. At present opencast extraction is restricted to the high Smithstone Basin, where the bauxitic clay is extracted with coal and is mainly exported to England to upgrade lower quality fireclays.

The Ayrshire Bauxitic Clay has been examined as a potential source of titanium dioxide (TiO₂) (Cameron, 1980), a substance which is used mainly as a pigment in paints and plastics. Although the bauxitic clay locally contains as much as 14 per cent TiO₂ (average 4.7 per cent), the mineral is not present in sufficient concentration, and is too finely divided, to attract any attention at present.

Future potential

In the Irvine district, the area with the greatest potential for exploitation of refractory clays is the northern outcrop of the Ayrshire Bauxitic Clay, especially in the High Smithstone area. The material is locally of very high quality and of refractory grade. In addition to being used for the manufacture of refractory products, it has also been used as a chemical feedstock in the production of aluminium sulphate. The high-alumina fireclays around the old Monkcastle Mine at Dairy also have potential for further exploitation, given a sufficiently high price for the raw material.

Sandstone

Sandstone as a building stone is becoming a more important commodity, though it still is not used as extensively as in former years. The well-known red sandstones of Ayrshire were won either from the Devonian Stratheden Group of the Irvine district or from the Permian Mauch-line Sandstone of areas to the south. White or buff coloured sandstones occur within the Carboniferous sequence and these were also used formerly for building. Not all the stratigraphical units in the Irvine district contain sandstones but those that do are described below. Those sandstones which have been worked in the past, either for building stone or for millstones, are likely to be the best prospects for future extraction.

Kinnesswood Formation and Ballagan Formation

These formations contain a mixture of sandstones, siltstones and mudstones, commonly with thin beds of dolomitic limestone (cementstone) or concretionary limestone (cornstone). Sandstones are most common in the Kinnesswood Formation but there is a variation in colour between sandstones and the colour at any one place tends to be inconsistent. Sandstones are not abundant in the Ballagan Formation but some have been worked in the past, mostly in the districts to the east and south. In the Irvine district the formation is usually at too great a depth to be a workable resource.

Limestone Coal Formation

This formation contains a mixture of sandstones, siltstones and mudstones with coals and ironstones. Channel sandstones of fluvial origin tend to be clean and well sorted but may be coarser grained towards the base. This group of rocks offers considerable potential for building stone.

Upper Limestone Formation

This formation comprises sandstones, siltstones, mudstones and limestones. The sandstones are generally medium to fine grained and are usually light grey to white. Massive channel sandstone units occur, and this formation may yield sandstones suitable for building stone.

Passage Formation

The sandstone-dominated basal member of this formation consists mostly of channel sandstones. The character of this member changes laterally, the proportion of seat-clay and seatearth increasing southwards, and potential resources are effectively restricted to the region north of the Dusk Water Fault.

Lower and Middle Coal Measures

These measures comprise sandstones, siltstones and mudstones with coals. Any sandstone resources are likely to relate to locally developed channel sandstones.

Upper Coal Measures

These measures were deposited in an environment very similar to that of the Lower and Middle Coal Measures but a general process of reddening has occurred due to subsequent oxidation. The reddening is often patchy and unpredictable, giving the sandstones within this unit a poor resource potential.

Silica sand

Sandstones above the Index Limestone, roughly equivalent to the Bishopbriggs–Cadgers Loan Sandstone sequence of Glasgow, have supplied silica sand. At Seven-acres Quarry [NS 328 451] approximately 10 m of white sandstone is worked to produce washed moulding sand and coloured chaff white') glass sand with an iron content of about 0.25 per cent. The adjacent Monkredding Quarry [NS 324 451] is at about the same horizon, and has produced moulding sands. It has been suggested that the sandstone farther west, at the south-west end of Ashgrove Loch, may also provide silica sand (Boswell, 1918). The sandstone in Lochcraigs Quarry [NS 2746 4405] is generally white, medium to coarse grained and friable, producing a sand with about 98 per cent Si02 and 0.02 per cent Fe205.

Silica sand also occurs in the superficial deposits. Dune sands in the Irvine area (see also sand and gravel, above) have been worked to supply locally required moulding sand and coloured (amber) glass sands. The sands form dunes up to 20 m high and, although composed mainly of quartz, contain shelly and coaly fragments. The sands have a reddish appearance due to iron staining, and are known to contain about 1.5 per cent iron oxide. Most of the workings are now disused, and many of the deposits are sterilised by the development of Irvine and Stevenston. The only silica sand now produced is from the Ardeer Pit and is used for washed moulding sand. Shewalton Quarry south of Irvine currently produces constructional sands, mainly from the raised beach deposits below the dunes. The dune sands do offer a local source of relatively impure silica sand for washed moulding sands and possibly for filtration beds.

Ironstone

The ironstones of the district were formerly important as the foundation for much of the heavy industry that was present in the area. From 1860 onwards the iron-making industry relied increasingly on imported ore, however, and the use of local ironstones declined. The ironstones are described in detail by Macgregor et al. (1920).

Two types of ironstone deposit occur; clayband ironstones and blackband ironstones. Clayband ironstone is a rock consisting of a mixture of clay minerals and siderite (iron carbonate). A blackband ironstone has, in addition, sufficient carbonaceous material to allow the ironstone to be self-calcining. Current interest in ironstones relates to the potential hazard posed by the old shallow workings.

The principal ironstones in the district occur within the lower part of the Limestone Coal Formation. They are common either as continuous beds up to about 0.5 m thick or as nodules within argillaceous sequences. The more prominent beds of ironstone were named, most important being the Dalry Clayband Ironstone and the Dairy Blackband Ironstone.

Dalry Clayband Ironstone

This deposit was an important resource, particularly in the Dalry and Glengarnock areas. It is the lowest workable ironstone in the sequence, occurring about 6 to 10 m above the Top Hosie Limestone. It varies in thickness from 0.15 to 0.50 m and has been worked extensively from Kilbirnie Loch south to Blair House. In the Lugton area it is about 0.35 m thick and was worked in shallow pits. Thicknesses of up to 0.50 in are common in the Kilbirnie area where 35.4 and 38.1 per cent Fe are recorded in calcined samples.

Pundeavon Ironstone

This deposit is poorly exposed in the Paduff Burn but in the Pundeavon Burn is 0.23 m thick and shows 'cone-incone' structure. It is commonly calcareous and even at the height of ironstone extraction was considered of little economic significance.

Garibaldi Ironstone

This deposit is a series of thin beds of clayband ironstone separated by mudstone. They occur between 4 and 10 m below the Dalry Blackband Ironstone and were worked at Glengarnock until 1914.

Dalry Blackband Ironstone

This deposit was formerly worked extensively in the Dairy area. In some parts of the ironstone field the bed is entirely replaced by volcanic ash. Cones of fine-grained, ferruginous ash appear to have been built up contemporaneously with deposition of the ironstone. The ironstone itself is of variable thickness; at Lylestone, for example, it varies from zero to 0.12 in, with a coking coal of between 0.15 and 0.30 m thick.

Logan's Bands

These bands consist of a series of clayband ironstones with intervening mudstone. The ironstones usually total between 0.30 and 0.48 m and are calcareous in places. Logan's Bands were worked to a minor extent at Laigh Dykes, Ardrossan and at Ryesholm Colliery, just north of Dairy.

Mineral resources:  metalliferous minerals

Vein mineralisation due to hydrothermal activity is believed to have been associated with all three periods of volcanic activity in the Irvine district, the Visean Clyde Plateau Volcanic Formation, the late Namurian Troon Volcanic Member and the early Permian Mauchline Volcanic Formation. Many of the veins are associated with west-north-west-trending faults and with late Carboniferous dykes. It has been suggested that some of the baryte mineralisation in the district to the north may be Palaeogene in age because of its association with the Palaeogene dyke swarm. However, radiometric (KAI) determinations from vein gouge clays at Muirshiel give ages ranging from 240 to 213 ± 3 Ma (Moore, 1979). This suggests that the main phase of mineralisation may have occurred during the Triassic period, at about the same time as baryte was emplaced at Strontian, in Glen Sannox and in northern England. Paterson et al. (1990) suggested that. the mineral veins and dykes were injected at different times into a fracture system which may date from the late Carboniferous.

Baryte

Baryte is associated with the Misty Law Trachytic Centre in the district to the north (Paterson et al., 1990). Muirshiel Mine [NS 282 649] was one of three haryte mines in southern Scotland (Muirshiel, Gasswater and Glen Sannox) which collectively accounted for about one third of UK annual production between 1946 and 1966. In the Irvine district veins of baryte are commonly found in the vicinity of the north-west-trending Palaeogene dyke swarm in the north of the district. The distribution of baryte veins and their relationship to the Palaeogene dykes have been discussed by MacGregor (1944), Scott (1967) and Gallagher (1968), and summarised by Stephenson and Coats (1983).

Three localities illustrate the way in which baryte occurs in the district. East of the Pundeavon Reservoir [NS 2916 5786] baryte occurs in a disseminated form within the fault gouge of a west-north-west-trending fault. Discrete veining is found in another fault of similar orientation in the Millside Burn [NS 3091 5850]. Richey et al. (1930) recorded a 0.3 m vein of baryte in the limestone quarry at Crawfield [NS 332 526].

Copper

Copper was worked from a vein at the Swinlees Mine [NS 2882 5277] around 1830 but by the time of the first geological survey (around 1870) the mine was abandoned (Stephenson and Coats, 1983). Two levels were driven along the vein but only the entrance to the upper level remains visible. Flow-banded rhyolite of the Clyde Plateau Volcanic Formation is in fault contact with the Dockra Limestone (Lower Limestone Formation) and the mineralised vein cuts both. The vein is recorded as being 60 to 100 cm wide and trends north-west. Along the faulted contact between the Dockra Limestone and the rhyolite, the fault breccia consists of limestone and rhyolite cemented by baryte with calcite, quartz, malachite, azurite and chalcopyrite.

Copper mineralisation is also recorded from Loanhead Quarry [NS 365 554] where Gribble (1992) noted the occurrence of native copper on joint surfaces along with malachite and the oxide, cuprite.

Lead

Richey et al. (1930) recorded that a pocket of galena comprising 'a few tons' was extracted during working of the limestone at Dockra Quarry. This is the only record of significant amounts of lead ore being found within the district.

Water resources

The district lies mostly within the catchment of the rivers Garnock and Irvine, and their major tributaries the Lugton Water and the Annick Water. The coastal areas in the extreme north and south of the district drain directly to the sea. Mean annual rainfall varies along the coast from only 850 mm at Troon to 1200 mm north of Hunterston. Rainfall increases inland particularly over the high ground in the north of the district which receives over 1600 mm. Potential evapotranspiration is generally less than 400 mm a-1; the ample difference between rainfall and evapotranspiration supports surface run-off as well as infiltration to groundwater reserves.

None of the strata in the district offers groundwater development potential of regional significance. However, locally important supplies are available particularly from the Coal Measures whilst valuable rural domestic supplies are obtained from other formations.

The dominant groundwater flow system parallels the surface water drainage; groundwater levels are greatest beneath the higher ground to the north, south and east, with flow towards the sea in the vicinity of Irvine (Robins, 1990). Much of the groundwater emerges as baseflow in the main watercourses. For the most part groundwater flow is shallow and essentially local, with flow paths of only a few kilometres length.

Transport of groundwater depends to a large extent on the presence of open cracks and fissures within the rock which provide the strata with secondary permeability usually far in excess of the intergranular permeability. Sustainable borehole yields are nevertheless modest except where boreholes actually intersect abandoned and flooded mine workings in the Limestone Coal Formation and the Coal Measures. A further limitation on groundwater development is water quality; groundwater, particularly in the Carboniferous strata, may be reduced in oxygen with high concentrations of bicarbonate, sulphate and iron in solution.

Devonian rocks

There are few records of any boreholes or wells in the Devonian rocks north of Ardrossan. Shallow groundwater flow occurs in some weathered zones and small supplies (< 0.3 l s⁻¹) have been obtained from boreholes 26 and 32 in deep at West Kilhride [NS 1829 4884].

An exploratory borehole at Hunterston penetrated 120 m of pebbly sandstone and conglomerate. Test pumped at 1.3 l s⁻¹ the specific capacity after three days of pumping was only 0.02 l s⁻¹ m⁻¹; groundwater flow is locally inhibited by the presence of dykes. In 1974 a programme of test drilling was carried out in the southern half of the Hunterston peninsula to evaluate the feasibility of storing oil in underground caverns (Price et al., 1974). The strata comprise compact sandstone with subordinate siltstone and conglomerate beneath a thick cover of till. It was found that groundwater flow was restricted to fissures in the uppermost 70 m. The Devonian rocks had a mean hydraulic conductivity of 0.5 m d⁻¹ and a storativity of between 10⁻⁵ and 10⁻⁴. The oil storage option was not pursued.

Strathclyde Group

The Clyde Plateau Volcanic Formation contains some groundwater which circulates in shallow cracks and fissures. However, groundwater flow paths are essentially local, and some sources, particularly springs, are ephemeral. Small domestic supplies from boreholes 14 to 82 in deep have been obtained around Dunlop and at Montgreenan. Boreholes penetrating the associated sandstones are more productive, for example 3 l s⁻¹, with a specific capacity of 0.2 l s⁻¹ m⁻¹ from a 30 m-deep borehole at Uplawmoor [NS 4017 5464]. The groundwater piezometric surface tends to occur at or near surface.

Clackmannan Group

The diverse lithologies of the Lower Limestone and Limestone Coal formations, offer poor prospects for groundwater flow. Sustainable borehole yields of 0.2 l s⁻¹ are recorded for a 45 in-deep borehole near Kilmaurs [NS 3898 4270] and also for a 31 m-deep borehole at Saltcoats [NS 2408 4212], the supply from the latter probably being augmented by the overlying Quaternary gravels. Various smaller supplies have been used over the years but few remain in service because of indifferent groundwater quality; the water is depleted in oxygen and is iron-rich with an alkalinity greater than 200 mg 1⁻¹.

The Lochwood No 2 Mine between Dalry and Stevenston used to be pumped for 10 hours in the day at 17 l s⁻¹ (average 7 l s⁻¹) in order to keep it dry (Table 1). This high yield from deep in the Limestone Coal Formation reflects the comparatively large open contact area of a mine to the aquifer compared with the very limited open area of a borehole.

Rocks of the Upper Limestone Formation offer more favourable hydraulic properties than older Carboniferous formations. Sustainable yields from boreholes in the Kilmaurs area [NS 4080 4071], [NS 3903 4162], [NS 4001 4244] attain 0.5 1st from boreholes ranging from 40 to 135 m deep. Groundwater quality is variable, usually moderately mineralised with alkalinity up to 250 mg 1⁻¹.

Few water boreholes penetrate the sedimentary members of the Passage Formation and none is recorded in the Troon Volcanic Member. However, former mine dewatering rates are instructive (Table 1); the Monkcastle Fireclay Mine and the Pict Mine were notably dry whereas the Dubbs Mine and Montgreenan yielded large quantities of water from relatively shallow workings.

Coal Measures

Boreholes in the Coal Measures are generally poor, yielding, for example, 0.1 l s⁻¹ from a 36 m-deep borehole at Cunninghamhead [ 3764 4189] and less from a 41 m-deep borehole at Kilwinning [NS 3369 4426]. Exceptions are Kilmaurs Creamery [NS 4981 4064], which sustains a yield of 4 l s⁻¹ with a specific capacity of 0.8 l s⁻¹ m⁻¹ from a 101 m borehole, and ICI at Stevenston with a yield of 40 l s⁻¹ and a specific capacity of 1.7 l s⁻¹ m⁻¹ from a 165 m-deep borehole which actually intersects flooded mine workings. The rest water level in the ICI borehole rises and falls by 3 m due to tidal influence. At Irvine [NS 350 365] a group of shallow boreholes penetrating mine workings was tested at yields up to 12 l s⁻¹. Long-term dewatering of the workings precluded the use of the boreholes, due to the probability that ground settlement would affect established industrial plants.

Mine dewatering records for the Coal Measures are listed in (Table 1). Although substantial volumes of water were often involved, the rates are modest compared with sonic mines in the Central and Midlothian coalfields. For the most part water quality from the Coal Measures is poor, often heavily mineralised, iron-rich and reducing (Table 2).

Quaternary deposits

Some groundwater is available from the drift, though none has apparently been exploited from alluvium. The raised marine deposits and blown sand centred on Irvine tend to be elevated and therefore well drained, except where peat horizons act to constrain groundwater flow.

However, a group of shallow boreholes [NS 350 365] near Irvine penetrated 9 m of sand, silt and gravel; the water table lay at 0.5 m below ground level. Borehole yields could only be sustained at between 1 and 3 1st because poor quality water from the underlying Coal Measures was drawn into the boreholes as pumping progressed. The range of specific capacity for the drift was 1.4 to 2.0 l s⁻¹ m⁻¹.

Another investigation for water supply was carried out for a new paper mill near Irvine [NS 335 353]. Here 3 to 11 m of medium-grained sand with thin peat horizons and gravel and cobble lenses overlie silt and clay (Robins and Ball, 1991). Bedrock is of Coal Measures. Substantial groundwater potential, although insufficient to supply the mill, was proven during construction. Long-term exploitation could not, however, be pursued because of the risk of land subsidence should the peat horizons become dewatered.

Two adjacent boreholes at Dreghorn each penetrated 7.5 m of sand and gravel saturated from 2.5 m below ground surface. The bores maintained a long-term yield of 1 l s⁻¹ and had a specific capacity of 2 l s⁻¹ m⁻¹. During 1982 the boreholes were used to evaluate the energy available in shallow groundwater for space heating. The groundwater was circulated through a heat pump where it was reduced from 10.5°C to 6.7°C to provide 3.2 kW mean power. However, the economics of conventional gas-fired heating were then marginally more attractive, especially in view of the initial capital outlay.

Groundwater from the drift is typically weakly mineralised, except where upwelling of deep groundwater is induced by shallow pumping (Table 2). Analysis of the groundwater drawn from the Dreghorn experimental boreholes is more typical of the drift: it showed a calcium-sulphate-type water with an alkalinity of only 38 mg

Landfill

The legacy of mining throughout the district has provided considerable scope for land reclamation through landfill disposal of domestic and industrial wastes. Potential drainage of landfill leachate to groundwater has not in the past been perceived as a problem because bedrock groundwater quality is generally poor, particularly in the Coal Measures. This perception has now changed, and increasing use is being made of landfill liners and engineered containment so that leachate can be drawn off and treated.

Made ground

Made ground is a feature of human activity. In the Irvine district, early human occupation left traces near Ardrossan Railway Station and on the south-west side of the Cannon (Castle) Hill [NS 230 422] in the form of a manmade mound of discarded molluscan and vertebrate remains overlying Flandrian raised coastal deposits (Smith, 1893, pp.355–366). Interestingly, amongst the diet of fish, land mammals and a diversity of molluscs revealed by these remains were abundant shells of the gastropod 'Trochus lineatus' [= Monodonta lineata], a species which is, as Smith (1893, p.357) observed, no longer found in the Clyde estuary. A more gruesome aspect of the mound was the evidence of cannibalism which could be deduced from the nature of the human remains there. Between Irvine and Shewalton Moor, Ordnance Survey maps show various sites from which Mesolithic implements have been recovered.

The most obvious examples of made ground in the district are related to industrial developments of the last two centuries. In many cases the characteristics of the material used to create the made ground is impossible to ascertain, especially in old quarries which have been filled with a variety of domestic or industrial refuse. Mined materials are easier to categorise as they depend on the character of the material that was worked. In the Dalry and Kilbirnie areas, coal and ironstone were extracted together and small bings are commonly associated with the sites of old mine shafts. Usually the waste is mudstone and these old bings have been used as a source of clay for brickmaking. The extraction of coal from the Coal Measures similarly produced waste which was usually a mixture of mudstone, siltstone, sandstone and seatclay. These formed the rather larger and more heterogeneous bings that used to be common around Kilwinning and Irvine. Most of these have now been removed or landscaped but thin veneers of made ground remain.

Made ground also marks the site of the former steelworks at Glengarnock and that of the explosives and chemicals factory at Ardeer. Made ground at Ardrossan and Hunterston has been produced by the reclamation of the intertidal zone.

Chapter 3 Lower Devonian

Rocks of early Devonian age occur on the south-west of the Hunterston peninsula in a small, isolated, fault-bounded area at Farland Head. They are associated with the Largs–Hunterston Fault Zone, a major structure which extends from Farland Head northwards along the coast to Largs (Figure 29). At Farland Head a series of parallel faults defines a zone of steeply dipping, commonly vertical strata about 200 m wide. The Lower Devonian strata occur in a wedge bounded to the east by the main fault and to the west by a splay fault.

This area includes two lithological units, one of highly contorted siltstones, silty mudstones and sandstones, the Sandy's Creek Beds, and the other of brown sandstones and conglomerates, the Portencross Beds. Both are fault-bounded. These rocks have been studied by Moore (1859), Bell (1885), Gunn et al. (1903) and Patterson (1949). On the basis of lithological comparisons, Patterson suggested that the Portencross Beds be assigned to the Lower Old Red Sandstone, and thereby implied an early Devonian age. The Sandy's Creek Beds are more problematical; the faulting relationships suggest that they are older than the Portencross Beds and Patterson suggested that they might be Lower Palaeozoic in age. Downie and Lister (1969) obtained poorly preserved miospores from the Sandy's Creek Beds and, although the evidence was inconclusive, they suggested a possible late Downtonian to Dittonian age. The two units are now regarded as formations, and the names Sandy's Creek Formation and Portencross Formation are therefore used here.

Sandy's Creek Formation

The Sandy's Creek Formation is a small but highly anomalous sequence of sedimentary rocks, first noted by Moore (1859) for its distinctive greenish grey colour and contorted structure. The sequence consists of siltstones, silty mudstones and sandstones, but the relationship between beds is obscured by tectonism. The lithologies have deformed in different ways: the sandstones show brittle deformation whereas the siltstones and mudstones have deformed in a more plastic fashion (Patterson, 1949). Some sandstone units may have been originally lens-shaped in cross-section and these show small-scale cross-bedding. Total thickness exposed is probably about 30 m. A thin, irregular, contorted, carbonated basaltic dyke up to 0.7 m wide has been intruded into the beds.

Downie and Lister (1969) considered that a crude lithostratigraphy could be established within the Sandy's Creek Formation, based on the variations in the charac teristics of the sediments. However, the evidence from the sedimentary facies and the spore assemblages indicates terrestrial deposition in a fluvial environment. In such an environment the expected lateral variation in lithology would be similar to that observed within the known vertical sequence.

The structure of the Sandy's Creek Formation is essentially an anticline associated with sinistral transcurrent movement in the fault zone (Patterson, 1949; Downie and Lister, 1969). The last-named authors have determined that the sheared siltstones were dragged out in the direction of faulting but the bedding orientation was maintained in the sandstone beds. Lenticular carbonate veins have also been deformed by these tectonic processes.

Portencross Formation

The Portencross Formation comprises brown and chocolate-brown, hard, compact sandstones commonly arranged in upward-fining units. Massive, cross-bedded and thinly bedded flaggy types of sandstone are present. Conglomerate is common as irregular patches, lenses and dyke-like bodies. The clasts within the conglomerate are up to cobble size, rarely exceeding 15 cm. Most are of quartzite but smaller proportions of vein quartz, chert, sandstone and lava are also present. Conglomerates are common in the lower part of the exposed sequence but higher in the formation sandstones dominate and conglomeratic units are confined to lenses and lag deposits in cross-bedded sandstones. Some calcareous cementation is present, characterised by honeycomb weathering. The sequence dips to the north and north-east at between 30 and 60°; and the total thickness is probably between 450 and 550 m.

Correlation with adjacent areas

Downie and Lister (1969) drew comparisons between the sequence of Lower Devonian rocks on Arran (Friend et al., 1963) and that at Farland Head. The grey siltstones and sandstones of the Sandy's Creek Formation are comparable to the Basal Quartz Sandstones, while the medium-grained lithic sandstones of the Portencross Formation are of similar lithology to the Torr Breac Sandstones of Glen Rosa (Table 3). Because of the tectonically isolated character of the Lower Devonian outcrops, it is not possible to extend correlation to other parts of Scotland.

Conditions of deposition and palaeogeography

The sedimentological character of these rocks and their spore content indicate that they are of terrestrial origin.

The regional palaeogeography is that of a river flowing south-westward along the axis of what is now the Midland Valley of Scotland. Alluvial fans on the northern margin provided coarse-grained detritus of metamorphic and volcanic origin derived from a Highland terrane. The finer-grained nature of the Sandy's Creek Formation suggests deposition in a low energy environment, possibly a lake or a river floodplain. The Portencross Formation comprises conglomerates and sandstones suggesting deposition in river channel environments.

From a regional perspective, early Devonian sedimentation took place in an elongate, north-east-trending basin, which extended up to, and probably over, the Highland Boundary Fault Zone; the fault itself did not limit sedimentation (Bluck, 1978; Morton, 1979; Mykura, 1991; Armstrong et al., 1985). Braided stream environments dominated, with clasts derived from the north-east along the axis of the basin. A second basin, of similar trend, lay to the south.

Chapter 4  Upper Devonian: Stratheden Group

Rocks of late Devonian age occur on the coastal strip from Largs in the north to Ardrossan in the south, with outcrop also present on Great Cumbrae. The rocks are mostly sandstones, commonly cross-bedded and red in colour, with subsidiary conglomerates. The sequence was previously termed Upper Old Red Sandstone but is now referred to as the Stratheden Group (Paterson and Hall, 1986). The age of these rocks has not been determined locally but comparison with lithologically similar rocks elsewhere in the Midland Valley suggests a Famennian age. The base of the Stratheden Group is not exposed in the Irvine district, but at Hunterston there is a faulted contact with Lower Devonian rocks (Chapter 3). The relationship to the overlying Inverclyde Group, of early Carboniferous age, is discussed in Chapters 5 and 6.

Classification

In the absence of any biota, the sequence is divided on a lithostratigraphical basis. The named units are consistent within structurally defined areas but correlation between these areas is speculative. The nomenclature is summarised in (Table 4). Lateral differences in the sequence are assumed to be due to contemporaneous fault movement. All the units were defined by Paterson and Hall (1986) and used in the adjacent district to the north by Paterson et al. (1990). The formations are distinguished on their lithological facies, and on the different patterns of sedimentation that these imply. The boundaries are probably all diachronous.

Relationship to earlier nomenclature

Bluck (1967; 1978; 1980) recognised two sets of formations in the region, but no definitions of these were published. He (1978, fig.8) subdivided the sequence along the Clyde coast north of the Irvine district into the Wemyss Bay Formation (sandstone) at the base followed by the Skelmorlie Formation (conglomerate) and the Kelly Burn Formation (sandstone). Above are sandstones with cornstones, which he called the Leap Moor Formation; this unit has all the characteristics of the Kinnesswood Formation, part of the Inverclyde Group (Chapter 6). He also (1980, fig.6) proposed a lithostratigraphy for the area of the Hunterston Peninsula and its hinterland. Here a basal conglomerate, the Hunterston Formation, is overlain by the Kilrusken Formation, a sandstone of lacustrine origin. The succeeding Seamill Formation consists of braided stream sandstones and the topmost unit, the Haupland Muir Formation, was deposited in an alluvial coast environment.

The lithostratigraphy proposed by Bluck (1980) for the Hunterston area has been modified here, for several reasons. First, the relationship between the base of the conglomeratic Hunterston Formation and the older Portencross Beds is controversial. Bluck stated 'the younger Hunterston Formation at Portencross rests unconfimnably on Lower Old Red Sandstone'. Mapping suggests that the contact between the two units here is faulted; the presence of an unconformity cannot be proved.

Secondly, Bluck suggested that there are differences in the provenance of the conglomerates of the Hunterston and Skelmorlie formations, with a higher proportion of quartzite in the former. While this is undoubtedly significant in terms of the basin evolution, in lithostratigraphical terms it is probably less meaningful. The Stratheden Group lithostratigraphy used by Paterson et al. (1990) in the adjacent Greenock district, to the north, has therefore been applied in the Irvine district, and the lowest conglomerate is referred to as the Skelmorlie Conglomerate.

Thirdly, it was found during resurvey that the Kilrusken Formation of Bluck (1980) could not be mapped on a regional basis. The lacustrine Kilrusken Formation and the braided stream Seamill Formation together probably equate with the Kelly Burn Sandstone Formation (Table 4). The Haupland Muir Formation was not described by Bluck but he suggested that it was deposited in a coastal alluvial area. It may relate to the interbedded fluvial and acolian deposits of the Fairlie Sandstone Formation.

Skelmorlie Conglomerate

The Skelmorlie Conglomerate is the lowest unit of the Stratheden Group outcropping in the Irvine district, and is faulted against older strata. The sequence, locally called 'Pudding Rock', consists of a boulder conglomerate about 30 m thick occurring on the I Iunterston peninsula [NS 18 50] immediately to the west of the Largs–Hunterston Fault Zone. The clasts are sub- to well-rounded cobbles and boulders averaging less than 30 cm, but locally attaining over 60 cm, in diameter. They are commonly of pale brown quartzite and less commonly of brown sandstone with white quartz pebbles. The matrix is a coarse-grained gritty sandstone with fragments of quartzite, greywacke, and brown and grey mudstone and siltstone. Palaeocurrent flow was to the north-north-east (Bluck, 1980). The dip of the beds is to the north-west, in part influenced by the orientation of the fault which defines the junction with the older Portencross Formation. Many of the quartzite clasts are fractured; the fracturing post-dates deposition and may be associated with the development of the Largs–Hunterston Fault Zone.

Bluck (1980) noted that the palaeocurrent directions at Hunterston are similar to those in the Skelmorlie Conglomerate at Wemyss Bay but suggested that the Hunterston conglomerate was formed in a different depositional basin, with different provenance. The stratigraphical relationship between the two conglomerates is problematical, as discussed above.

Kelly Burn Sandstone Formation

The Kelly Burn Sandstone Formation crops out only to the west of the Largs–Hunterston Fault Zone, on the north-west part [NS 18 51] of the Hunterston peninsula, at Largs [NS 20 59] and on Great Cumbrae. The formation comprises a series of red and reddish brown medium-grained sandstones with pebbly sandstone and pebble conglomerate in lenses up to 1 m thick. The clasts are commonly subangular white vein quartz with subrounded quartzite and greywacke. The subangular vein quartz may represent redeposited clasts which were fractured by jointing or faulting in a pre-existing conglomerate, so their subangular character is not necessarily a function of immaturity. The beds show trough cross-bedding and are arranged in crudely upward-fining units. At Hunterston the palaeocurrent flow was to the north-east (Bluck, 1980) and the formation is probably over 400 m thick. It is very much thicker on Great Cumbrae, and in the Greenock district the thickness is estimated to be about 1500 m (Paterson et at, 1990).

Seamill Sandstone Formation

The Seamill Sandstone Formation crops out to the east of the Largs–Hunterston Fault Zone and is best seen on the coast section from Farland Head [NS 177 484] on the Hunterston peninsula south to near Ardrossan. It comprises a series of reddish brown, medium- to fine-grained sandstones occurring as generally upward-fining units. Thin beds and lenses containing pebbles mostly of white vein quartz form the base of each unit (Plate 1). The vein quartz clasts are generally angular, fracturing along mineral boundaries, but some well-rounded clasts are also present. As discussed above, the angularity of these clasts is not necessarily a sign of immaturity. Brown mud flakes occur as rip-up clasts, although discrete beds of mudstone are rare. The formation is made up of packets of sedimentary rock usually 2 to 3 m thick, massive and poorly bedded at the base but becoming well bedded upwards with the development of trough cross-bedding (Plate 2). The cross-bedding indicates sediment derivation from the south-west, in accord with data from Bluck (1980). Local minor disconformities within the sequence are common. Slump structures are present, commonly in proximity to the Largs–Hunterston Fault Zone, and may have resulted from tectonic instability along that line.

Fairlie Sandstone Formation

The Fairlie Sandstone Formation crops out in the area to the north of Fairlie Glen, stratigraphically above the Seamill Sandstone Formation and immediately below the Clyde Plateau Volcanic Formation. It is best seen in Kelburn Glen [NS 216 567], the Keppen Burn [NS 214 557] and in Fairlie Glen [NS 217 547]. It consists of white or cream coloured sandstones with beds of vein-quartz conglomerate. The formation is made up of sedimentary packets of varying thickness but usually around 2 to 3 m thick. These units fine upwards from a basal, rather massive conglomeratic sandstone to a medium-grained sandstone with cross-bedding and finally to a fine-grained and well-laminated sandstone. The topmost part is well-sorted within single laminae and occasional well-rounded grains can be seen, usually in coarser laminae. The laminae are flat lying and are similar to those illustrated from the Knox Pulpit Formation of central Fife (Chisholm and Dean, 1974). The latter is now considered to be aeolian in origin (Hall and Chisholm, 1987). These authors noted the co-existence of aeolian and fluvial environments in several parts of the Scottish Midland Valley and used the criteria of the characteristic fine lamination, well-rounded grains and absence of micaceous laminae or larger clasts to distinguish strata of aeolian origin from those of fluvial origin. These features can also be identified in the Fairlie Sandstone Formation and an origin by alternation of fluvial and aeolian processes seems probable. Grain-size analyses of the aeolian component of the Fairlie Sandstone Formation ((Figure 4), discussed further below) show a clear similarity to those of the undoubtedly aeolian Penrith Sandstone of the Vale of Eden.

The southward extension of this formation is enigmatic. It can be traced south to the Fairlie Glen, where it is about 200 m thick, but in the outcrops south of here, the Fairlie Sandstone Formation is missing and the Clyde Plateau Volcanic Formation rests directly on Seamill Sandstone Formation. The Fairlie Sandstone Formation is either not developed or else is no longer preserved south of the Fairlie Glen Fault. The latter is the preferred solution as the loss of 200 m of the Fairlie Sandstone lithofacies in less than a kilometre seems unlikely. The Fairlie Glen Fault also corresponds to the northern margin of a significant negative gravity anomaly (Chapter 13).

Conditions of deposition and palaeogeography

The changing environments of deposition during the late Devonian are reflected in the changing average grain size through the Stratheden Group, from conglomerates at the base to sandstones at the top.

The Skelmorlie Conglomerate at Hunterston is considered by Bluck (1980) to be an alluvial fan deposit. The high degree of sphericity shown by the clasts would further suggest that this conglomerate was reworked from a previous one. The distribution of the Stratheden Group conglomerates in the west part of the Midland Valley appears to have been controlled by faulting; the regional work of Bluck (1980) suggests that basin development at this time was related to sinistral movement along the main bounding fault, the Highland Boundary Fault, with extensional half-graben structures generated by normal accommodation faults trending south-east. In the context of the Irvine district, the outcrop of the Skelmorlie Conglomerate is about 10 km south-east from the Highland Boundary Fault, which transects the northern part of Arran. The Largs–Hunterston Fault Zone is probably a splay from the major structure and lies close to the outcrop of the Skelmorlie Conglomerate, but exposure is insufficiently extensive to demonstrate a conclusive relationship between movement on the fault and deposition of the conglomerate.

The Kelly Burn Sandstone Formation and Seamill Sandstone Formation, which are probably lateral equivalents, show similar sedimentary characteristics, including the absence of fine-grained overbank deposits and the presence of trough cross-bedding with few significant higher-order erosive surfaces. These features are interpreted as indicating deposition in a braided stream environment on top of the alluvial fan. Bluck (1980) suggested that some of these deposits are of lacustrine origin but, while ephemeral lakes were probably part of the alluvial environment, finer-grained sandstones interpreted as lacustrine deposits seem to be preserved only locally.

On a regional scale palaeocurrent directions indicate flow from the south-west, a direction corresponding to the main axis of the depositional basin (Figure 3). The derivation of the vein quartz pebbles which are characteristic of these formations is problematical. The virtual monomirteralic character of the clasts suggests that they may be the remnants of older clasts that were derived originally from the Highland metamorphic area but which have been subject to repeated cycles of erosion, fracturing and deposition.

The Fairlie Sandstone Formation is a further development of the depositional environments of the Seamill and Kelly Burn Sandstone formations. The interpretation is of a continuing braided stream environment but with decreasing rainfall. Sediment movement in water would have become less frequent as a result of the drying out, and movement of sand-sized grains by wind would have increased. This model accounts for the existence of upward-fining units, in which conglomerates and trough cross-bedded sandstones pass up into fine-grained sandstone with fine lamination. Grain-size analyses of the Seamill and Fairlie sandstones in the Irvine district, and of the aeolian Penrith Sandstone of the Vale of Eden are given in (Table 5) and (Figure 4). The samples from the Fairlie Sandstone Formation were all taken from beds that display the criteria suggestive of aeolian origin. The curves in (Figure 4) show similarities in shape between the Fairlie and Penrith sandstones, both comprising well-sorted sandstones with a slight tail in the fine-grained silt fraction. The Seamill sandstone is more poorly sorted, with a markedly asymmetrical curve and a high proportion of coarser grain sizes. The grain-size analyses, taken in the context of the other criteria, described earlier, confirm that parts of the Fairlie Sandstone Formation were deposited by aeolian processes.

Chapter 5 Carboniferous: stratigraphical classification

The Carboniferous rocks in the Midland Valley of Scotland were formerly divided into six major units, the Calciferous Sandstone Measures, Lower Limestone Group, Limestone Coal Group, Upper Limestone Group, Passage Group and Coal Measures (Macgregor and MacGregor, 1948; MacGregor, 1960). The name Calciferous Sandstone Measures proved unsatisfactory and the classification of that part of the sequence was revised by Paterson and Hall (1986). Subsequent revisions to the rest of the sequence (Browne et al., 1996) have brought the lithostratigraphical nomenclature of the Midland Valley into line with established international practice with the definition of groups, formations and members. The revised lithostratigraphy is adopted for the Irvine district (Table 6).

Chronostratigraphy

The base of the Carboniferous has not been established in Scotland (Lumsden and Wilson, 1979), but it probably lies high in the Stratheden Group or low in the Inverclyde Group (Paterson and Hall, 1986). Macrofossils from the lowest beds of the Inverclyde Group do not include any definitive zonal forms. A miospore zonation has been erected in Scotland (Neves et al., 1973), but no evidence has been found for the presence of the two lowest zones in the scheme proposed for the British Carboniferous as a whole (Neves et al., 1972). The base of the Carboniferous in Scotland has, in practice, been drawn arbitrarily at the base of the Inverclyde Group.

Series and stages of the Carboniferous System, based on biostratigraphy, have been applied throughout Britain (Ramsbottom et al., 1978; George et al., 1976). These divisions are generally supposed to approximate to time units and so should be widely applicable, but as they are identified on the basis of biostratigraphical evidence, difficulties arise where faunas are restricted or absent. These difficulties have made it impossible to identify all the stages in the Irvine district, though the series boundaries can be drawn with reasonable certainty. The biostratigraphical framework for the Carboniferous strata of the Irvine district is described in Chapter 10.

(Table 6) summarises the chronostratigraphical nomenclature of the British Carboniferous and relates it to the lithostratigraphical nomenclature for the Irvine district, which reflects the particular lithological characteristics of the succession in that area. Marker horizons are used to define formation boundaries where that is appropriate. The relationship between chronostratigraphy and lithostratigraphy is only approximate, because lithostratigraphical units are by their nature diachronous, and because the stratigraphically significant faunas may not have been found in the area.

Inverclyde Group

The Inverclyde Group includes the Kinnesswood, Ballagan and Clyde Sandstone formations (Figure 5). These formations are characterised throughout the Midland Valley by sandstones with pedogenic carbonate ('cornstone') in association with silty mudstones containing thin beds of dolomite and limestone ('cementstone'). Colours range from grey, greenish grey and buff to shades of red and purple; the varied colouration differs from the dominant brick red of the underlying Stratheden Group. Absence of volcanic rocks, and rarity of black shales, rootbeds, coals and marine faunas serve to distinguish the Inverclyde Group from the overlying Strathclyde Group.

In the Irvine district the Kinnesswood Formation consists of sandstones, some of them coarse grained, with cornstones. These may form discrete carbonate beds or may occur as nodules. The Ballagan Formation consists mostly of mudstones with beds of cementstonc. Nodular carbonate also occurs. The Clyde Sandstone Formation consists mainly of arenaceous rocks with beds and nodules of cornstone similar to those in the Kinnesswood Formation. The three formations are essentially conformable with one another, but only the Kinnesswood Formation is widely present in the Irvine district.

Strathclyde Group

The Strathclyde Group in the west part of the Midland Valley includes the Clyde Plateau Volcanic Formation and the Kirkwood Formation, both largely of volcanic origin, and the sedimentary Lawmuir Formation. In general it differs from the Inverclyde Group in the presence of volcanic rocks, black shales, rootbeds, coals and marine bands, and the absence of cornstones and cementstones. The Strathclyde Group shares these features with the overlying Clackmannan Group but is distinguished from this by the relative paucity of marine bands and by its stratigraphical position below the Hurlet Limestone. The base of the Strathclyde Group in the western Midland Valley is generally disconformable on various formations of the Inverclyde and Stratheden groups.

In the Irvine district the Clyde Plateau Volcanic Formation consists mostly of basaltic lavas and associated tuffs and was formerly known as the 'Clyde Plateau Lavas' (Geikie, 1897). The Kirkwood Formation consists of tuffaceous mudstones and tuffs overlying the lavas. The tuffaceous lithologies are locally intercalated with nontuffaceous sedimentary rocks, including some marine strata. The Lawmuir Formation is laterally equivalent to the Kirkwood Formation and consists largely of sandstone. It is known only in the south of the district.

Clackmannan Group

The Clackmannan Group in the Irvine district consists, in upward sequence, of the Lower Limestone, Limestone Coal, Upper Limestone, and Passage formations. Much of the sequence is characterised by a generally cyclical pattern of elastic sedimentation with thin marine bands and terrestrial coals, though the highest part is of volcanic origin. Most of the boundaries between formations are conformable. The base of the group is drawn at the base of the Hurlet Limestone, the first marine marker horizon that can be recognised with certainty in all parts of the Midland Valley. The presence of strongly cyclical sequences, and the frequency of marine bands, also distinguish the Clackmannan Group from underlying groups.

The Lower Limestone Formation consists of marine limestones and mudstones with other subsidiary elastic lithologies. The base is drawn at the base of the Broadstone (Hurlet) Limestone. The Limestone Coal Formation is characterised by a series of mainly upward-coarsening cyclical sequences of mudstone, siltstone and sandstone with coal and ironstone. The lowest beds are distinguished as the Kilbirnie Mudstone Member. The base of the formation is conformable on the Lower Limestone Formation and is taken at the top of the Top Hosie Limestone. The Upper Limestone Formation is characterised by cycles in which limestone is overlain by mudstone, siltstone and sandstone capped by rootbeds and coal. The base of the formation is drawn wherever possible at the base of the Index Limestone, a marker horizon recognised over a wide area of the Midland Valley. The top is taken generally in the Midland Valley at the top of the Castlecary Limestone but this is not known to be present in the Irvine district and a disconformable boundary with the Passage Formation is inferred here. The basal unit of the Passage Formation consists of pale grey medium- to coarse-grained sandstones with seatearths and seatclays, which are well developed in places. Subsidiary siltstones and mudstones are also present, some with marine faunas. The base is drawn at the erosive base of a sandstone, and may be disconformable. The overlying Troon Volcanic Member of the Passage Formation consists of basaltic lavas, with some sedimentary intercalations. The base is marked by the lithological change from elastic strata to basaltic lavas. The Ayrshire Bauxitic Clay Member of the Passage Formation immediately overlies the Troon lavas and is characterised by the presence of bauxitic clay and related rock types. Interbedded sediments such as coals, seatclays and mudstones are also included. The base is taken at the lithological change from basalt to bauxitic clay; this junction is normally gradational.

Coal Measures

The Coal Measures is a group comprising three formations: Lower, Middle and Upper Coal Measures. The sequence comprises repeated elastic cycles with rootbeds and coal. The strata are generally grey in colour but are extensively reddened towards the top. The base of the Coal Measures in Scotland is generally taken at the base of the Lowstone Marine Band but in the Irvine district this horizon has not been found and the line is drawn at the litho-logical boundary between the Ayrshire Bauxitic Clay and the coal-hearing strata above. The Lowstone Marine Band lies a little way above the base of the Langsettian (Westphalian A) stage as defined at the base of the Subcrenatum Marine Band, a horizon that has not been identified with certainty in Scotland.

The Lower and Middle Coal Measures in the Irvine district both show grey elastic lithologies and contain many coal seams, some of them workable. The boundary between the two formations, as elsewhere in Britain, is drawn at the Vanderbeckei (Queenslie) Marine Band, which also defines the Langsettian/Duckmantian stage boundary. The Queenslie Marine Band has not been found in the Irvine district, and the top of the Shale Coal is taken as the boundary here (Chapter 10).

The Upper Coal Measures in the Irvine district are commonly reddish brown and purplish grey in colour. Coal seams are not common. The line between Middle and Upper Coal Measures in Scotland is drawn at the Aegiranum (Skipsey's) Marine Band, which is also the Duckmantian/Bolsovian stage boundary. This practice differs from that adopted in England and Wales, where the line is drawn at a higher level in the Bolsovian Stage, at the Cambriense Marine Band. The existence of this band has not been proved in Scotland.

Chapter 6 Tournaisian: Inverclyde Group

Over most of the Irvine district, the Inverclyde Group consists only of the Kinnesswood Formation; developments of the Clyde Sandstone and Ballagan formations are very localised (Table 7). The Kinnesswood Formation is present in the coastal exposures at Ardrossan and a little to the north, with outcrops also on Horse Isle and Great Cum-brae. The Ballagan Formation crops out on Great Cum-brae and is known from boreholes in the area south of the Inchgotrick Fault, but in the Irvine district does not occur in the region between the Inchgotrick Fault and the Largs–Hunterston Fault Zone. The Clyde Sandstone Formation is present on Great Cumbrae, and in a very localised development between the Kinnesswood Formation and the Clyde Plateau Volcanic Formation at Meikle Busbie.

Each formation within the Inverclyde Group is litho-logically distinct but together they represent environments transitional between the purely fluvial regimes of the late Devonian and those subject to increasing marine influence through Visean times.

Kinnesswood Formation

The Kinnesswood Formation in the district consists of beds of yellow and white sandstone commonly with red or green mottling. Conglomerate beds with clasts consisting almost exclusively of vein quartz pebbles are common. The beds are arranged in upward-fining cycles, 2 to 3 m thick, with the conglomerates and more massive sandstones at the base and finer-grained cross-bedded sandstones at the top. On Great Cumbrae a mudstonedominated unit, the Foul Port Mudstone Member, occurs between two arenaceous members (Table 7).

The Kinnesswood Formation is characterised by the presence of carbonate, which may be present as a cement filling the intergranular spaces of the sandstones or, more commonly, as nodular growths which have disrupted or replaced the primary fabric. Where appreciable quantities of carbonate are present the rock is termed cornstone. The chemical composition of the carbonate in the Kinnesswood Formation, as determined in three hot-tholes in the Greenock district (Paterson et al., 1990), ranges from calcite to dolomite.

Over most of the mainland area between the Inchgotrick Fault and the Largs–Hunterston Fault Zone, the Kinnesswood Formation is overlain disconformably by the Clyde Plateau Volcanic Formation. However, to the south of the Inchgotrick Fault, and to the west of the Largs–Hunterston Fault on Great Cumbrae, the Kinnesswood Formation is overlain by the Ballagan Formation.

The sequence is lacking in fossils but there is some evidence from outside the district that the Kinnesswood Formation may be partly of late Devonian and partly of early Carboniferous age.

Environment of deposition

The lithologies of the Kinnesswood Formation indicate a depositional environment only slightly different from that of the underlying Seamill and Kelly Burn sandstone formations of the Stratheden Group (Chapter 4). The sedimentary structures and grain-size range are generally similar, so deposition in braided stream environments is likely to have continued. The principal difference is in the presence of 'cornstones', the carbonate nodules which were probably formed in soil profiles soon after deposition.

The overall environment can be interpreted as a broad alluvial plain, which occupied a large part of the Midland Valley at this time, with a generally eastward-flowing river system. The palaeogeography would have been similar to that of the Stratheden Group (Figure 3), though with a larger area of deposition. It is inferred that the cross-bedded sandstones were deposited in the river channels and that the cornstones formed in soil profiles which developed on the associated floodplains under the influence of a fluctuating water table in a semi-arid climate. The rare argillaceous deposits were probably deposited in ephemeral lakes within this system. The cornstones developed either in sandstones at the top of channel-fill sequences after the channels had been abandoned or in dried-out overbank areas. The cornstones range from immature, in which the sandstones have a partial carbonate matrix with ill-defined concretions, to mature, in which well-defined vertically elongated nodules are associated with massive carbonate beds. Time is an important factor in the development of a mature cornstone, so these may be associated with slower rates of sedimentation, a view supported by the occurrence at Meikle Busbie of a mature cornstone profile associated with a thin development of Kinnesswood Formation.

Clyde coast north of the Inchgotrick Fault

The Kinnesswood Formation is best seen at Meikle Busbie [NS 238 459] where a massive cornstone about 8 m thick overlies about 2 m of mottled sandstone with carbonate nodules (Plate 3). The formation rests conformably on the Seamill Sandstone Formation, the boundary being transitional over a few metres. The Kinnesswood Formation at this locality is overlain by up to 5 m of Meikle Busbie Sandstone (see below) and then by the first of a series of lava flows in the Clyde Plateau Volcanic Formation. The cornstone bed forms a prominent topographical feature, as does the base of the lava pile. Traced laterally these features show that the lavas overstep the Meikle Busbie Sandstone and the Kinnesswood Formation to rest on the Seamill Sandstone.

Similar rocks are exposed on the foreshore north of Ardrossan at Burnfoot Bridge [NS 225 434] where cornstone is developed as discrete carbonate nodules within a sequence of cross-bedded sandstones and vein-quartz conglomerates. Sedimentary structures in these are similar to those of the underlying Seamill Sandstone Formation but the change in colour from red to pale mottled yellow and the addition of carbonate, indicate the change to Kinnesswood Formation. South of Ardrossan Harbour and on Horse Isle [NS 21 42], the Kinnesswood Formation comprises pale yellow sandstone and vein quartz conglomerate with only a few carbonate concretions. The boundary with the Seamill Sandstone Formation is not exposed.

Great Cumbrae

The Kinnesswood Formation on Great Cumbrae comprises three mutually conformable members.

Doughend Sandstone Member

This member at the base of the sequence, consists mainly of sandstone with a few cornstone concretions. The member is about 30 m thick and crops out at Doughend Hole [NS 149 545] on the south-west of Great Cumbrae, a short distance beyond the Irvine district boundary. The base is apparently conformable on the Kelly Burn Sandstone.

Foul Port Mudstone Member

This member, above, consists mainly of red-brown silty mudstone with concretions of cornstone. The member is  over 200 m thick and crops out on the shore [NS 157 544] at Millport, to the west of the Foul Port Fault Zone.

West Bay Cornstone Member

This member at the top of the sequence consists mainly of variegated sandstone with concretions of cornstone. The member is about 16 m thick and crops out on the shore [NS 157 543] west of Millport. It is overlain conformably by the Ballagan Formation.

South of the Inchgotrick Fault

A full sequence of the Kinnesswood Formation, about 186 m thick and resting unconformably on Lower Devonian lavas, was proved in the Deacon-hill Borehole. It consists of sandstone with numerous beds of cornstone up to 7 m thick. Cornstones are less common towards the top and base of the formation, and the individual cornstone beds are thinner in the lower half of the formation than in the upper. Conglomerate beds with quartzite pebbles are common, and towards the top of the formation mudstone beds are also present. The top of the formation is marked by a bed with mudstone clasts indicating a period of erosion, and possibly a disconformity, beneath the Ballagan Formation.

Ballagan Formation

The Ballagan Formation is characterised by the presence of 'cementstones' — thin beds or nodules of dolomitic limestone or dolomite — in a generally grey argillaceous sequence. The base is marked by a change in grain-size from the sandstones of the underlying Kinnesswood Formation up into mudstones. The change may take place over a few metres, or may be abrupt. The top may be conformable, as with the Clyde Sandstone Formation on Great Cumbrae, or disconformable, as to the south of the Inchgotrick Fault.

The Ballagan Formation occurs at outcrop only on Great Cumbrae, where it is about 22 m thick. In the mainland part of the district the formation is not present north of the Inchgotrick Fault, but south of the fault it is known from outcrops and boreholes immediately to the east of the district. A full sequence of Ballagan Formation, 150 m thick, was encountered in the Deaconhill Borehole and the top part was penetrated in the Harelaw Borehole. As proved in the boreholes, the sequence is made up of medium to dark grey mudstones with subsidiary beds of siltstone, and some thin ripple cross-laminated or cross-bedded sandstones. Cementstone beds and nodules are common. Most have no apparent internal structure but some are brecciated and a few are laminated. Gypsum ribs up to about 1 cm thick are common, and pyrite occurs within the mudstones as irregular nodules between 1 and 3 cm across. The overlying unit in the Deaconhill Borehole is the Lawmuir Formation. Similar lithologies, but without the gypsum and pyrite, are exposed in the Great Cumbrae sections.

Environment of deposition

The cementstones are generally dolomitic as in other parts of the Midland Valley (Paterson et al., 1990; Francis et al., 1970). The occurrence of the evaporite mineral gypsum, and the lack of other potential sources of magnesium, suggest that the magnesium in the dolomite was derived from hypersaline sea-water which had been concentrated by evaporation. It has been suggested that the laminated cementstones were deposited as carbonate muds (Belt et al., 1967) and that the nodular beds are diagenetic in origin, probably formed at an early stage and at a shallow depth in the sediment (Andrews et al., 1991). The mudstones are normally grey in colour, poorly laminated, and may have desiccation cracks, features which are compatible with deposition in a shallow-water area with restricted circulation. The abundance of gypsum veins may be due to the conversion of anhydrite to gypsum since, as suggested by Shearman et al. (1972), this change involves an increase in volume of 63 per cent in the solid phase; gypsum veins occurring in association with gypsum nodules are likely to represent, at least in part, the additional volume created by this conversion. Scott (1986) recorded textural and mineralogical features in the Ballagan Formation of Berwickshire which he interpreted as evidence for the former existence of gypsum or anhydrite. He also pointed out that the deposition of gypsum would reduce the amount of dissolved calcium, thus increasing the relative amount of magnesium in the brine and promoting the requisite Mg/Ca ratio for the formation of dolomite.

Clyde Sandstone Formation

The Clyde Sandstone Formation is a mainly arenaceous sequence with cornstones, lying between the Ballagan Formation and the Clyde Plateau Volcanic Formation. Within the Irvine district it occurs in two unconnected outcrops, one on Great Cumbrae and one on the mainland at Meikle Busbie.

Great Cumbrae

Millport Cornstone Member

This member crops out along the shore at Millport [NS 161 548]. It comprises mainly grey sandstone and red-brown silty mudstone with cornstone nodules. It is about 70 m thick and rests on the Ballagan Formation. These rocks probably represent a return to the type of environment present when the Kinnesswood Formation was forming.

Eileans Sandstone Member

This member, which rests on the Millport Cornstone Member, crops out on the Eileans [NS 164 546] in Millport Bay, just to the west of the district, so is likely to occur within the district on the sea bed of Millport Bay [NS 162 542]. It consists mainly of white cross-bedded sandstones and has a minimum thickness of 150 m.

Meikle Busbie

Meikle Busbie Sandstone Member

This member forms a small outcrop at Meikle Busbie [NS 238 459] about 3 km north of Ardrossan. It is a medium-grained light grey sandstone with plant fragments throughout. The base is a disconformity at the top of the highest cornstone of the Kinnesswood Formation and the top is a local angular unconformity with the overlying lavas of Clyde Plateau Volcanic Formation. At its maximum it is about 5 m thick; laterally it is cut out by the unconformity. The presence of plant material suggests that the member might alternatively be regarded as a basal unit of the Strathclyde Group.

Chapter 7 Visean: Strathclyde Group and lowest part of Clackmannan Group

The Strathclyde Group and Lower Limestone Formation are of Visean age (Table 6). The main area of outcrop is in the northern part of the district (Figure 2). The Strathclyde Group includes the volcanic rocks of the Clyde Plateau Volcanic Formation, the volcaniclastic rocks of the Kirkwood Formation and the sedimentary rocks of the Lawmuir Formation. The sedimentary Lower Limestone Formation is the basal division of the Clackmannan Group. The distribution of these units has a topographical expression in that the relatively harder rocks of the Clyde Plateau Volcanic Formation form the upland area of the Renfrewshire Hills.

Volcanism had an important influence on sedimentation during the Visean in the Irvine district, changing the surface topography and providing a lithologically distinct sediment source. Lateral thickness variations of the volcanic rocks appear to have been produced by differential subsidence, with contemporaneous faulting playing an important role. Thickness variations occur across the Dusk Water Fault, and the Inchgotrick Fault scarp all but stopped the southward extension of the lavas.

The lateritic weathering of the lavas provided the source for the distinctive sediments of the Kirkwood Formation. These strata are, in general, considerably thicker than estimated by Richey et al. (1930). They are iron-rich, and kaolinite is the dominant clay mineral, with some montmorillonite. The Kirkwood Formation was deposited from ephemeral streams and rivers, and infills hollows in the volcanic land surface. Lateritic weathering continued after deposition, further altering the deposit.

The land surface formed by the volcanic rocks and their mantle of volcaniclastic material had little relief but as it subsided it controlled the extent of marine deposition. A 'gulf' which had confined the early marine deposition to an area round where the Old Mill and Kirkwood boreholes are sited (Figure 11), (Figure 12) became less significant with each successive transgression.

The depositional environments of the Lower Limestone Formation were dominantly shallow marine. Such sedimentation is particularly sensitive to changes in relative sea level. A local pedogenic horizon, the 'White Post', formed on emergent mounds of the Dockra Limestone during a relative fall in sea level in the immediate post-Dockra Limestone period. Many of the other limestones within the Lower Limestone Formation also have rooted tops, indicating changes in relative sea level.

Contemporaneous faulting was not significant after the volcanic eruptions ceased, and bed-for-bed limestone correlations can be made across most of the major faults. However, the area to the south of the Inchgotrick Fault formed a tectonic high subject probably to erosion during the volcanic period, and sedimentation did not resume there until Brigantian times, when the Lawmuir and

Lower Limestone formations were deposited. The top of the latter is eroded here, indicating a period of localised uplift and erosion in the early Namurian (Chapter 8).

Clyde Plateau Volcanic Formation

The Clyde Plateau Volcanic Formation is the thickest and most extensive volcanic sequence of Dinantian age in Scotland. These rocks, which have frequently been referred to as the 'Clyde Plateau Lavas', form most of the high ground to the north, west and south-west of the Clyde valley from Stirling westwards to Greenock and from there south-eastwards to Strathaven. The formation attains a maximum thickness of some 1000 m in the northern part of the Renfrewshire Hills (Paterson et al., 1990). The lavas are mostly basaltic in composition and were erupted subaerially. Major fault lines, mostly trending north-east, divide the outcrop into several blocks, each having its own petrological characteristics and volcanic succession. The fault lines may therefore have been active or existed as topographical features at the time of the eruptions, controlling the extent of individual lava fields (see Chapter 14). In the Irvine district the volcanic rocks occur in three such fault-bounded blocks: the Kilbirnie Hills, constituting the southern extension of the Renfrewshire Hills of the Greenock district; the south-western part of the Beith–Barrhead Hills; and the western end of the Dunlop–Eaglesham Hills.

The base of the formation crops out along the western side of the district from Largs to Ardrossan. In the Kilbirnie Hills it rests upon the Fairlie Sandstone Formation in the north, and upon the Seamill Sandstone Formation in the south. Further south, between the extension of the Paisley Ruck and the Dusk Water Fault at Ardrossan, the lavas rest upon the Kinnesswood Formation. To the north of the district the formation rests on strata ranging up to the Clyde Sandstone Formation (Paterson et al., 1990). This widespread basal disconformity is a result of earth movements and erosion in the early Carboniferous, following deposition of the Inverclyde Group.

The top of the formation crops out on the south-eastern edge of the Kilbirnie Hills and around the Beith and Dunlop hills. In all of these outcrops the subaerially eroded surface of the lava succession is overlain by volcanic detritus of the Kirkwood Formation.

The radiometric age of the Clyde Plateau Volcanic Formation has been suggested as 335 to 325 Ma, based upon K-Ar whole-rock and mineral dates of the fresher lavas and intrusions (De Souza, 1982). Data from the Irvine district included in this calculation are basalt flows from Hapland, Dunlop (326 ± 7 Ma) and from Cock Law in the Kilbirnie Hills (327 ± 7 Ma) (De Souza, 1979). Both are whole-rock determinations and, as individual data points, should be regarded with caution.

The volcanic rocks are predominantly basaltic, although the Renfrewshire–Kilbirnie Hills and Dunlop–Eaglesham blocks contain a significant proportion of more evolved compositions (hawaiite, mugearite and, particularly just outwith the Irvine district, trachyte and rhyolite). Pyroelastic and epiclastic rocks are subordinate within the main basaltic successions, but in the Knockside Hills, 7 km north-west of Kilbirnie, a wide apron of pyroclastic deposits surrounds a group of trachytic plugs. Large trachytic intrusions also occur at Lochlands Hill and Brownmuir Plantation, 3 km north-east of Beith, and it is possible that trachytic deposits were once more widespread, since trachytic and felsite pebbles are relatively common in detritus of the Kirkwood Formation within the district.

The sources from which the dominant basaltic lavas were erupted are not apparent. Six basaltic plugs are recorded in the Kilbirnie Hills (Chapter 12) but, since they are only obvious where they cut through sedimentary rocks, others may occur unrecognised within the lava outcrop. It seems likely that many of the basaltic flows were erupted from fissures, as has been proposed for the northern part of the Renfrewshire Hills (Paterson et al., 1990). Possible feeder dykes for such eruptions are concentrated in a zone which continues the Dumbarton–Fintry volcano-tectonic line of the Campsie and Kilpatrick hills (Whyte and MacDonald, 1974) south-westwards through the Renfrewshire Hills to Largs and South Bute. This zone intersects the north-western corner of the Irvine district, where abundant north-east- trending dykes crop out, particularly on the east coast of Great Cumbrae. It is possible that much of the volcanicity was related to a deep crustal fracture along this line, inherited from the Caledonian Orogeny. Several basaltic agglomerate-filled vents in the area of the Kilbirnie Hills were formerly regarded as post-Dinantian in age (Richey et al., 1930). The evidence is somewhat circumstantial and it is possible that some, at least, may he contemporaneous with the Clyde Plateau lavas.

The eruptions seem to have been entirely subaerial, with periods of tropical weathering between successive flows; the lava tops are commonly marked by red-brown lateritic boles underlain by zones of deep weathering. Basaltic to mugearitic flows are between 3 and 10 m in thickness, much of which may be slaggy, amygdaloidal and hydrothermally altered, with only the central part consisting of fresh, massive, cross-jointed lava. Columnar jointing is rare. The alternation of massive and slaggy parts of flows gives rise to good trap featuring in places. Individual flows can seldom be traced for any great distance, however, on account of their lenticular character, the faulted nature of the ground, and the lack of exposure in many areas.

Petrology of the lavas

The lavas of the Clyde Plateau Volcanic Formation in general constitute an alkali-basalt series with a wide range of compositions throughout the range ankaramite-basaltbasaltic hawaiite-hawaiite-mugearite-benmoreite-trachyte-rhyolite. Comparable assemblages are found in other Dinantian volcanic formations of central and southern Scotland (MacGregor, 1937; 1948; Tomkeieff, 1937) and Ireland. General details of the geochemistry of the Dinantian lavas are given by Macdonald (1975) and Smedley (1986a). More detailed studies of the trace element and isotopic compositions of the most primitive members of the series have been used to investigate the nature of the mantle source from which the magmas were melted (Macdonald, 1980; Smedley, 1986b, 1988). Analyses of the Clyde Plateau Volcanic Formation show that the most basic lavas are mainly hypersthenenormative, mildly alkaline or transitional rocks with some of slightly silica-undcrsaturated, nepheline-normative composition. The more fractionated rocks are all hypersthene-quartz-normative and are usually subdivided geochemically according to their Differentiation Index and normative plagioclase composition, so as to facilitate comparison with the classification scheme of Macdonald (1975) (Figure 8). Similar subdivisions are achieved using the TUGS recommended Total Alkalis versus Silica diagram (Le Maitre, 1989). In terms of that classification these rocks are intermediate between typically sodic and typically potassic series, which leads to problems of nomenclature. Analyses can fall on either side of the divide between the two series, but in general the names of the sodic series, hawaiite-mugearite-benmoreite, are used on the basis of petrographical criteria. The more general names trachybasalt and trachyandesite, used on earlier editions of the maps, are retained where petrographical features characteristic of more potassic rock are recognised.

Few analyses are available from the Clyde Plateau Volcanic Formation of the Irvine district. Ten analyses in the range basalt to hawaiite from Smedley (1986a) and one of rhyolite from Richey et al. (1930), all from the Kilbirnie Hills, are listed in (Table 8) and plotted on (Figure 8). However, numerous analyses are available also from the northward continuations of the Kilbirnie and Beith hills in the Greenock district (Paterson et al., 1990) and these have influenced the following discussion.

The lavas, particularly the more basic varieties, can be remarkably fresh, although olivine is usually replaced by red-brown pseudomorphs. Less fresh material shows varying degrees of albitisation, chloritisation, oxidation, hydration and replacement by carbonate. Albitisation in particular can lead to considerable difficulties in petrographical and petrochemical classification.

The more basic rocks are almost all porphyritic, with various combinations and proportions of plagioclase (white), olivine (red-brown) and clinopyroxene (black) phenocrysts. On this basis they have been classified in the field and in thin sections into six main types (Table 9), following MacGregor (1928). Of these, the Dalmeny (map symbol BD), Dunsapie (BD„) and Hillhouse (BD) types are all olivine-basalts, and the very mafic Craiglockhart type (lick) could be termed ankaramite. The feldsparphyric lavas, allotted on a petrographical basis to the Markle and Jedburgh types, have compositions in the range olivine-basalt through basaltic hawaiite to hawaiite. Within this range, most of the basaltic Markle lavas as defined by MacGregor (1928) contain microphenocrysts of olivine visible in hand specimen, and are identified on the map by the symbol Bps. Such lavas occur in the Dunlop area and in the Beith–Barrhead Hills. Other feldsparphyric lavas, notably in the Renfrewshire–Kilbirnie Hills contain no olivine phenocrysts and have little or no visible olivine in the groundmass. Analyses indicate that the majority of these are basaltic hawaiites or hawaiites, although some may be classed as basalts. The flows of this group are referred to by the general term 'Markle lavas' and are indicated on the map by the combined symbol fBW which represents feldsparphyric (f) lavas of basaltic (B) to hawaiitic (W) composition. A similar distinction is made between true ledburgh basalts' (Bj) and jedburgh lavas' of unspecified basaltic to hawaiitic composition (BW).

Nonporphyritic lavas in the district are usually somewhat paler than the basalts and typically show a pronounced platy jointing due to flow-alignment of feldspar crystals. In accordance with usual practice in the Scottish Dinantian volcanic sequences, these rocks have been classified in the field as mugearites (Wm), although available analyses from areas outwith the present district show that they range from hawaiite through mugearite to benmoreite or trachyandesite in composition. In thin section they consist mainly of small fluxioned oligoclase feldspar laths with a relatively high proportion of iron-titanium oxide grains. Clinopyroxene and olivine are commonly absent but where present are less abundant than in the basalts and are usually altered.

Trachybasalt and trachyandesite flows are abundant further east, in the adjacent Kilmarnock district, but are present only on the eastern margin of the Irvine district. They are distinguished from the mugearites by the presence of K-feldspar in the groundmass and commonly have phenocrysts of amphibole in addition to sporadic microphenocrysts of plagioclase and more rarely olivine and clinopyroxene. The amphibole phenocrysts are commonly resorbed. Trachyte lavas do not occur in the Irvine district.

A rhyolite flow or plug near Swinlees [NS 29 52] is highly altered. Sporadic phenocrysts of feldspar, up to 2 mm long, are represented by zoisite pseudomorphs and relics of ferromagnesian minerals may be present. Most of the rock consists of a felsitic mosaic of tiny feldspar laths enclosed in cloudy quartz.

Detailed petrographical descriptions of all of the igneous rock types in the district and of many individual samples are given in the previous editions of this memoir (Richey et al., 1930).

Kilbirnie Hills

The Kilbirnie Hills are the southern continuation of the Renfrewshire Hills of the Greenock district, where the Clyde Plateau Volcanic Formation has been divided into six members (Paterson et al., 1990, table 6). An overall gentle regional dip towards the south-east throughout the block causes the basal units to crop out on the western side of the hills, with successively younger units towards the east. The total thickness in the district was estimated by Richey et al. (1930) as at least 300 m, but modern gravity and magnetic modelling suggest that the total thickness may be closer to 500 m (see Chapter 13). Distribution of the members is shown on (Figure 6). The two lowest members of the Renfrewshire Hills succession, the Noddsdale Volcaniclastic Beds and the Largs Lavas, are not recognised at outcrop south of Largs, so are unlikely to extend far into the Irvine district. The lowest lavas in the Kilbirnie Hills are predominantly basaltic to hawaiitic lavas of Jedburgh type and hence correlate with the Greeto Lavas. The bulk of the lava sequence, however, constitutes a thick succession of basaltic to hawaiitic Markle type lavas and mugearites characteristic of the Strathgryfe Lavas. More mafic basaltic lavas, characteristic of the upper part of the Renfrewshire Hills succession (the Marshall Moor and Kilbarchan lavas), extend for only a short distance into the Irvine district around Ladyland [NS 32 57].

Noddsdale Volcaniclastic Beds And Largs Lavas

Basaltic and ankaramitic lava flows of Dalmeny and Craiglockhart type crop out in the lower part of the Gogo Water, Largs, and extend southwards on the lower hill slopes of Douglas Park to the A760 road. Volcaniclastic deposits seen beneath these flows in the Largs Borehole are not exposed but are inferred to be present.

Greeto Lavas

Lavas of this member form the main scarp feature at the top of the hills behind Largs and Fairlie. The scarp extends south-south-east on the east side of the Caaf Water valley. An outlier caps Kaim Hill [NS 22 53] to the west of the valley. The member is about 100 m thick in the Largs area, but thins southwards and lenses out at Wardlaw [NS 24 51]. Thin beds of purple nodular tuff with intercalated fine-grained red marls are developed locally beneath the lavas and are exposed in a few stream sections, for example east and south of Bradshaw [NS 24 53]. These beds are regarded here as basal to the Greeto Lavas and not as part of the more widespread Noddsdale Volcaniclastic Beds which form the base of the overall succession to the north. The lavas are basaltic to hawaiitic, mainly of Jedburgh type with a few flows of Markle type. A single flow of ankaramite within the member on the north slope of Castle Hill at Largs represents a return to the more basic eruptions of the previous member.

Strathgryfe Lavas

The outcrop of this member covers most of the Kilbirnie Hills to the east of the main escarpment. Its maximum thickness in the Irvine district is probably no more than 250 m, compared with up to 750 in in the Renfrewshire Hills. The extensive outcrop reflects gentle dips. The member extends southwards beyond the limit of the Greeto Lavas and rests directly upon the Seamill Sandstone Formation. It thins markedly to the south and in the area of the Busbie Muir and Munnoch reservoirs [NS 24 46] and [NS 25 47], where it is cut out by the extension of the Paisley Ruck fault system, it is represented by a few flows only. On the south-eastern edge of the Kilbirnie Hills this member is overlain directly by the Kirkwood and Lower Limestone formations.

The sequence is composed mainly of mugearites and lavas of Markle type in approximately equal amounts, together with a few lavas of Jedburgh type. As in the Renfrewshire Hills, many of the Markle and Jedburgh type lavas contain no olivine phenocrysts and have a relatively felsic groundmass. It seems likely, therefore, that they are of hawaiite or basaltic hawaiite, rather than basalt composition.

In the Knockside Hills, on the northern edge of the Irvine district, there is an extensive outcrop of pyroclastic breccia in the middle part of the member. The breccias are predominantly trachytic and are cut by plugs of trachyte (Chapter 12). They constitute the most southerly part of the Misty Law Trachytic Centre, the remnant of a trachytic-rhyolitic central volcano developed within the Strathgryfe Lavas of the Greenock district (Wilson, 1916; Johnstone, 1965; Paterson et al., 1990). The pyroclastic breccias were formerly regarded as occupying a large vent, but it seems more likely that, for the most part, they constitute an apron of extrusive pyroclastic rocks surrounding a series of smaller vents in the Irish Law–Knockside Hills area [NS 25 58]. True vent breccias most probably do occur in the vicinity of the trachytic plugs, but they cannot be distinguished on the map.

Farther south, rhyolites occur at the top of the lava succession to the west of Swinlees Farm [NS 29 52]. Here the rock has steeply inclined fluxion banding and, at its most northerly exposure, is vesicular and brecciated at the junction with the rotted top of an underlying mugearite. The rhyolite forms an upstanding mass on Carwinning Hill [NS 28 52], but has a sheet-like extension to the south-west. It is not clear whether it represents an intrusion or a viscous lava dome spreading from a plug. Close to a faulted margin with sedimentary rocks, the rhyolite is host to copper mineralisation and the rock has been worked in recent times for aggregate (Chapter 2).

Marshall Moor and Kilbarchan Lavas

Outcrops of relatively mafic basalt on the east side of the Maich Water around Ladyland [NS 32 57] are probably at the southern limit of these two members which, farther north, constitute a distinctive upper part of the volcanic succession of the Renfrewshire Hills. A flow of macroporphyritic basalt of Dunsapie type (like the Marshall Moor Lavas) is overlain by at least three microporphyritic flows of Hillhouse and Dalmeny type basalt (like the Kilbirnie Lavas). Junctions with overlying sedimentary rocks are normally faulted at outcrop but Dalmeny type basalts at the top of the volcanic succession were penetrated beneath sedimentary strata in the base of the Lora Burn Borehole.

Meikle Busbie to Saltcoats

A thin volcanic sequence between the extension of the Paisley Ruck and the Dusk Water Fault (Figure 6) appears to be a continuation of the Kilbirnie Hills outcrop, which in this area consists entirely of the Strathgryfe Lavas. However, the presence of mafic basalts of Dalmeny type suggests that this sequence may be part of a different succession, possibly related to that of the Beith area, which lies in the same fault block. The sequence normally rests unconformably on Kinnesswood Formation but at Meikle Busbie [NS 238 459] a wedge of up to 5 m of white sandstone with numerous indeterminate plant remains, the Meikle Busbie Sandstone, intervenes (Chapter 6). The lavas are predominantly basalts of Dalmeny type in the lower part of the sequence but the upper part consists of basaltic to hawaiitic flows of Markle type and some mugearites. The total thickness is probably no more than 150 m.

Beith Hills

The Beith–Barrhead Hills constitute an area of relatively high ground, 16 km by 6 km, extending north-eastwards into the adjoining districts of Kilmarnock, Greenock and Glasgow. The hills are bounded by a north-east-trending extension of the Paisley Ruck and by the Dusk Water Fault. Between these faults, the lava succession is folded into a shallow north-east-trending structure, the Beith Anticline, which plunges to the south-west. Dips on the limbs seldom exceed 8°. Richey et al. (1930) recognised four divisions within the succession. These have been traced northwards into the Greenock and Glasgow districts and have been given member names (Paterson et al., 1990; Hall et al., 1998):

The distribution of the members in the whole of the Beith–Barrhead Hills is shown in (Figure 7). Their overall thickness may be no more than 250 to 300 m.

The lavas are almost entirely olivine-basalts with rare mugearites. As such they correspond most closely with the upper part of the Renfrewshire Hills succession (the Marshall Moor and Kilbarchan lavas) which crop out mainly in the Greenock district, although the Beith–Barrhead Hills have a greater variety of lava types. Direct correlation between the two blocks is not possible owing to the intervention of the Paisley Ruck faults. It is possible, however, that the lower members of the Renfrewshire Hills succession die out across this fault zone leaving only equivalents of the higher members, and possibly some younger lavas, in the Beith–Barrhead Hills. There is also some evidence in the Glasgow district (Hall et al., 1998) that lower members of the Beith–Barrhead Hills succession die out eastwards towards the Dusk Water Fault.

Gleniffer Lavas

Owing to the plunge of the Beith Anticline, these lavas crop out in the northeastern part of the district, to the east of Barcraigs Reservoir [NS 39 57]. The base of the member is exposed farther to the north-east, beyond the district boundary, and was penetrated by the Glenburn Borehole where volcaniclastic sedimentary rocks at the base of the lavas rest upon the Kinnesswood Formation. The member is characterised by olivine-basalts of a highly distinctive Markle type having abundant platy phenocrysts of plagioclase up to 25 mm long and microphenocrysts of red-brown olivine.

Serglantlaw Lavas

On the north-western limb of the Bcith Anticline these lavas crop out to the north-east and south of Barcraigs Reservoir. They are well seen in the area between Townhead of Threepwood [NS 39 55] and Gillies Hill [NS 38 54], where they define the closure of the anticline, but in the Irvine district they cannot be traced on the southeastern limb, north-east of Rigfoot [NS 40 55]. A small outcrop of Dalmenytype basalts north-east of Caldwell House Hospital [NS 41 54] may be part of this member. The lavas are mafic olivine-basalts, generally of Dunsapie and Dalmeny type, but in the area of the anticline closure the basal flow is a distinctive ankaramite (Craiglockhart type).

Fereneze Lavas

These lavas are best exposed on the north-west limb of the anticline, to the west of Barcraigs Reservoir, where they form a good scarp and dip slope topography ('trap features'). The flows are mainly olivine-basalts of Markle and Dunsapie type, but a borehole at Crummock Park, Beith, which penetrated 120 in into the member, encountered two flows of mugearite in its lower part. Three flows and a discontinuous bed of tuff are well seen in Loan-head Quarry [NS 365 555], 4 km north-east of Beith (Gribble, 1992). The flows are 15 in, 8 m and over 10 m thick, and are separated by boles up to 50 cm thick. The flows are largely amygdaloidal with much hydrothermal alteration giving a range of secondary minerals such as calcite, prehnite, analcime, thomsonite, natrolite, bowlingite and heulandite. The quarry is well known for copper mineralisation with native copper, malachite and cuprite recorded (Chapter 2). On the south-eastern limb of the anticline, the member is difficult to trace due to poor exposure and it is likely that it is for the most part faulted out. However, at Blaelochhead [NS 39 53], a flow of mugearite forms a massive scarp with an extensive dip slope to the south. This and underlying flows of Markle type are probably part of the member.

Beith Lavas

These lavas, which are almost entirely olivine-basalts of Dalmeny type, crop out on lower slopes at the southwestern limit of the Beith–Barrhead Hills. They occur on the north-western limb and around the closure of the anticline, but are faulted out on the south-eastern limb. At Beith the member is probably about 90 m thick, the lowest 50 m of which were penetrated by the Crummock Park Borehole. This borehole confirmed surface observations that large parts of the flows are extensively altered and decomposed. Red boles are commonly developed between flows and interbedded chocolate-coloured marls are present. South of Roebank Glen [NS 353 554] a flow of Markle-type basalt is exposed within the member. Volcaniclastic sedimentary rocks of the Kirkwood Formation are inferred to rest upon the member wherever the top crops out.

Dunlop

The lava outcrops around Dunlop are at the south-western end of a block of lavas, bounded by the Dusk Water and Annick Water faults, which covers an area of 8 km by 24 km between Dunlop and Eaglesham. This block is part of a large area of lavas that extends eastwards through the adjacent Kilmarnock and Hamilton districts to Darvel and Strathaven. The overall volcanic succession in this area contains a wide range of lava compositions from olivine-basalts of various types, through mugearite to trachyandesite and trachyte, together with extensive trachytic pyroclastic deposits (Richey, 1928; Richey et al., 1930). As such it is comparable with the succession in the Renfrewshire–Kilbirnie Hills.

In the area around Dunlop only the upper part of the sequence crops out, and the lavas are principally olivinebasalts of Dalmeny and Jedburgh type. The trachytic rocks, which characterise the succession as a whole, occur beneath these lavas and crop out to the east of the present district, apart from two very small areas of hornblende-bearing trachyandesite on the margin of the district east of Bourock [NS 40 51]. The succession in the immediate area of Dunlop (modified after Richey et al., 1930) is:

• Olivine-basalt of Dalmeny type, approaching Hillhouse type
• Olivine-basalt of Markle type
• Mugearite
• Olivine-basalts of Jedburgh type (extensive)
• Mugearite
• Hornblende-bearing trachybasalt
• Ankaramite (Craiglockhart type) — relative position uncertain
• Olivine-basalt of Markle type

Between Dunlop and Barrmill is an extensive outcrop of olivine-basalt of Dalmeny type with minor flows of Jed-burgh, Markle and Dunsapie type in the north-west. The area is too broken by faults for a sequence to be determined, but the flows must be near the top of the succession since interbedded and overlying volcaniclastic sedimentary rocks are exposed in old railway cuttings [NS 3929 5118] and in the Lugton Water. In the latter exposure [NS 3982 5115], a volcaniclastic sequence within the lavas, includes volcanic detritus with some fragments of trachyte like that of the Lochlands Hill intrusion [NS 37 55], and ostracod-bearing shales.

Kirkwood Formation

The Kirkwood Formation directly overlies the lavas of the Clyde Plateau Volcanic Formation and is composed dominantly of tuffaceous mudstones and tuffs which vary in colour from dark reddish brown to greenish grey. Non-tuffaceous rocks — sandstones, siltstones and limestones, generally grey in colour — are intercalated with the volcaniclastic material in a places. The limestones have been correlated, on faunal grounds, with the Blackbyre and Hollybush limestones of the Greenock district ((Figure 10); Wilson, 1979); these are members of the Lawmuir Formation. The Lawmuir and Kirkwood formations commonly interdigitate (Paterson et al., 1990, figure 7), but in the Irvine district, north of the Inchgotrick Fault, the dominant lithological characteristics are those of the Kirkwood Formation and the strata are considered here as a single formation. A thin development of Lawmuir Formation is recognised south of the Inchgotrick Fault (see below), where the Clyde Plateau lavas and Kirkwood Formation are absent.

At outcrop the Kirkwood Formation occurs mostly as isolated exposures of brown tuffaceous mudstone. One such exposure in the Dusk Water [NS 3124 4744] shows grey and brown tuffaceous mudstones with beds containing igneous pebbles. These rocks are demonstrably overlain by the Broadstone Limestone of the Lower Limestone Formation but the relationship with the underlying lavas, and the total thickness of the formation, cannot be determined here.

The Kirkwood Formation was proved in the Kirkwood, Old Mill, Muirlaught, Ardrossan, Lora Burn, Dunniflat and Coalhill No. 2 boreholes (Figure 11). The type section is from 38.90 to 75.45 m depth in the Kirkwood Borehole. The thickest development was proved in the last-named, where 41.39 m of strata were encountered. This is a minimum thickness for the formation, as the borehole did not reach the lavas. (Figure 11) shows that the Kirkwood Formation includes all the strata beneath the Broadstone cycle, where both Broadstone and Dockra limestones are present, and all beneath the Dockra cycle where the Dockra Limestone only is present.

The key features shown in the boreholes are that the Kirkwood Formation varies in thickness, the developments in the Lora Burn, Dunniflat and Ardrossan bore-holes being markedly thinner than in the other boreholes; that the lithofacies within the Kirkwood Formation vary laterally, marine horizons being impersistent or passing into other lithologies; and that all the tuffs and some of the tuffaceous mudstones are graded.

Thickness variation

The Kirkwood Formation thins towards the north-west, north and north-east, with a local depocentrc north-east of Ardrossan (Figure 9). In general the thinning takes place towards the areas where the Clyde Plateau lavas are thick, and so probably reflects the banking up of the volcanic detritus against the lava pile. Localised thinning in boreholes close to the Dusk Water Fault may reflect contemporaneous movement on the fault.

Lithological variation

The Kirkwood Formation consists predominantly of a volcanic lithofacies, with subordinate marine intercalations.

Volcaniciastic lithofacies

The volcaniclastic lithofacies is typically developed in the Coalhill No. 2, Muirlaught, Ardrossan, Lora Burn and Dunniflat boreholes (Figure 11), where it is composed almost exclusively of mottled tuffaceous mudstone with beds of coarse-grained tuff. The coarse tuffs invariably grade upwards into tuffaceous mudstones. In hand specimen the coarse tuffs are greenish grey in colour, poorly bedded, and composed of pebbles with a generally discoid shape. In thin section the pebbles are seen to be highly altered to clay minerals and in some cases are barely distinguishable from the matrix. Less altered pebbles are basaltic with relatively fresh ferromagnesian minerals preserved.

The tuffaceous mudstone is blocky, and generally mottled dusky red and greenish grey. Small, rounded, altered clasts of reworked tuffaceous mudstone or of volcanic material may be present, though these tend to be scattered and are usually less than 5 mm across. Some of the mottling may be associated with contemporaneous soil formation, but undoubted root structures are rare.

Kaolinite, as determined by X-ray diffraction techniques, is the dominant clay mineral in these tuffs and tuffaceous mudstones. Montmorillonite is also present in some of the tuffaceous mudstones, which causes them to swell in contact with water.

Marine lithofacies

Locally, the volcaniclastic lithofacies is intercalated with a marine lithofacies, mostly limestones and sandstones, which reflects a partial marine transgression on to the land area produced by the lava pile and its mantle of volcaniclastic material. The marine intercalations are present in a restricted area only ((Figure 12), centre), and are best demonstrated in the Old Mill and Kirkwood boreholes (Figure 11).    

In the Old Mill Borehole, 6.21 m of tuffaceous mudstones and tuffs rest on the Clyde Plateau lavas. Above lies the marine lithofacies, 1.3 m of greenish grey sandstone overlain by the Lower Old Mill Limestone. The limestone is coarse grained, medium to brownish grey at the base, fining upwards into a medium to dark grey well-bedded calcareous mudstone. Tuffaceous mudstone, 7.32 m thick, separates the Lower Old Mill Limestone from the Upper Old Mill Limestone, another intercalation of the marine lithofacies. This consists of coarse-grained limestone, made up largely of crinoid columnals and shell fragments. The rock has little argillaceous matrix material at the base but it fines upwards into a nodular limestone with greenish grey carbonate concretions in an argillaceous matrix. In-situ root developments are recorded immediately beneath the Upper Old Mill Limestone.

In the Kirkwood Borehole two marine horizons have been recognised. The lower consists of greenish grey siltstones with thin beds of concretionary limestone containing black algal patches; poorly preserved crinoid columnals have been recorded from the siltstones, and ostracods from the overlying tuffaceous mudstones. This development is 1.48 m thick and rests directly on the Clyde Plateau lavas. It may correlate with the Lower Old Mill Limestone in the Old Mill Borehole. A further marine intercalation occurs between 51.83 and 53.06 m in the borehole and is a bed of dusky brown tuffaceous mudstone containing rare, poorly preserved brachiopod moulds. The remaining strata within the Kirkwood Formation are of the volcaniclastic lithofacies, consisting of tuffaceous mudstones with frequent beds of greenish grey siltstone and fine-grained sandstone. In-situ root developments are recorded at the top of the formation, from immediately below the upper marine horizon, and at various levels in tuffaceous mudstones and siltstones between the two marine horizons.

Environment of deposition

The volcaniclastic lithofacies may have accumulated by direct ash, falls (either subaerially or into water), by reworking of weathered igneous material and consolidated volcanic ash, or by a combination of both processes.

The variable origin of this lithoficies is evidenced by the presence of montmorillonite-rich beds. These are characteristically formed by the in-situ weathering of fine-grained basic ash-fall deposits under tropical conditions; this, allied to the poorly sorted character of some of the tuffs, suggests that little reworking has taken place. Elsewhere, well-sorted and graded tuffs with rounded discoid-shaped clasts suggest reworking of igneous material and consolidated volcanic ash. It is therefore probable that both processes operated in the formation of the volcaniclastic lithofacies.

There is some evidence to suggest that the reworked sediments were themselves subjected to alteration and weathering following deposition. Kaolinite is a common constituent of the bauxitic clays that form by intensive weathering under present-day tropical or subtropical conditions. The common presence of kaolinite in the weathered tuffs and tuffaceous mudstones of the Kirkwood Formation may indicate that these were subjected to similar weathering processes. A period of postdepositional weathering is also suggested by the present condition of rounded clasts of volcanic or tuffaceous origin. These were capable of withstanding abrasion during transportation but are now altered to a soft, highly weathered state.

Assuming climatic factors remained constant, the areas that were exposed for longest should show the most mature weathering profiles. At Carnals Castle [NS 5682 4041], east of the boundary of the Irvine district, the Dockra and Broadstone limestones are absent and the weathering profile is mature, with beds rich in kaolinite and with kaolinite aggregations infilling cavities. This would suggest that here the Kirkwood Formation was subjected to quite intense weathering before being covered by younger strata.

The marine lithofacies contains two sets of lithologies, each deposited in a slightly different environment, a sandstone–siltstone suite and a limestone suite.

The sandstones and siltstones are generally greenish grey in colour and contain some volcanic material. Plant fragments, commonly coalified, are preserved in places. Mudstone clasts are common in the basal beds of the siltstone and sandstone units of the Kirkwood Borehole, some units having a marked erosive base. Erosive lower junctions, with mudstone clasts in the basal beds, suggest infilling of channels. The sandstone in the Old Mill Borehole is coarser-grained and better sorted than those at Kirkwood, and immediately precedes a marine limestone; although coalified plant fragments are present, the sandstone may represent a deposit marginal to the marine environment, such as a beach. Although no marine fossils have been recorded from this sandstone, it is unlikely that a mature sandstone like this could have been derived from a local hinterland composed largely of lavas, tuffs and tuffaceous mudstones. Longshore drift is a mechanism whereby relatively mature sand might have been moved into the area.

The limestone lithologies are marine in origin. Deposition of the Lower Old Mill Limestone in the Old Mill Borehole was initially in clear agitated water. Progressive upward increase in the mud content shows that the energy in the environment decreased with time, either to a deeper water setting or, more likely, to a quiet lagoon protected by barrier bars. The lateral equivalent of this horizon at Kirkwood is a sequence of fine-grained nodular limestones interbedded with siltstones containing some volcanic material and a few marine fossils. Here the environment of deposition must have been close to the shore. A restricted marine environment subject to some wave or current action is envisaged, where algae could flourish and be broken up to form the dark patches noted above.

The Upper Old Mill Limestone in the Old Mill Borehole shows a similar development to the Lower Old Mill Limestone at the same locality. The interpretation of environment of deposition is similar, though the top of the Upper Old Mill Limestone is concretionary, a feature most probably post-depositional in origin. The lateral equivalent of this horizon in the Kirkwood Borehole is a thin bed of tuffaceous mudstone with a poorly preserved marine fauna, and must represent the interface between the environments of the tuffaceous mudstone and the limestone — a shoreline.

The geographical extent of the two marine transgressions is approximately the same as the extent of deposition of the Old Mill limestones as shown in (Figure 12). In a wider context, if the Upper and Lower Old Mill Limestones correlate with the Blackbyre and Hollyhush limestones of the Glasgow–Paisley area (see above), then the V-shaped inlet into the Dalry Basin presumably opened in a northward direction.

Lawmuir Formation

In the Irvine district the Lawmuir Formation is known only from boreholes south of the Inchgotrick Fault. The Harelaw Borehole and the Deaconhill Borehole both showed that a sequence of arenaceous strata intervenes here between the top of the Ballagan Formation and the Broadstone Limestone at the base of the Lower Limestone Formation. The presence of coal and absence of cornstone show that these sandy beds belong to the Lawmuir Formation rather than the Clyde Sandstone Formation. In the Deaconhill Borehole the Lawmuir Formation is 6.72 in thick and is mostly sandstone with a thin bed of silty mudstone, a coal 0.07 m thick and seat-clays. The base of the formation appears conformable with the underlying Ballagan Formation. In the Harelaw Borehole the sequence is 11.63 in thick and is mostly sandstone with thin beds of siltstone and mudstone; in contrast to the Deaconhill Borehole there is an erosive boundary between the basal sandstone and the underlying Ballagan Formation. It is likely that the basal contact in this area represents a period of erosion or non-deposited, during which the Clyde Sandstone Formation and Clyde Plateau 'Volcanic Formation were deposited elsewhere.

Lower Limestone Formation

The Lower Limestone Formation is the lowest division of the Clackmannan Group. It comprises a cyclic sequence of limestones and mudstones with subsidiary siltstones and sandstones. Coals with seatclays are also developed locally. The cyclical nature of sedimentation is most apparent in the borehole sections (Figure 11), as these are more complete than the outcrop sections. The cycles show an alternation between marine and terrestrial conditions. Swamp conditions starved of elastic detritus and accumulating organic material (rooty beds and coal) gave way to marine conditions, with limestones and calcareous mudstones being deposited as the area was inundated by a relative rise in sea level. Swamp conditions then returned. In terms of the depositional environment the cycles suggest an area marginal to a shallow sea, with elastic sedimentary rocks formed as small local deltas advanced into the sea. The cyclicity has been discussed further by Monro (1982a).

In the Irvine district the formation includes the Broadstone Limestone, the Wee Post Limestone, the Dockra Limestone and the Hosie limestones (Figure 11). As the base of the formation is regionally defined at the base of the Hurlet Limestone, the correlation of this horizon with the limestones of the Irvine district is critical. Three possibilities have been suggested (Figure 10).

Macnair (1917) correlated the Hurlet Limestone with the lowest of the Hosie limestones, while Carruthers and Richey (1915) correlated it with the Dockra Limestone. Both views are based on a lithological comparison with the succession in the Paisley area. The third correlation, proposed by Burgess (1965) was based on the distribution of certain algal species of the genus Calcifolium within the succession. Burgess considered Calcifolium punctatum Maslov to he characteristic of the Hurlet Limestone, and C. okense Shretzov and Binna to characterise the overlying Blackhall Limestone. In the Irvine area only C. punciatum was found, and that within the Broadstone Limestone, leading Burgess to correlate this bed with the Hurlet, and the Dockra Limestone with the Blackhall Limestone.

An extensive examination of these strata, and of much new faunal evidence (Wilson, 1979; Monro, 1982b) has supported the correlation of the Broadstone Limestone with the Hurlet. The base of the Lower Limestone Formation within the Irvine district is therefore taken at the base of the Broadstone Limestone, where this is present. Where this limestone is not present the base of the formation is drawn at the upward lithological change from volcanic detritus to limestones and mudstones. In those areas where the Dockra Limestone rests directly on the Kirkwood Formation the nature of the junction is problematical. A disconformity is probable where kaolinisation of the strata occurs at the top of the Kirkwood Formation. The junction between the Lower Limestone Formation and the overlying Limestone Coal Formation is drawn at the top of the Top Hosie Limestone, the highest of a group of generally thin limestones characterised by the presence of Posidonia corrugata.

Limestone members

Broadstone Limestone

At the type locality, the quarries at Broadstone [NS 3615 5297] to [NS 3660 5334], the limestone is at least 5.20 m thick and consists of dark grey micrite with beds and laminae of calcareous mudstone. Borehole provings (Figure 11) show the Broadstone Limestone to be between 1.5 and 7.7 In thick.

Wee Post Limestone

The Wee Post Limestone lies close above the Broadstone Limestone and commonly comprises about 0.5 m of micrite. It is separated from the Broadstone Limestone by usually less than 1 m of seatclay, and may rest on a thin coal locally. The type locality is in the quarries at Broadstone [NS 3615 5297] to [NS 3660 5334].

Dockra Limestone

The Dockra Limestone is named from Dockra Quarry [NS 364 525] where it is over 6 m thick. Boreholes show that it range up to 13.7 m in thickness. The lithology varies laterally, and two distinctive facies can be identified. The 'Trearne' facies consists of pale bioclastic limestone, with argillaceous material confined to fine laminae and to the base of the limestone. It istypically developed at the south end of Trearne Quarry [NS 3729 5314]. The lugton' facies consists of dark grey argillaceous limestone, more variable in bioclastic content than the Trearne facies and containing beds of calcareous mudstone. It is typically developed at Lugton Quarry [NS 4108 5268]. The 'White Post', a distinctive lithology at the top of the Dockra Limestone, was described by Richey (1946) and has been recorded from Geirston [NS 2991, 5519], and from Trearne, Hessilhead, Middleton and Old Mill quarries.

Mid and Main Hosie limestones

The Mid and Main Hosie limestones are represented by a sequence of dark grey micritic limestones and calcareous mudstones. These are about 2.65 m thick at Paduff Burn [NS 3020 5490] and are separated from the Dockra Limestone below by about 2.4 m of mudstone.

Top and Second Hosie limestones

The Top and Second Hosie limestones are similar in Ethology to the Mid and Main Hosie limestones, and at Paduff Burn are about 1.4 m thick.

Descriptions of sections

The deposition of the Dockra Limestone marks the first marine transgression which extended virtually throughout the district (Figure 12), though in some small areas in the adjacent Kilmarnock district the Dockra Limestone was not deposited. This transgression marks a change in the overall pattern of sedimentation and so is a convenient horizon at which to divide the formation for description purposes.

Base of formation to base of Dockra Limestone

On the shore at Horse Isle [NS 2141 4256] and Cleeves Cove [NS 3178 4745] the Broadstone Limestone rests directly on tuffaccous mudstones of the Kirkwood Formation. The Wee Post Limestone is not visible on Horse Isle, and at Cleeves Cove it is either lost by faulting or has passed laterally into mudstones with beds of sandstone. A similar succession can be seen at low tide on the foreshore at Ardrossan [NS 2288 4168], where the Broadstone -Limestone again rests directly on tuffaceous mudstones. The thickness of individual units here is complicated by faulting but the Broadstone Limestone is overlain by about 4 m of massive quartzose sandstone with no development of a Wee Post Limestone.

Sections above the Broadstone Limestone at Broadstone [NS 3624 5404] and Langside Mine, and north and south of the Roebank Bridge [NS 3490 5580] and [NS 3493 5570], demonstrate lateral variation of the strata around the level of the Wee Post Limestone. The limestone, well developed at Broadstone, passes laterally into siltstones and mudstones at Langside Mine, 1 km to the east. A thin shelly lamina in the siltstones above the Broadstone Limestone (recorded here by Richey et al., 1930), may he the equivalent of the Wee Post Limestone. The localities at Roebank Bridge are only 100 m apart and demonstrate the replacement of the Wee Post Limestone northwards by sandstone and silty mudstone.

The more significant borehole sections are shown in (Figure 11). Though of differing thickness the Wee Post and Broadstone limestones have similar lithologies, which are maintained laterally. Generally dark or medium grey in colour, they are both argillaceous with beds of calcareous mudstone. Root structures are commonly preserved in the topmost beds of both limestones, and a nodular structure distorts the original bedding. Both limestones carry a marine fauna, with a rich variety of species in places (Chapter 10), but in the Wee Post Limestone in the Old Mill Borehole only algal structures were seen. The Wee Post Limestone is missing only in the Montgreenan Borehole where the entire Wee Post cycle is absent. It is assumed that the Wee Post transgression did not extend everywhere, and that in small areas coal or seatclay continued to develop while Wee Post Limestone was deposited elsewhere. Sandstones are recorded immediately above the Wee Post Limestone in the Ardrossan and Coalhill No. 2 boreholes. In the Dunniflat and Lora Burn borcholes the Dockra Limestone rests directly on the Kirkwood Formation. In the Montgreenan Borehole the Broadstone Limestone rests on 2.30 m (not bottomed) of medium- to coarse-grained sandstone, which is regarded as a sedimentary member of the Kirkwood Formation.

In the areas where the Broadstone Limestone is present, the Broadstone, Wee Post and Dockra limestones can generally be correlated with certainty. These correlations can also be made across major structures such as the Dusk Water Fault, and the thickness variations of the strata up to the base of the Dockra Limestone show no systematic change across the structures (Figure 11). It follows that the area north of the lnchgotrick Fault was tectonically inactive during this time, with little change in sedimentation rates across these more northerly faults.

Dockra Limestone

The Dockra Limestone represents the first marine transgression to have covered the whole area where the Lower Limestone Formation is known (Figure 12). The bed exists in two facies, called the 'Trearne' facies and 'Lugton' facies by Richey (1946). The 'White Post', a local development at the top of the Dockra Limestone, was also described by Richey (1946).

Trearne and Lugton facies

The Trearne facies consists of light to medium grey or brownish grey limestone, with argillaceous material confined to fine laminae. It was described by Richey (1946) as 'consisting of dominantly white or cream coloured limestones (though usually containing small dark patches) often in massive beds. It is also characterised by an abundance of well-preserved fossils, especially large encrinite stalks, many of them an inch in width, and medium-sized Productid shells'. The Lugton facies consists of dark grey argillaceous limestones, more variable in hioclastic content than the Trearne facies and containing beds of calcareous mudstone; it was described by Richey (1946) as 'marked by a succession consisting of dark grey limestone beds interleaved with dark grey calcareous shale'.

Richey's (1946, fig. 1) map showing the geographical distribution of the Trearne and Lugton facies has now been revised (Figure 13). At Auchenskeith [NS 3928 5258] the Dockra Limestone is dark grey and argillaceous and has been referred to the Lugton facies, not the Trearne facies to which it was assigned by Richey. An area of dark limestone of Lugton facies shown by Richey around Barr-mill has been omitted. This is because detailed field observations have shown that these dark grey limestones are in faulted contact with a small outcrop of Dockra Limestone of Trearne facies. The dark grey limestones are now considered to he Broadstone Limestone.

The lateral transition from Trearne to Lugton facies can be examined in Trearne Quarry and, less well, in Old Mill Quarry. The north end of Trearne Quarry, in Lugton facies,  consists of a series of posts of dark grey argillaceous limestone interleaved with beds of calcareous mudstone (Plate 5). Some of the limestone posts are light grey in colour and mottled, a lithology referred to by Shiells and Penn (1971) as 'multicomponent mudstones' and by Brown (1975) as 'mottled limestones'. Southwards the beds of calcareous mudstone thin until they become fine argillaceous laminae, though a muddy development remains towards the base of the limestone. The limestone lithologies also change southwards. Midway along the quarry, coarser hioclastic limestones appear resting on beds of fine-grained light grey mottled limestone (the 'multicomponent mudstones' of Shiells and Penn and the 'mottled limestones' of Brown). At the southernmost point in the quarry this type of limestone has disappeared by lateral passage into a succession of beds of light brownish grey hioclastic limestone with fine argillaceous laminae (Trearne facies). Low mounds or thickets of Siphonodendron junceum (Plate 6) usually 20 to 30 cm high are common. The 'reef limestone' shown by Shiells and Penn (1971, fig. 1A) has not been recognised, and beds of calcareous mudstone exposed in the floor of the quarry at its southern end are confirmed by borehole evidence as the basal beds of the Dockra Limestone.

The important features of the lateral facies transition are:

1. An increase in mud content from Trearne to Lugton facies, both within the limestone beds and as separate beds of calcareous mudstone.
2. Presence of light coloured bioclastic limestones only in Trearne facies.
3. Presence of 'mottled limestone' or 'multicomponent mudstone' lithologies within a transition zone.

The Dockra Limestone was deposited over a wide area (Figure 12), and contains a fauna characteristic of an open marine environment. Thickness variations of the limestone shown in (Figure 13) can be compared with the facies distribution given on the same diagram. Overall, the Dockra Limestone of Trearne facies is thicker than that of Lugton facies. Also, there is considerably more mud within the Lugton facies, which implies a greater degree of compaction. It is possible to estimate the variation in sea-bed topography by tracing an individual calcareous mudstone bed in the Lugton facies into a muddy lamina and thence into a bedding plane in the Trearne facies. It becomes apparent that at any time the difference in sea-bed level was not more than 1 or 2 m. The absence of even fine-grained clastic sediment from the Trearne facies suggests that some process was stopping the uniform spread of argillaceous sediment across the area. Brown (1975) discussed two possibilities, a carbonate shoal model and a biotic barrier model. The first envisages a shoal developing on a site rich in organisms, which builds on itself by being more attractive to organisms which require more light, better circulation and a freedom from terrigenous sediment. The end product would be a buildup of mud-free limestone within a central area. The second model is one where a localised abundance of crinoids traps the incoming clastic sediment, leaving areas starved of clastic detritus.

White Post            

The lithology of the upper part of the Dockra Limestonehas been modified locally to produce the White Post.

Reddening may occur in the top few centimetres of the White Post limestone, as in the Old Mill Quarry, and lenticular chert nodules may be present at around 2 m below the top surface, as in the Old Mill Borehole. The White Post is well exposed in the Paduff Burn 150 m north-east of Geirston Farm [NS 2991 5519] and in the Trearne, Hessilhead and Old Mill quarries, 3 to 4 km east of Beith. A White Post lithology has also been recorded from the top or near the top of the Dockra Limestone in the Old Mill, Muirlaught and Lora Burn boreholes, but none was recorded in the Dunniflat, Montgreenan, Ardrossan and Coalhill No. 2 boreholes. The distribution of the White Post in relation to the thickness and lithofacies of the Dockra limestone is shown in (Figure 13).

The occurrence of the White Post in the Paduff Burn at Geirston was recorded and commented on by Richey (1946). It consists of 1.20 to 1.60 m of fine-grained light grey limestone and rests on bedded dark grey argillaceous limestone of the Lugton facies. The junction between the White Post and the underlying limestones is irregular, with wedges of the White Post penetrating the under lying beds. These wedges have distorted the bedding in the underlying argillaceous limestone (Plate 7).

Thin sections from this locality show that the top 4 cm or so consist predominantly of a brownish grey, low birefringent clay mineral, which commonly forms discrete laminae. Reddish brown coatings of haematite occur around grains in the top 5 mm. Carbonate grains are scattered throughout the clay matrix and increase in proportion downwards into a dominantly carbonate rock below 4 cm. Quartz grains of detrital origin are also present throughout, as is a scattering of pyrite. An XRD scan of a bulk sample of this rock confirmed the quartz-calcite mineralogy and showed the dominant clay mineral to be kaolinite. The rock has been extensively bioturbated by roots or burrowing organisms. Foraminifera and other shell fragments indicate a marine origin. Below the top 4 cm the White Post consists of a massive light brownish grey micritic limestone with scattered shell fragments. In thin section the bulk of the rock is seen to consist of fine-grained calcite, with sparry calcite infilling open joints. Extensive bioturbation has occurred either by roots or burrowing organisms.

The White Post is also developed at the Old Mill Quarry [NS 3964 5277] and in the nearby Trearne and Hcssilhead quarries. A reddened crust around 5 cm thick marks the top of the White Post, here a bed of light. grey fine-grained limestone around 0.75 m thick. Beneath the White Post is a series of dark grey argillaceous limestone beds with layers of calcareous mudstone. Black lenticular chert nodules are developed in the top metre of these beds. Stigmarian roots are present within the reddened crust at IIessilhead and Old Mill. In thin section the reddened top of the Dockra Limestone at Old Mill Quarry can be seen to contain abundant rhombs of dolomite and patches of kaolinite. XRD determinations show the kaolinite to be well ordered. The fauna is characterised by ostracods and occasional algal structures, some of which form oncolites. The transition from the reddened crust to the White Post is irregular and frequently disrupted by roots.

The western face in the northern part of Old Mill Quarry shows the relationship between the White Post and the rest of the Dockra limestone (Plate 8). The White Post, with its reddened crust, is present at the north and south ends of the face and must cross-cut the bedding in the rest of the limestone. The White Post is therefore not a bed within the Dockra Limestone but an alteration of the top surface of the limestone: that surface is also a local disconformity, representing a period of some erosion of the limestone.

The key to the origin of the White Post lies in the recognition of this erosive surface and the reddened crust. These features are consistent with subaerial erosion and the finillation of a residual deposit rich in iron. Stigmarian roots within the White Post, and cavities in the reddened crust filled with well-ordered kaolinite, are also indicators of a subaerial environment in a warm climate. The White Post therefore formed by peclogencsis on a slightly uplifted area of Dockra Limestone. This process may also explain the occurrence of chert nodules sonic. way down the profile. Silica, dissolved in groundwater in an uplifted area of the limestone, would move clown the profile until it reached the water-table where it would be reprecipitated.

The level of the silica nodules, now seen in the limestone, might therefore represent the site of the water-table at that time. The subaerial weathering of the topmost beds of the limestone marks an upward transition from marine to terrestrial conditions. This is also evidenced by faunal changes in the top few centimetres of the Dockra Limestone at Old Mill. From a marine fltuna rich in brachiopods, crinoids and corals, the fauna alters to one dominated by ostracods and containing algal stomatolites (Plate 9).

Dockra Limestone to Top Hosie Limestone

The general lithological pattern is of mudstone and calcareous mudstone with limestone beds, as at Paduff Burn [NS 3020 5490] and Blair Mill Bridge [NS 3210 4750], and proved in several boreholes. Thickness variation is shown in (Figure 14).

Individual beds of limestone are seldom more than a metre thick. They are argillaceous, usually bedded though often bioturbated, and generally contain a brachiopod/crinoid fauna. The proportion of bioclastic material to matrix, and the content of material derived from different organisms, vary from bed to bed. Roots may penetrate the limestone, disrupting the structure.

Calcareous mudstones are similar to the limestones but are less bioturbated and have retained both bedding and fissility. The fauna present in the calcareous mudstones is similar to that in the limestones, though bryozoa seem to be confined to the former.

Non-calcareous mudstones differ from calcareous mudstones in several respects. They are dark grey rather than medium grey, and are usually better bedded, commonly weathering to give the appearance of 'paper shales'. Lingula is the most common fossil, but other marine organisms, including calcareous brachiopods and goniatites, have been recorded.

Strict bed-for-bed correlations between sections are not possible. The Top Hosie and Dockra limestones can be recognised by their place in the sequence and by their fossil content, but between these two horizons it is not possible to distinguish with certainty the four Hosie Limestones recognised elsewhere (Richey et al., 1930). However, correlation of sedimentary cycles is generally possible, and these may be recognised by the occurrence of medium- to coarse-grained sandstones and/or rooty beds. These are developed commonly at two horizons, one immediately above the Dockra Limestone, the other midway between the Dockra and Top Hosie limestones (a position correlated with the Hosie Fireclay of the Glasgow district).

In the Ardrossan Borehole a thick sandstone, coarse-grained in places, occurs just above the Dockra Limestone and is correlated with a thin rooty horizon at the same level in the Kirkwood, Coalhill No. 2, Muirlaught and Montgreenan boreholes (Figure 11). This horizon has not been recognised with certainty in the Old Mill, Dunniflat and Lora Burn boreholes.

The higher rooty horizon is recorded in the Kirkwood, Coalhill No. 2, Muirlaught, Montgreenan, Ardrossan and Lora Burn boreholes, and at outcrop in Paduff Barn and Blair Mill Bridge. It is represented in the Dunniflat Borehole by rooty sandstones, and by sandstone in the outcrop sections at Cunningham–Baidland [NS 275 513] and the Gurdic railway cutting [NS 334 524].

Sandstones at the same two stratigraphical positions have been noted also in old borehole data. The known lateral extent of these sandstone bodies is shown in (Figure 14). A marked north-easterly linear shape can be demonstrated for the sandstone in the position of the Hosie Fireclay but the geometry of the lower sandstone is less well constrained.

Lower Limestone Formation south of the Inchgotrick Fault

South of the Inchgotrick Fault, strata of the Lower Limestone Formation are known only in the Harelaw and Deaconhill boreholes. The sequences are different from those north of the fault, reflecting the role of the fault during and after sedimentation. The thickness of Lower Limestone Formation present is small: 6.80 m in the first borehole and 4.59 m in the second. The base of the formation is drawn at the base of the Broadstone Limestone, which in both boreholes rests conformably on strata of the Lawmuir Formation. The same boreholes showed that there is an unconformity at the top of the Lower Limestone Formation: in the Harelaw Borehole the strata above the Dockra Limestone are not present and in the Deaconhill Borehole the Dockra Limestone also is missing. This unconformity is discussed further in Chapter 8.

Chapter 8 Namurian: main part of Clackmannan Group

The Clackmannan Group consists of the Lower Limestone Formation, Limestone Coal Formation, Upper Limestone Formation and Passage Formation (Table 6). The Lower Limestone Formation is of Visean age and is described in Chapter 7. The remainder, described in this chapter, are largely of Namurian age, straddling the Pendleian to Ycadonian stages. A small thickness of strata at the top of the group may be early Westphalian in age. The main outcrop is in the northern part of the Irvine district, with smaller outcrops south of the Inchgotrick Fault. These areas are generally rolling farmland, with the lavas of the Troon Volcanic Member of the Passage Formation occupying relatively higher ground. Regional thickness variations are shown in (Figure 15).

Classification

A comparison of the succession in the Irvine district with that for the central part of the Midland Valley is given in (Table 10).

The Limestone Coal Formation consists of repeated cycles of sandstone, siltstonc and mudstone with seatclay and coal. The basal Kilbirnie Mudstone Member is not recognised outside Ayrshire. In the Upper Limestone Formation elastic cycles continue but marine limestones appear as an additional element. The Passage Formation is in three parts. The lowest is dominantly sandstone with minor proportions of other elastic rocks. Above, the Troon Volcanic Member is lithologically distinctive in that it consists of a uniform pile of basaltic lavas of Dalmeny type. The products of the lateritic weathering of these lavas form the Ayrshire Bauxitic Clay Member. The two formally named members are not found in other parts of the Midland Valley.

Limestone Coal Formation

The formation consists mainly of grey mudstones, siltstones and sandstones arranged in upward-coarsening cycles, with coals, seatearths and ironstones. The characteristic cyclical pattern of sedimentation, evident subjectively in many sections, has been analysed more objectively using transition matrix analysis of data from the Montgreenan Borehole (Monro, 1982a). The analysis confirms that the typical cycle shows a simple upward-coarsening trend, with seatcarth and coal resting on sandstone at the top of the cycle. The basal Kilbirnie Mudstone Member consists of mudstone with some ironstones.

The formation rests conformably on the Lower Limestone Formation, its base being drawn at the top of the Top Hosie Limestone. The top of the formation is drawn at the base of the Index Limestone or, if this is absent, at the erosive base of a sandstone below the limestone-bearing cycles of the Upper Limestone Formation. A generalised section for the area north of the Dusk Water Fault is shown in (Figure 16).

The thickness of the formation varies greatly within the area north of the Inchgotrick Fault, from 35 m in a slowly subsiding 'block' area between the Dusk Water and Annick Water faults, to 245 m or more in an adjacent basin to the north. The formation is absent, or very condensed, over a structural high to the south of the Inchgotrick Fault (Chapter 14). Some impression of these variations, and the way they relate to the faults, can be obtained from the isopachs shown in (Figure 17). The main marker beds within the formation can be traced across both the Dusk Water and Annick Water faults, and it is clear that the thickness differences were accumulating throughout the time interval concerned.

Marker bands

In attempting to correlate recent cored borehole sections with those proved in older boreholes it becomes apparent that the reliably correlatable horizons are not those defined solely on a lithological basis, such as the ironstones, but those which also carry a distinctive fauna. These faunas are generally marine, though commonly only Lingula is present. The most important marker horizons in the Irvine district are the Johnstone Shell Bed, the Maich Shell Bed, Logan's Bands and a Lingula band in the roof of the Smithy Coal (Figure 16).

The Johnstone Shell Bed is present throughout the Irvine district as a sequence of mudstones of variable thickness, usually less than 8 m, with a marine fauna. It can be correlated with little change northwards into the Glasgow district (Forsyth, 1978).

The Maich Shell Bed at its type locality in the Maich Water is about 4 m thick and has a fauna identical to that of the Linwood Shell Bed, its correlative in Glasgow (Wilson in Forsyth, 1978). The Linwood Shell Bed fauna is considered to have lived in brackish water, where only a few species could exist in abundance. The Maich Shell Bed is not easily recognised outside the type area and may be represented by a bed which only bears Lingula or Naiadites. This inference is based largely on borehole evidence, and while in the stream section at Maich Water a full Maich Shell Bed fauna was obtained, in the nearly Maich Water Borehole only Lingula squamiformis, Paracarbonicola cf. pervetusta and ?Naiadites sp. were found.

Logan's Bands are thin beds of ironstone. They occur in mudstones or silty mudstones which usually contain Lingula or nonmarine bivalves. The mudstones are of variable thickness and appear to be laterally persistent. They have been correlated with the Black Metals of the Glasgow district (Richey et al., 1930).

The Lingula band in the mudstone roof of the Smithy Coal is persistent laterally but may be absent, as may the Smithy Coal, in areas where sandstones are common.

Facies variation

A lithofacies map (Figure 18) constructed from borehole data shows variation in the ratio of sandstone to siltstone and mudstone for the Limestone Coal Formation. The map shows a general increase in mudstone from east to north-west, with a localised area between the Annick Water and Dusk Water faults having a high mudstone content. Other ways of contouring the data points are possible, and the apparent restriction of this high-mudstone area to the region between the faults may be an artefact of the method used.

Volcanism

Locally, in the area to the west of Dairy, tuffs and volcanic rocks are interbedded with the sedimentary strata. Richey et al. (1930, figs. 19 and 20) demonstrated the presence of tuffs in the Linn Pit [NS 2824 4906], the Bankhead Pit [NS 2704 4566] and the High Monkcastle Borehole [NS 2885 4781]. More recently, boreholes sunk to determine foundation conditions for development around Dairy have shown that the tuffs extend eastwards at least to the area of the town. In some of these bores thin ferruginous limestones up to 0.53 m thick are developed within the tuff layers. The stratigraphical position of all the tuff beds is not known, but one at least may occur some 10 m below the Borestone Coal. Bankend No. 3 Borehole proved an 8 m tuff bed below the Borestone Coal and a 1 m bed of tuff below Logan's Bands. It is not possible to analyse the geometry of the volcanic rocks because individual beds cannot be correlated laterally. However, a basic magma source is envisaged, with a vent centred on the thickest development of the tuffs, some way to the west of Dahy.

Environment of deposition

The coarsening-upwards clastic cycles characteristic of the formation can be interpreted in terms of repeated delta progradation into a shallow bay or lake. The mudstones represent. prodelta or lake-floor deposits, the main bulk of sandstones and siltstones represent the delta front, and the coals and seatclays, with local distributary channel sandstones, represent the delta plain (Monro, 1985, 1986).

Regional variations in the proportion of sandstone to siltstone and mudstone give an indication of the geometry of the delta (Figure 18), while regional thickness variations (Figure 17) suggest that contemporaneous fault movements affected the area across which it advanced.

The general fining of the sequence from the east to the north-west indicates an eastern source for the delta. This would imply that marine conditions were more prevalent in the north-west, a supposition which is borne out by changes in the Maich Shell Bed faunas.

Few palaeocurrent measurements are available for the Limestone Coal Formation. Good cross-bedding is preserved in a small quarry south of Lugton at [NS 4067 5228] where the mean of eight readings shows palaeocurrent flow to 230°, a direction compatible with the lithofacies interpretation.

Kilbirnie Mudstone Member

The Kilbirnie Mudstone Member is a series of mudstones with ironstones and occasional marine bands, and follows conformably on the limestone–mudstone sequence of the Lower Limestone Formation. The top is taken at the base of the first significant sandstone-bearing cycle. This boundary is diachronous, and the Johnstone Shell Bed, which is commonly developed within the Kilbirnie Mudstone Member, can be traced into areas where sandstones occur below it. The main outcrops are in the northern half of the district (Figure 2) and the thickness varies from 5 to 45 m within this area. A generalised section for the region north of the Dusk Water Fault is shown in (Figure 16). The Kilbirnie Mudstones are not present south of the Inchgotrick Fault, and may never have been deposited there.

Environment of deposition

The Kilbirnie Mudstones represent a transitional environment between the open marine conditions of the Lower Limestone Formation and the deltaic facies of the main part of the Limestone Coal Formation. Many of the mudstone beds, like the Johnstone Shell Bed, are marine in origin. The passage upwards into deltaic sediments is transitional, with sandstone lithologies becoming progressively more common. Marine horizons echo this transition, the variety of faunal species present becoming more restricted upwards.

Outcrop sections

Match Water

The Limestone Coal Formation section in the Maich Water is the most complete within the Irvine district. Exposures extend from a fault [NS 3260 5700] 300 m west-south-west of Langstilly Farm downstream to the Maich Bridge [NS 3321 5581]. Generally the rocks dip at around 8° to the south-east. The oldest strata, exposed at the northern end of the section, are correlated with the Johnstone Shell Bed on the basis of their fauna, principally brachiopods, and their position in the sequence. Some 28 m of Kilbirnie Mudstones are present. Sandstone beds begin to appear 250 m downstream from the outcrop of the Johnstone Shell Bed and thereafter the strata occur in repeated upward-coarsening cycles of mudstone, siltstone and sandstone. At Kirnhole at [NS 3272 5668] a bright coal 14 cm thick rests on a light brownish grey sandstone with a rooty top. This coal, locally called the Kirnhole Coal, is the lowest seam in the formation in this district.

It is succeeded by 2.6 m of mudstone coarsening upwards through 0.8 m of siltstone into 3 m or more of sandstone. Above, similar cycles, commonly incomplete and of variable thickness and lithological proportion, are developed downstream to a cliff section [NS 3287 5631] 280 m south of Nervelston Farm. At the base of the cliff, mudstones with two ironstones each 30 cm thick overlie mudstones containing a marine fauna, the type section of the Maich Shell Bed.

The sequence between the Maich Shell Bed and a fault trending at 120° and downthrowing to the south-west at[NS 3288 5610] consists of cycles of mudstone, siltstone and sandstone, some capped by thin coals. The thickness and proportion of each lithology within the cycle is variable and any part, particularly the coal, may be absent. A mudstone at least 4 m thick, cropping out in the stream 350 m south of Nervelston at [NS 3296 5620] contains thin ironstone bands and carries a fauna containing Lingula. This is a typical development of Logan's Bands.

Strata south of the fault lie in the top part of the formation. Sandstone, light grey with ripple lamination and larger-scale cross-bedding, occurs downstream to the railway viaduct [NS 3297 5588]. This bed is between 10 and 15 m thick and passes up into seatearth with a coal 20 cm thick, probably the Smithy Coal. The roof mudstone of this coal contains Lingula. Above the Smithy Coal further coal-bearing cycles are present but minor faulting precludes correlation of the seams. Exposure ends south of the Maich Bridge.

The section in the abandoned railway cutting between Jeffreystock Farm and West Lochridge Farm [from 3385 5710 to 3230 5508] is in the upper part of the formation. Small-scale faulting is common and sedimentary cycles coarsening upwards from mudstone and siltstone to sandstone are typical. Thin coals are often developed above the sandstones.

Paduff Burn

The section in the Paduff Burn begins 200 m north of Place at [NS 3028 5483], and consists of some 30 m of Kilbirnie Mudstones overlying the Top Hosie Limestone. The dip is generally to the south at 8°. Within this sequence two horizons can be identified: the Dairy Clay-band Ironstone [NS 3031 5478] and, slightly higher in the sequence, the Johnstone Shell Bed [NS 3035 5473]. The Dairy Clayband Ironstone is a medium grey argillaceous ironstone, 50 cm thick, identified solely on its stratigraphical position. The Johnstone Shell Bed is at least 2.20 m thick, consisting of mudstones with a marine fauna of brachiopods and bivalves.

The strata above the Johnstone Shell Bed crop out downstream as far as Kilbirnie Place Bridge [NS 3073 5442], and are faulted and intruded by dolerite dykes. This part of the sequence includes sandstones, with beds commonly over 4 m thick, which mark the transition into the main part of the Limestone Coal Formation. The strata occur in upward-coarsening cycles, some capped by thin coals. Immediately to the south of West Bankside Farm at [NS 3060 5462] the following section is seen: the coal may be the Kirnhole seam.

Between Kilbirnie Place Bridge and the centre of Kilbirnie [NS 3148 5450] the exposure is intermittent, in lithologies and sequences similar to those described above, though no coals have been observed. The section from a point [NS 3092 5436] 200 in downstream from Kilbirnie Place Bridge is:

The general nature of the sequence would suggest that these strata lie around the middle of the formation, but the section is too limited to allow correlation of the Lingula band with any of the other known occurrences of Lingula.

Powgree Burn

The section in the Powgree Burn extends from the weir [NS 3326 5241] eastwards to a point [NS 3420 5225] 250 m east of Brackenhills. The Kilbirnie Mudstones rest on the Top Hosie Limestone and consist of mudstones with occasional ironstone ribs and nodules including the prominent Dalry

Clayband Ironstone, which has been worked locally. The dip is to the south. At a railway viaduct [NS 3350 5230] mudstones containing a marine fauna including calcareous brachiopods and Lingula have been identified as the Johnstone Shell Bed. Eastwards from here exposure is intermittent, with sandstones and siltstones among the mudstones. A Lingula band is exposed 100 m south of Brackenhills Farm at [NS 3398 5214].

Pitcon Burn

The section in the Pitcon Burn begins [NS 2951 5308] 900 m south of Balgray and extends southwards to where it disappears beneath the alluvial plain [NS 2983 5244]. The section is highly faulted but displays similar lithologies and sequences to those described above. A tuff bed 1.50 in thick occurs among the sedimentary strata 400 m west of Swinlecs Burn at [NS 2979 5280]. Similar beds are also exposed in Hardcroft Burn [NS 2940 5183 and 2917 5180]. An exposure in a small stream 200 m west of Haw-hill at [NS 2890 5127] shows a highly altered lava resting on tuff.

Tuff beds also occur at Dykehead Bridge at [NS 2568 4757], in the disused railway cutting [NS 3310 4740] 550 m north-east of South Lissens where a 1 m band of coarse tuff is exposed, and 250 m south-west of Caddell at [NS 2696 4708] where rotten basaltic lava, 1 m thick, is overlain by sandstone.

Ardrossan shore

A very short section is exposed on the shore at Ardrossan [NS 232 419]. The Johnstone Shell Bed is occasionally exposed (depending on shifting sand) at the topmost eastern end of the section. Mudstoncs with ironstones below the Johnstone Shell Bed include beds of sandstone 2 to 3 in thick. The presence of sandstones indicates that here the top of the Kilbirnie Mudstones has to be taken below the Johnstone Shell Bed, emphasising the diachronous nature of the boundary. The southernmost end of the section includes strata which have been folded. This folding is significant, being on the projected line of the Dusk Water Fault.

Borehole sections

Montgreenan Borehole

This borehole was drilled in the area of reduced sedimentation (Figure 17). The full thickness of the Limestone Coal Formation, about 62 in, was proved. The basal Kilbirnie Mudstones were about 17 m thick, excluding dolerite intrusions, a figure similar to those given by old borehole records nearby.

Maich Water Borehole

This was drilled to establish the stratigraphical relationship between the Maich Shell Bed and the Johnstone Shell Bed in the area of increased thickness near Kilbirnie. It encountered 33 m of sandstone-bearing cyclic deposits above about 20 m of Kilbirnie Mudstones, but neither the top nor the base of the formation was proved.

Upper Limestone Formation

The Upper Limestone Formation, like the Lower Limestone Formation, is a cyclic sequence of elastic lithologies with marine limestones. It crops out mainly in the northern part of the district (Figure 2). The Index Limestone, which marks the base of the formation in the Midland Valley generally, is present over much of this area. In those places where the limestone is absent the base of the formation is drawn arbitrarily where coal-bearing cycles typical of the Limestone Coal Formation give way to cycles containing marine limestones. In these cases the boundary is normally taken at the erosive base of a coarse-grained sandstone. The Castlecary Limestone, defined as the top of the Upper Limestone Formation (MacGregor, 1960), has not been identified in the Irvine district. In consequence the top of the Upper Limestone Formation here is drawn arbitrarily where limestone-bearing cycles give way to seatearths and sandstones with occasional marine mudstones, an assemblage typical of the overlying Passage Formation. The boundary is usually drawn at the erosive base of a coarse-grained sandstone.

The main limestones in the formation are named in (Figure 19). All contain varied marine faunas, as detailed in Chapter 10. Most can be correlated confidently with limestones elsewhere in the Midland Valley (Cameron and Stephenson, 1985; Ramsbottom et al., 1978), the exceptions being the Diddup and Corsankell limestones.

Index Limestone

This bed is recognised in most parts of the Midland Valley. In the Irvine district it is generally about 3 m thick and comprises dark grey biomicrite and calcareous mudstone. It was formerly well seen in quarries at Highfield [NS 318 500] and has, in the past, been referred to as the High-field Limestone.

Third Post Limestone

This consists of about 3.5 m of dark grey biomicrite and calcareous mudstone. It crops out in the Putyan Burn, north-west of Dalry at [NS 2861 4975]. It correlates with the Lyon-cross Limestone of the Central Coalfield.

Lower Linn Limestone

Typically developed in the Caaf Water at the Lynn Bridge [NS 2870 4863], this comprises over 5 m of dark grey biomicrite, in places concretionary, with beds of calcareous mudstone. It correlates with the Orchard Limestone of the Central Coalfield.

Upper Linn Limestone

At Lynn Spout [NS 2828 4853] this comprises 14.7 m of dark grey micrite with beds of calcareous mudstone in the lower and upper parts and with fine argillaceous laminae throughout. It correlates with the Calmy Limestone of the Central Coalfield.

Diddup Limestone

This is a marine limestone, 2.90 m thick, occurring in the Diddup Borehole between 64.60 and 67.60 m from surface. It lies above the Upper Linn Limestone and below the Corsankell Limestone. It is exposed at outcrop in only one locality, 120 m south of Diddup Farm at [NS 2657 4440]. Its correlative outside the Irvine area is not known, though it may equate with one of the Plean Limestones (Chapter 10).

Corsankell Limestone

This is a marine limestone, 0.71 m thick, encountered between 55.85 and 56.56 m from surface in the Diddup Borehole, where it lies stratigraphically above the Diddup Limestone. No other occurrence is known. Its correlative outside the Irvine district is not known, though like the Diddup Limestone it may be one of the Plean Limestones.

Thickness

The thickness of the Upper Limestone Formation varies between about 15 and 175 m within the district. The greatest change of thickness appears to take place across the Dusk Water Fault, implying contemporaneous movement along that line. The isopach map (Figure 20), constructed using borehole data, has been contoured to take this inference into account. The presence of very thin Upper Limestone Formation sections south of the Inchgotrick Fault may imply some contemporaneous movement along that fault line also.

Lateral variations

The Diddup Borehole encountered six limestones within the Upper Limestone Formation: the Index, Third Post, Lower Linn, Upper Linn, Diddup and Corsankell limestones (Figure 19). In the other boreholes shown in (Figure 19) the Upper Linn, Diddup and Corsankell limestones were not present, and in the Kilmaurs Borehole the Index Limestone was also absent. The lateral variations proved by these recent cored boreholes can he plotted out over a larger area from old borehole records re-correlated on the basis of the more recent drilling.

The Index Limestone and associated marine mudstones are present over much of the northern part of the Irvine district and, with the exception of one occurrence, can be traced throughout the area north of the Annick Water Fault. South of this fault the Index Limestone is missing locally, the relevant part of the sequence being represented usually by coarse sandstone units with erosive bases.

Strata above the Lower Linn Limestone are missing over a much wider area of the Irvine district. The most complete sections are present north of the Dusk Water Fault, as in the Diddup Borehole (Figure 19). However, the relationship of the strata there to the adjacent outcrop sections, and particularly the presence of limestone clasts in a section where the higher marine strata are missing, suggest that there is an unconformity at the top of the Upper Limestone Formation. Between the Dusk Water and Annick Water faults the unconformity is identified by the absence of the Upper Linn Limestone and all the marine strata above that bed. Immediately to the south of the Annick Water Fault the same strata are again missing but the Upper Linn Limestone reappears in the south-west around Fenwick. The identification of the precise plane of unconformity in boreholes is difficult because the sandstones of the overlying Passage Formation also contain erosion surfaces. It is not known whether any of the missing strata were ever deposited in the area between the faults.

Only two recent boreholes, Harelaw and Deaconhill, have proved Namurian sequences south of the Inchgotrick Fault. Both boreholes proved attenuated successions, with only one or two thin marine horizons, to represent the Upper Limestone Formation.

Lithofacies

Three lithofacies can be identified within the Upper Limestone Formation. The marine lithofacies includes limestones and mudstones with a marine fauna. The upward-coarsening facies consists largely of elastic sediments (sandstones, siltstones and mudstones) arranged in upward-coarsening units. The upward-fining facies consists of elastic strata arranged in upward-fining units and in places includes coarse-grained erosive-based sandstones. Both elastic lithofacies contain seatclays and seat-earths with coals.

Marine lithofacies

This includes the principal named horizons in the formation, the Index, Third Post, Lower Linn, Upper Linn, Diddup and Corsankell limestones. Each of these has a different geographical distribution. All the limestones are dark grey and argillaceous but while those above the Lower Linn have no roof mudstones developed, the Lower Linn Limestone and those below generally have some thickness of marine mudstone above the limestone. In environmental terms the marine lithofacies indicates deposition in a shallow arm or bay of the sea.

Upward-coarsening lithofacies

The upward-coarsening lithofacies dominates the sedimentation pattern in the Upper Limestone Formation. The cycles which occur, like those in the Limestone Coal Formation, are thought to have been deposited in a deltaic environment. The mudstones represent deposition in the prodelta area while the siltstones and sandstones represent deposition at the delta front. Deposition on the delta plain is represented by the frequent presence of seatclays and coals. Sequences of this type are repeated by shifting of the delta lobes or by successive deltas building out as the area subsided.

Upward-fining lithofacies

The upward-fining lithofacies is less common than those described above but is locally important in the area south of the Inchgotrick Fault. North of the Inchgotrick Fault it is also known to be present, as between 34.90 and 38.00 m in the Hullerhirst Borehole and between 41.45 and 56.20 m in the Kilmaurs Borehole. In environmental terms this lithofacies generally represents the infilling of distributary channels within the delta lobe, though south of the Inchgotrick Fault the channel seems to have been cut into older marine sediments.

Cyclicity

An examination of the cyclicity in the Upper Limestone Formation (Monro, 1982a) has shown that within the elastic parts of each cycle the upward-coarsening pattern tends to be dominant. There is no clear relationship between the elastic and the carbonate parts of the cycle, suggesting that the cycle was deposited by more than one process. The elastic pattern suggests deposition in a pro-grading delta, as in the Limestone Coal Formation, and the marine strata must have been deposited as the delta was abandoned and flooded. Compaction, tectonic subsidence and eustatic rise in sea level are all likely to have contributed to the flooding process. Ramsbottom (1977) has suggested that deposition of the Lower Linn Limestone (equivalent to the Orchard Limestone) resulted from a eustatic rise in sea level.

Environment of deposition

The environment of deposition as it affects the pattern of sediment distribution can be considered during two periods, before and after deposition of the Lower Linn Limestone. This bed is the only marine horizon consistently present throughout the area (Monro, 1985, 1986).

Pre-Lower Linn Limestone

North of the Dusk Water Fault the sea transgressed on to a low-lying topography, commonly with coal swamps, and at first produced open marine conditions. As the deltaic coastline advanced into the sea, argillaceous strata, coarsening up to sandstone, were deposited. Delta lobe development culminated in coal swamp conditions on the delta plain. Three marine transgressions occurred, each marked by deposition of a limestone. Between the Dusk Water Fault and the Inchgotrick Fault a similar pattern is found, though slightly thinner, indicating some downthrow to the north on the Dusk Water Fault. In this area some distributary channels were formed which locally eroded underlying marine marker horizons. South of the Inchgotrick Fault only channel-fill deposits are present.

During the pre-Lower Linn Limestone period, therefore, contemporaneous movement on the Inchgotrick Fault, and to a much less extent on the Dusk Water Fault, did much to control the pattern of deposition. The Inchgotrick Fault probably marked the limits of marine transgression and bounded a block of uplifted older strata across which river channels were cut. Deltas built out from the coast north-westwards, but distributary channels probably did not extend beyond the Dusk Water Fault.

Post-Lower Linn Limestone

Following deposition of the Lower Linn Limestone the pattern of sedimentation north of the Dusk Water Fault line was again deltaic, with up to three marine horizons preserved. The two highest of these were commonly removed by erosion and replaced by channel-fill sandstones of the Passage Formation. South of the Dusk Water Fault channel-fill deposits probably all belong to the Passage Formation (Figure 19). South of the Inchgotrick Fault the Upper Limestone Formation is very thin, and was often raised above sea level, so that pedogenesis produced a series of seatclays and seatearths.

Outcrop sections

Outcrop sections through these strata are not common in the Irvine district. Such as do occur are interrupted either through poor exposure or by faulting. The outcrop data are, however, valuable for comparing with subsurface data from the many boreholes in the district.

Care Water

The section in the Caaf Water extends from a point [NS 2910 4850] 120 m west of Caaf Bridge to Pinnoch Point [NS 2820 4822] and includes most of the strata above the Lower Linn Limestone. This last crops out 120 m west of Caaf Bridge, at the Lynn Bridge [NS 2870 4863] and at West Lynn [NS 2860 4872]. The limestone is over 5 m thick, consisting of dark grey, argillaceous, occasionally concretionary limestone with calcareous mudstone layers. A marine fauna is present throughout.

The strata between the Lower and Upper Linn limestones, although not continuously exposed, are seen to consist largely of sandstones with subordinate mudstones and siltstones. The Upper Linn Limestone crops out in Lynn Quarry [NS 2843 4857] and at Lynn Spout [NS 2828 4853]. It consists of 14.7 m of dark grey argillaceous limestone, with calcareous mudstone layers at the top and towards the base, and with thin muddy partings throughout. The limestone carries a marine fauna and is underlain by mudstones with the bivalve Edmondia pundatella.

Above the Upper Linn Limestone, the strata consist of thick cross-bedded sandstone units with subordinate siltstones and mudstones. Downthrown against these, to the west of a fault at Pinnoch Point, are sandstones, seatearths and seatclays, lithologies more typical of the Passage Formation.

Putyan Burn

Sandstones below the Third Post Limestone are the lowest strata exposed in the Putyan Burn. The Third Post Limestone crops out twice, 180 m east of Broadlie Cottage at [NS 2841 4975] and 210 m north-west of North Kirkland at [NS 2861 4975], in an anticlinal structure. It consists of dark grey argillaceous limestone and calcareous mudstone, 3.50 m thick, with a marine fauna. Above the Third Post Limestone the exposures in the stream are dominantly of sandstone.

The strata from the Lower Linn Limestone to the Upper Linn Limestone crop out between a point 80 m north of Broadlie Cottage at [NS 2828 4978] and Broadlie House at [NS 2782 4996]. The sequence is similar to that in the Caaf Water. The Broadlie Coal, no longer exposed, occurs below the Upper Linn Limestone and was formerly worked.

Monkcastle Burn

The Lower Linn Limestone crops out in the core of an anticlinal structure in the burn at Monkcastle Bridge [NS 2933 4729]. The Upper Linn Limestone is exposed to the west by Monkcastle at [NS 2911 4735] and where the Monkcastle Burn enters the Garnock Water [NS 2962 4724]. The limestone lithologies and thicknesses, and the nature of the intervening strata, are similar to those in the Caaf Water.

Diddup Burn

The youngest exposed strata of the Upper Limestone Formation are seen at outcrop in the Diddup Burn. The Lower Linn Limestone crops out 330 m north-north-east of Diddup Farm at [NS 2650 4455]. It is a dark grey medium-to fine-grained limestone, dipping to the south-east at 25°. Above the Lower Linn Limestone are occasional outcrops of sandstone, with the Upper Linn Limestone cropping out in a quarry 240 m north-east of Diddup Farm at [NS 2670 4469] and by Diddup Farm itself at [NS 2658 4448]. The Upper Linn Limestone is a brownish grey limestone over 3 m thick, with a marine fauna, and rests on 1.00 m of dark grey silty mudstone, 0.70 m of medium-grained sandstone, 0.15 in of dark grey siltstone to silty mudstone and 0.15 m of coal.

Above the Upper Linn Limestone the strata are generally sandy. A thin iron-rich limestone crops out 140 m south of Diddup Farm. This horizon is tentatively correlated with the Diddup Limestone of the Diddup Borehole. Above the limestone are intermittent exposures of sandstone which dip to the north-west, as opposed to the south-east dips in the northern part of the section. It is suspected that a fault occurs immediately to the south of the Diddup Limestone.

Glen Burn

In the upper part of the Glen Burn isolated exposures of the Third Post Limestone [NS 2602 4405] and the Lower Linn Limestone [NS 2605 4404] occur with some sandstone between. The strata dip to the south-east at around 40°, and about 50 m downstream from the Lower Linn Limestone the Upper Linn Limestone crops out [NS 2607 4494]. The Upper Linn is a dark grey argillaceous limestone with a pronounced joint pattern, and rests directly on sandstones. About 20 m downstream from the Upper Linn Limestone and at a higher stratigraphical level is a bed of breccia with angular fragments of limestone, sandstone and siltstone. No limestones are found in this section at a stratigraphical level higher than the Upper Linn Limestone and the limestone fragments in the breccia are thought to be eroded remnants of the Diddup Limestone. This breccia, and the sandstones and siltstones above, belong to the Passage Formation.

Ardrossan Harbour

At Ardrossan Harbour a major north-westerly fault down-throws to the south-west. On the downthrow side, 40 m north-west of Castle Craigs at [NS 224 417], is a sequence of strata lithologically comparable to the Upper Limestone Formation. The limestones are correlated tentatively with Upper Linn, Lower Linn and Third Post limestones. At the top of the Upper Linn Limestone a lateral change in lithology, from limestone to sandstone, can he observed locally.

The details are:

Lugton Water

The section in the Lugton Water extends from 300 m east of Bridgend of Montgreenan at [NS 3490 4558] downstream to 460 in west of Montgreenan Cottage at [NS 3422 4507]. The Index Limestone is exposed at the base of the section east of Bridgend of Montgreenan. It consists of 2 m of dark grey calcareous mudstone and 1.1 m of dark grey argillaceous limestone on dark grey mudstone, and carries a marine fauna. Between the Index Limestone and the Third Post Limestone at Montgreenan Bridge, the strata are largely sandstones with subordinate siltstones, the sandstones becoming rooty upwards.

The Third Post Limestone is exposed at Montgreenan Bridge at [NS 3472 4534] and at various points in the Lugton Water westwards for 470 m, the general trend of the valley being along the strike. It is a medium to dark grey limestone, about 1.1 m thick with calcareous mudstone laminae. The strata immediately above and below are largely sandstones with subordinate siltstones.

A north-south-trending fault 470 m west of Montgreenan Bridge at [NS 3425 4532] throws older strata up against the Third Post Limestone. Immediately west of the fault the strata around the Third Post Limestone are poorly exposed, those above being generally silty sandstones. The Lower Linn Limestone crops out as a hard, dark grey limestone with pronounced rectangular jointing, 460 in west of Montgreenan Cottage at [NS 3422 4507].

Borehole sections

The principal features of the sequences encountered in four cored boreholes, drilled in the northern part of the district at Diddup, Hullerhirst, Kilmaurs and Montgreenan, are shown in (Figure 19). Two more cored bore-holes drilled at Harclaw and Deaconhill, south of the lnchgotrick Fault, also proved sections through the Upper Limestone Formation.

Diddup Borehole

This borehole proved the most complete sequence known in the Irvine district. It is dominantly sandy, but marine limestones and calcareous mudstones also occur. The Index Limestone is about 1.8 m thick, the Third Post Limestone is 4.4 m thick, the Lower Linn Limestone is 9.5 m thick, the Upper Linn Limestone is 15.5 in thick, the Diddup Limestone is 2.9 m thick and the Corsankell Limestone is 0.7 m thick. The top of the formation is drawn at 53.70 m depth, where the lithology passes up into sandstone and siltstone assigned to the Passage Formation.

Hullerhirst Borehole

The site of the Hullerhirst Borehole is about 2 km south-south-east of the Diddup Borehole but the sequence is markedly different (Figure 19).

The Index Limestone is 0.61 in thick and a sequence of marine strata 2.15 in thick, at 40.65 in depth, is correlated with the Third Post Limestone. Above, at 33.00 in depth, 4.45 in of dark grey argillaceous limestone, with beds of calcareous mudstone and a roof consisting of silty mudstones with a rich marine fauna, is correlated, partly on palaeontological grounds (information from R Wilson, 1981), with the Lower Linn Limestone. The top of the formation is drawn at 23.77 m depth, below a sequence of sandstones and seatearths assigned to the Passage Formation.

Montgreenan Borehole

The Index Limestone consists of 2.24 m of marine calcareous mudstone and nodular argillaceous limestone. The Third Post Limestone consists of 1.38 in of marine argillaceous limestone and calcareous mudstone overlain by silty mudstone with a rich marine fauna. The Lower Linn Limestone consists of 1.7 m of dark grey argillaceous limestone and calcareous mudstone with a marine fauna. A rich marine fauna is present also in the silty mudstones of the roof. The top of the Upper Limestone Formation is drawn arbitrarily at 9.70 m depth, at the base of a sandstone unit above which Passage Formation lithologies prevail.

Kilmaurs Borehole

The Kilmaurs Borehole started immediately below the Troon Volcanic Member of the Passage Formation, as evidenced by an old quarry in basaltic lavas about 10 in to the south of the site. The Index Limestone is not present in this section and the base of the Upper Limestone Formation is taken arbitrarily at the erosive base of a coarse-grained sandstone at 56.20 m depth. The Index Limestone may have been removed by penecontemporaneous erosion. The Third Post Limestone is a 1.12 m thick bed of medium dark grey argillaceous limestone with a marine fauna. The strata between the Third Post and Lower Linn limestones are sandstones fining downwards into silty mudstones and mudstones. The Lower Linn Limestone comprises 1.7 m of medium to dark grey calcareous mudstone and limestone with a rich marine fauna which extends up into the overlying mudstones. The top of the formation is drawn arbitrarily at the erosive base of a coarse-grained sandstone at 13.67 in depth, above which the strata have all the lithological characteristics of the Passage Formation.

Harelaw Borehole

This started in the Troon Volcanic Member of the Passage Formation and finished at a depth of 69.76 m in the Ballagan Formation. The sequence is quite different from that in the northern part of the Irvine district, the Upper Limestone Formation being reduced to just one marine limestone and the strata below consisting almost entirely of coarse-grained sandstones. Faunal evidence is insufficient to identify the range of strata represented. The base of the coarse sandstones is a marked erosion surface, and such surfaces are also present within the sandstone body. The strata below the sandstones are some way down into the Lower Limestone Formation and the base of the sandstones may therefore represent a significant time break.

Deaconhill Borehole

This started immediately below the Troon Volcanic Member of the Passage Formation and finished at a depth of 323.46 m in lavas of early Devonian age.

The Upper Limestone Formation is thin here also, being confined to two marine horizons. Faunal evidence is not conclusive (information from R B Wilson, 1982) but the upper horizon is possibly the Upper Linn Limestone which the lower horizon may be the Lower Linn. The sequence below the limestones is finer-grained than that in the Harelaw Borehole, containing mudstones and rooty beds. A sharp junction with underlying strata occurs at 40.60 m depth. The faunal evidence from a limestone at 44.40 m suggests a correlation with the Broadstone

Limestone (information from R B Wilson, 1982). Therefore, the basal strata of the Upper Limestone Formation, all the Limestone Coal Formation and much of the Lower Limestone Formation have been eroded off here, or were never deposited.

Passage Formation

In the Irvine district the Passage Formation is divided into three parts: a sandstone-dominated member at the base, the Troon Volcanic Member in the middle and the Ayrshire Bauxitic Clay Member at the top. The thickness varies between about 30 and 180 in in the district.

Sandstone-dominated member

The lowest part of the Passage Formation consists of sandstones with subsidiary siltstones, mudstones and seat-earths, an assemblage similar to that which characterises the whole formation in most other parts of the Midland Valley. The outcrops are in the northern part of the district (Figure 2), where four cored boreholes have proved typical sections (Figure 19). The thickness here ranges from 5 to about 20 in. The base of this member is drawn arbitrarily where the limestone-hearing cycles of the Upper Limestone Formation pass up into a sandstoneseatearth sequence. The line is commonly drawn at the erosive base of a sandstone. The top of the member is drawn where lavas of the Troon Volcanic Member overlie the sedimentary strata.

There is one named subdivision, the Douglas Fireclay, which comprises about 6 m of seatclays and seatearths, some with high-alumina characteristics.

Environment of deposition

The borehole evidence suggests that the strata thicken northwards, with lateral facies changes, across the Dusk Water Fault. The sequence of the Diddup Borehole seems to continue the marine to deltaic pattern of sedimentary environments established in the Upper Limestone Formation, though the marine strata are less common and much thinner. The pattern of sediment distribution northwards to the Monkcastle area can be interpreted in a similar way, though seatbeds are much more common.

South of the Dusk Water Fault the thinner developments may also reflect deltaic sedimentation with occasional marine transgressions. The strata in the Kilmaurs Borehole are different, however, consisting of sandstones fining upwards from an erosive-based pebbly basal unit. These strata probably represent the infilling of a distributary channel.

Sections north of the Dusk Water Fault

The sandstone-dominated member is recorded in the Diddup Borehole. The sequence is 25.4 in thick, consisting largely of sandstone with beds of siltstone and mudstone (Figure 19). A mudstone at a depth of 50.40 m contains some shell traces. Towards the top of the sequence roots become more abundant and seatearths are present. Palynological investigation of samples from the Diddup

Borehole (B Owens, written communication, 1980) show that the strata immediately below the lavas lie in the KV miospore zone, so are of Kinderscoutian to early Marsdenian age. North of Diddup, in the Monkcastle area, horeholes drilled to prove fireclay show that seatclays and seatearths, with some fireclay, dominate the sequence. Fireclay is currently (1994) being extracted in the area. Additionally, thin coals and beds of mudstone with Lingula make up a total thickness only slightly less than that in the Diddup Borehole.

In outcrop, the sequence in the Caaf Water at Pinnoch Point [NS 282 482] is similar to that developed in the Monk-castle boreholes. Sandstones, seatclays and seatearths occur with up to four thin marine mudstone bands (Richey et al., 1930). Southwards towards Diddup seat-clays and seatearths are less abundant and in the stream section at Glen Burn [NS 2608 4395] the outcrop is largely sandstone with beds of siltstone. An intraformational breccia marks the base of the Passage Formation here (see above).

Sections south of the Dusk Water Fault

The basal sandy member was proved in boreholes at Hullerhirst, Montgreenan and Kilmaurs (Figure 19). The sequence in the Hullerhirst Borehole is lithologically similar to that at Diddup, though only 12.57 m thick. It is dominated by sandstones with occasional thin beds of mudstone, one of which, at 16.12 m depth, has a poorly preserved marine fauna. Rooty horizons occur towards the top of the member. In the Montgreenan Borehole some 5.15 m of strata immediately below the Troon Volcanic Member are assigned to the basal sandy member. They consist of fine-grained sandstones with mudstone layers, generally greenish grey in colour and with red and yellow mottling. The Kilmaurs Borehole shows an upward-fining series of sandstones with occasional thin siltstones and a pebbly basal sandstone resting erosively on mudstone at 13.67 m depth. No fossiliferous beds occur.

The basal sandy member is poorly exposed south of the Dusk Water Fault except for a short section in the Lugton Water [NS 3380 1485] where the Troon lavas rest on about 2 m of cross-bedded sandstone which pass down into 3 m of dark grey mudstone. South of the Inchgotrick Fault the Troon lavas rest directly on the Upper Limestone Formation: the basal sandy member is absent.

Troon Volcanic Member

The Troon Volcanic Member is recognisable over an area that extends from Ayrshire south to Stranraer and west to Arran, Kintyre and possibly to Northern Ireland (Richey et al., 1930, fig. 25). It consists generally of basaltic lavas but both borehole sections and outcrops show that the sequence can be interbedded with sedimentary rocks. In the Diddup Borehole the latter have yielded a miospore flora from the KV zone (B Owens, written communication, 1980) indicating an age within the Kinderscoutian to early Marsdenian stages of the Namurian. A minimum K-Ar age for the Troon Volcanic Member is given by De Souza (1979; 1982) as 305 ± 6 Ma. Other published ages are given in (Table 11).

The lavas are generally much decomposed and altered both in surface outcrops and in horeholes. They have a characteristic red speckled appearance due to varying amounts of ferruginous alteration. Advanced decomposition produces pseudo-stratified greenish blue clays with only residual pseudomorphs after plagioclase visible. Much of the alteration is thought to be due to penecontemporaneous weathering, similar to that which gave rise to the lithologies of the overlying Ayrshire Bauxitic Clay.

The base of the member is marked by the incoming of lavas above the basal sandy unit, and its top is taken at the change from basalt to bauxitic clay; this junction is normally gradational. The isopach map for the volcanic member in Ayrshire (Figure 21) is based on data assembled by Richey et al. (1930) and modified to take account of more recent drilling. The lavas form a general dome structure with a maximum thickness of about 160 m just north of Troon; the trends of variation suggest contemporaneous movements along the Inchgotrick and Dusk Water faults.

Petrography

The Troon lavas are almost all olivine-basalts similar to the Dalmeny type of the Visean lavas (Table 9). Microphenocrysts of olivine pseudomorphed by dark red 'iddingsite' are conspicuous in all samples and black clinopyroxene is present in some. Microphenocrysts of plagioclase occur in only a few flows. The dimensions of the olivine microphenocrysts (0.5 to 2 mm) and of the groundmass plagioclase (0.5 to 1 nun) are on average greater than in the Dinantian lavas. The augite is commonly subophitic and/or in glomeroporphyritic aggregates, and rod-like iron-titanium oxide is common. Interstitial patches of orange-red or green isotropic material occur in many samples. More detailed petrographical descriptions are given by MacGregor (in Richey et al., 1930; in Eyles et al., 1949).

Analyses of fresh lavas (including six from the Irvine district) have been published by Macdonald et al. (1977), and (Table 12) lists 18 analyses from the district by Wallis (1989) plus one by Richey et al. (1930). The lavas form a 'transitional' basalt series which straddles the critical plane of silica undersaturation, almost equal numbers of available analyses being silica saturated (hypersthene normative) and silica undersaturated (nepheline normative). Only one analysis has sufficient normative nepheline (6.3 per cent) to be classed at a basanite. The compositional range is illustrated by a plot of plagioclase composition against differentiation index (Figure 22). This shows a clear contrast with the plot for the Visean lavas (Figure 8).

The source of the Troon lavas has not been identified and is assumed to lie beneath the lavas themselves, or else offshore. Geochemical studies (Macdonald et al., 1977) show that they do not fit into the trend of increasing silica undersaturation established in the Visean and early Namurian; they initiate a second cycle of increasing silica undersaturation which extends into the Permian. More detailed geochemical studies, and discussion in the context of Carboniferous to Permian igneous activity in south ern Scotland as a whole, are to be found in Macdonald et al. (1977), Macdonald (1980) and Wallis (1989).

Sections

Exposures are uncommon as much of the outcrop area is covered by Quaternary deposits.

North of the Dusk Water Fault

The principal section in this part of the district is in the Glen Burn [NS 261 439], where the total thickness of the member is about 60 m, including intercalations of sedimentary rock. The bottom 25 m of the member were cored in the nearby Diddup Borehole and included 12 m of sedimentary strata, mostly sandstone and seatearth. Outcrops around High Smithstone [NS 281 457] and Laigh Smithstone [NS 287 467] also show the lavas intercalated with sedimentary strata including thin beds of tuff.

South of the Dusk Water Fault

The topmost few metres of the Troon Volcanic Member are exposed on the coast [NS 239 413] at Saltcoats and have been proved in boreholcs at Stevenston and Kilwinning. They are also exposed in the Annick Water around Cunninghamhead Mill [NS 3745 4225] where the basalts are highly decomposed. A full sequence through the lavas is exposed in the Lugton Water east of Sevenacres Mains [NS 3320 4470], but the total thickness here is only about 10 m and the lavas are highly decomposed. Relatively fresh lavas are seen in the quarries at Auchenharvie Castle at [NS 363 443] and Rashillhouse at [NS 3828 4330]. To the south, in the Barassie area, the Troon Volcanic Member is exposed at the Stinking Rocks [NS 322 334]. Neither the base nor the top of the member is exposed here but the thickness was proved to be about 175 m in the Gailes Borehole nearby.

South of the Inchgotrick Fault

The member crops out from Symington to the edge of the Irvine district at Stone Calsey [NS 408 325]. Throughout this area there are small isolated exposures of decomposed, commonly amygdaloidal basalt; the volcanic rocks rest directly on the Upper Limestone Formation.

Ayrshire Bauxitic Clay Member

The Ayrshire Bauxitic Clay is best developed along the northern side of the Coal Measures outcrop, extending from the coast at Saltcoats [NS 24 41] eastwards to Cunninghamhead [NS 37 42], resting everywhere on the Troon lavas. In the southern part of the district the member is only sporadically developed. It is absent in the Troon area, where the Coal Measures rest directly on the Troon lavas, but south of the Inchgotrick Fault it is developed around Symington. The thickness varies from 0 to about 20 m. The top of the member is drawn at the lithological change from bauxitic clay to the elastic cycles of the Coal Measures. There is no marine marker band at the boundary in this district.

The distinctive bauxitic clay lithology was first described by John Smith (1895a), though it was not until 1922 that a detailed description was published by G V Wilson. Wilson emphasised the massive structure, conchoidal fracture, unusual density and induration of the bauxitic clay, and the variety of textures ranging from fine-grained to oolitic, pisolitic and coarsely elastic. He also noted that the rock is 'bauxitic in the chemical sense of containing more Al₂O₃ than can be accommodated with SiO₂ to give kaolinite'. From its association with altered basaltic lavas he concluded that it developed partly as a residual crust on the lavas and partly as a transported sedimentary deposit derived from erosion of that crust. He believed that the bauxitic clays were later subjected to diagenetic changes involving silicification of some of the bauxitic minerals to form a second generation of kaolinite.

Subsequently, De Lapparent (1936) identified the aluminium oxyhydroxide minerals boehmite and diaspore in selected samples. Although these proved to be very much subordinate to kaolinite, he agreed with Wilson's use of the term bauxitic clay. A number of authors (Bosazza, 1947; Keller, 1967; Robertson, 1971; Loughnan, 1978) have drawn attention to similarities between the bauxitic clay and the flint clays of North America, South Africa, Australia and elsewhere, and have suggested that the Ayrshire clay is in effect a typical flint clay.

The variation in sedimentary pattern and thickness of the member is evident from the sections shown in (Figure 23). At Smithstone [NS 279 456], north of the Dusk Water Fault, alternations of bauxitic clay and coal are developed up to a thickness of around 20 m. Large masses of tree trunks, Sidullaria sp., occur in the bauxitic clay, and there is no evidence to suggest that the coals were ever covered by hot lava flows or ash falls. A single marine band is locally preserved at the base of the bauxitic clay in this area. South of the Dusk Water Fault, at Dubbs [NS 280 423], the member is considerably thinner, with a bauxitic clay unit of around 4 m. The transition at the base of the member, from lava to bauxitic clay, is never sharp and is usually marked by a transitional clayrock rich in sphaerosiderite. In this respect the section at Sevenacres [NS 334 446] is typical of a large part of the outcrop. Other lithologies may be interbedded. In Torranyard No. 6 Borehole, for example, a bauxitic clay bed occurs within the basal clay-rock, while at Annick Water [NS 372 423] mudstones with plant fragments and a seatclay occur within the bauxitic clays.

A petrographical, chemical and mineralogical study of the core of the Torranyard No. 6 Borehole (Monro et al., 1983) gave the following results.

Petrography

The basal clayrocks contain microlites of kaolinite aligned parallel to the bedding, and clasts with original volcanic textures. The lower bauxitic clay contains abundant ooliths and pisoliths. The upper clayrocks also contain microlites of kaolinite aligned parallel to the bedding but do not contain clasts with original volcanic textures. The upper bauxitic clay contains coarse angular clay clasts resembling the underlying clayrocks and is rich in ooliths and pisoliths towards its top.

Chemistry

Some mobile elements, namely sodium, potassium and yttrium, are depleted in the bauxitic clay, the basal clay-rocks and the altered basalts beneath. Other mobile elements, namely calcium, magnesium, ferrous iron, rubidium, barium and strontium, show an erratic trend through the profile. Some immobile elements, namely zirconium and niobium, are concentrated to the same degree as aluminium although both are somewhat enriched in the bauxitic clay layers. Other immobile elements, namely titanium, chromium and vanadium, show relative depletion within the bauxitic clay layers.

Mineralogy

Neither quartz nor illite occurs in the bauxitic clay or clavrocks. Siderite is the most abundant ferruginous mineral. Kaolinite is the dominant constituent of the bauxitic clay, clayrocks and altered basalts.

In a recent review of the titanium oxide content of the bauxitic clays of Ayrshire, Cameron (1980) quoted a wide range of values from 1.2 per cent to 14.17 per cent. The higher values are above the levels that would he expected in a residual deposit of this type, and Cameron suggested that they might be clue to reworking of a residual deposit and the localised formation of titanium-bearing heavy mineral placers. Such a mechanism might also account for enrichments in other immobile elements, and the corresponding depletions.

Origin of bauxitic clay

Wilson (1922) recognised two modes of occurrence: autochthonous bauxitic clay, formed by in-situ chemical weathering of basaltic lavas; and allochthonous bauxitic clay formed as a true sedimentary deposit on the floors of shallow lagoons.

Study of the Torranyard Borehole core (see above) and observations of the Ayrshire Bauxitic Clay throughout the district highlight a number of features which are inconsistent with in-situ development.

These features are: the interbedding of bauxitic clay and unaltered coal at Smithstone; the presence of a marine band at the base of the bauxitic clay at Smithstone; the presence of laminated mudstone with plant remains beneath bauxitic clay at Annick Water; the presence of randomly orientated clasts consisting of parallel-aligned microlites of disordered kaolinite within the bauxitic clays of the Torranyard Borehole; and the presence of anomalously high or low amounts of immobile elements.

On the other hand, the nature of some of the lithological junctions supports the idea of in-situ development: the bauxitic clay generally grades down into clayrock which in turn grades into altered basalt. However, in-situ development can only be assumed where there is no evidence that former sharp junctions have been obscured by later lateritic weathering or diagenesis.

On this basis, the batixitic clay north of the Dusk Water Fault is regarded as largely allochthonous in origin, though some of the basal clayrock lithologies may have developed by in-situ weathering of the underlying basalts. South of the Dusk Water Fault the outcrops are more extensive, the profiles more varied and the conclusions less positive. Here the basal clayrocks which grade into altered basalts may be an in-situ development but, as at Torranyard, there are sections which are largely aliochthonous in origin.

Environment of deposition

The presence of bauxitic clays is a significant factor in the interpretation of the late Namurian environment. Laterites, of which bauxitic clay is a specific type, are residual deposits and as such require significant time to form in any thickness. Their occurrence therefore indicates a period when there was little or no sediment input. Laterites only form in tropical climates, so their presence also shows that the British Isles lay in low latitudes at that time.

Despite the variation in the thickness of the underlying lavas, the topography of the lava surface was low, as only local reworking of the weathered crust has taken place. The proximity of an area of sea is indicated by the presence of marine bands, though their local distribution suggests that the limits of incursions are sinuous. The ooliths and pisoliths, which Wilson took to indicate the existence of shallow bodies of water, are now known to form within lateritic profiles. However, the reworking and accumulation of weathered material in some parts of the Irvine district, and the presence of log mats of Sidullaria sp. at Smithstone, suggest that areas of standing water were present.

Sections

There are few wholly natural sections into the Ayrshire Bauxitic Clay, and quarries and former workings are the main source of outcrop information. Natural sections are accessible at Saltcoats, at Sevenacres and in the Annick Water, however.

On the shore at Saltcoats [NS 2401 4150], the section into the bauxitic clay is approximately 1.2 to 1.5 m thick and consists of a massive, light grey to buff-coloured clayrock with steep joints. Dark grey plant scraps are common and ooliths and pisoliths are present. Sphaerosiderite is common and varies in abundance through the bed. There is a gradational passage into an underlying clayrock which further grades into basaltic lavas of the Troon Volcanic Member.

In the Lugton Water at Sevenacres [NS 3353 4450], the section is typical of the succession south of the Dusk Water Fault. There is some interdigitation of sedimentary rocks within the Troon Volcanic Member.

In the Annick Water at Cunninghamhead [NS 3728 4220] the Ayrshire Bauxitic Clay is exposed and here consists of over 3 m of bauxitic clay, generally grey-brown with darker carbonaceous beds. A black laminated mudstone of variable thickness from a few centimetres to about a metre, and containing plant material, has been recorded from this locality (Wilson, 1922).

At High Smithstone, a section into the Ayrshire Bauxitic Clay is seen beneath about 2 m of sandstone in a quarry. It is typical of the thick sequence developed north of the Dusk Water Fault but is incomplete in that the underlying

Troon lavas are not exposed. The section in the quarry face is as follows:

Chapter 9 Westphalian: Coal Measures

Introduction

The Coal Measures form an irregular basinal structure in the southern part of the Irvine district (Figure 2). On the northern edge of the outcrop these strata rest disconformably on the Ayrshire Bauxitic Clay, and to the south the outcrop is bounded by the Inchgotrick Fault. Coal Measures also occur south of the Inchgotrick Fault in a strip along the southern margin of the district. The Coal Measures contain most of the workable coal seams in the district. Fossils are quite common (Chapter 10).

The Coal Measures is a lithostratigraphical group of three formations: Lower Coal Measures, Middle Coal Measures and Upper Coal Measures. These consist of repeated cycles of sandstone, siltstone and mudstone with coal and seatclay or seatearth, arranged in both upward-fining and upward-coarsening sequences. The strata are generally grey in colour but are extensively reddened towards the top. The base of the Coal Measures in Scotland, the Lowstone Marine Band, has not been identified in the Irvine district, so the line is drawn at a disconformity at the top of the Ayrshire Bauxitic Clay or, where this is absent, at the top of the Troon Volcanic Member. The nonmarine bivalve faunas (Chapter 10) suggest that the base as thus defined lies slightly above the level of the Lowstone Marine Band. The top of the Coal Measures is defined by the base of the Mauchline Volcanic Formation (Chapter 11).

The principal named units in the Coal Measures of the Irvine district and their relationship with the chrono- and lithostratigraphy, are shown in (Figure 24).

Environment of deposition

The styles of sedimentation in the Lower and Middle Coal Measures were very similar. The generally cyclical pattern of mudstones, siltstones and sandstones in association with seatbeds and coals indicates deposition in prograding lacustrine deltas. Coals and seatbeds formed in wetland forests where organic matter could accumulate under anaerobic conditions. Sandstones formed in mouthbars at prograding delta fronts, in distributary channels on delta plains, or as crevasse splays from the channels. Major fluvial channel-fill sandstones are rare in the Lower Coal Measures but do occur in the Middle Coal Measures, an example being the Ardeer Sandstone above the Five Quarter Coal. Deposition of siltstones and mudstones took place in low-energy environments such as lakes and inter-distributary hays. The abundance of nonmarine bivalve faunas suggests that the water was normally fresh, or possible brackish; marine faunas are known only locally, at one or two levels above the McNaught Coal.

The style of sedimentation in the Upper Coal Measures differed from that at lower levels in that thick fluvial sandstones are more common. These are usually much coarser in grain size than sandstones deposited in delta front environments and are also more mature, both texturally and mineralogically. Rip-up clasts towards the top of these channel-fill sequences indicate erosion of contemporaneous overbank material. Marine faunas are known at two levels only.

The palaeontology of the Coal Measures over the whole of the Ayrshire Coalfield, from the Irvine district in the north to the Dalmellington and New Cumnock areas in the south, has been reviewed by Brand (1983). This study also considered variations in sediment thickness, which show the relative stability of the Irvine district compared with the area around the Kerse Loch Fault, to the south. Brand demonstrated that the faults had little effect on the faunal distribution pattern, suggesting that subsidence was matched by sedimentation. The faunal distribution gives an indication of the palaeogeography, with open sea lying to the south-west, and this implies that the general direction of river flow and delta progradation was also to the south-west.

Lower Coal Measures

The Lower Coal Measures extend from the top of the Ayrshire Bauxitic Clay to the top of the Shale Coal. They consist mainly of sandstone, siltstone and mudstone in repeated cycles with seatclay or seatearth and coal at the top. Upward-coarsening trends are common. The mudstone and siltstone is usually grey to black, while the sandstone is fine- to medium-grained and off-white to grey in colour. Argillaceous strata predominate, with little development of thick sandstone bodies either as sheet sandstones or as channel-fill deposits. Coal seams are common and many exceed 0.3 m in thickness. Minor lithologies include cannel coal and blackband and clay-band ironstone, the latter occurring in nodular as well as bedded form. The characteristic 'musselbands' are beds composed mainly of nonmarine bivalves and usually occur in mudstone or ironstone.

Description of strata

Raise Coal is the name given to a seam that lies immediately above the Ayrshire Bauxitic Clay. This may not be the same seam in all areas, because sedimentation took place over a slightly uneven surface, initially infilling hollows. Bed-for-bed correlation can only be made with confidence from the Kilwinning Main Coal upwards. The Raise Coal has been worked locally, particularly around Mayfield [NS 25 42] where it reaches a maximum thickness of 1.8 m.

The strata between the Raise and Kilwinning Main coals are up to 20 m thick and consist largely of seatclays with thin impersistent coals. Other strata, where present, are usually mudstones or siltstones.

The Kilwinning Main Coal is commonly just over 1 m thick, but locally it splits into up to three leaves with seatclays between. The seam crops out on the coast [NS 2410 4115] at Saltcoats, where it has been altered to coke by the overlying picrite sill; the coal has in turn altered the lower margin of the sill to 'white trap'. The Kilwinning Main Coal has been extensively worked, particularly in the area to the east of Irvine and from Stevenston eastward to Kilwinning. Smaller areas of extraction also occur north-east of Springside [NS 37 38].

The strata between the Kilwinning Main and Kilwinning Lower Wee coals are about 4 m thick and largely argillaceous, coarsening upwards from mudstones to siltstones with seatclays at the top. Sandstones may be developed locally. The mudstones immediately above the Kilwinning Main Coal commonly carry a fauna characterised by Carbonicola pseudorobusta.

The Kilwinning Lower Wee Coal is usually about 0.40 m thick but can be as much as 0.85 m. In places the Lower Wee Coal is absent because of the intrusion of a dolerite sill. The coal has been worked in small areas such as Broom and Dubbs [NS 28 42] and Byrehill [NS 29 42] to the southeast of Stevenston, to the east of Kilwinning and at Sourlie [NS 34 41]. The position of the seam in the harbour at Saltcoats is marked by black carbonaceous mudstone with plant fragments.

The strata between the Kilwinning Lower Wee Coal and the Kilwinning Stone Coal are 4 to 6 m thick and commonly quite sandy, though more argillaceous strata are associated with both the coals.

The Kilwinning Stone Coal is usually up to 0.45 m thick but it is variable in thickness, sometimes occurring as more than one leaf with beds of carbonaceous seatclay between.

The Stone Coal in Saltcoats Harbour [NS 2426 4090] is close to a dolerite sill and comprises as 0.30 m of burnt coal, on black silty mudstone with plant fragments. Elsewhere, boreholes have shown dolerite intrusions at this position and the coal is sometimes preserved as a raft of burnt coal within a dolerite sill. Workings in the Kilwinning Stone Coal are extensive in the area south-east of Kilwinning and at Annicklodge [NS 35 41], and limited areas of coal have been extracted at Warrix [NS 33 37], Perceton [NS 34 40] and north of Dreghorn [NS 35 38].

The strata between the Kilwinning Stone Coal and the Under Ell Coal are generally about 8 m thick. They normally consist of mudstones and silty mudstones with ironstone nodules and bands, passing up into seatearth and seatclay.

The Under Ell Coal varies in thickness from a few centimetres up to 0.35 m. It is laterally impersistent and is not developed in the Ardeer area.

The strata between the Under Ell and Kilwinning Ell coals vary from 4 to 8 rn in thickness and are generally sandy, coarsening upwards from mudstones and siltstones which occur immediately above the Under Ell Coal.

The Kilwinning Ell Coal is usually about 0.6 m thick. Carbonicola sp. and Naiadites sp. are common in its roof beds. In some older borehole records the Ell Coal is apparently absent, and it is difficult to determine in these cases whether the coal had already been worked there, or was never present. The coal has been extensively worked south-east of Kilwinning and east of the River Garnock, at Annicklodge [NS 35 41], East of Perceton [NS 34 40] and in the Dreghorn [NS 35 38] and Springside [NS 37 38] areas. There was more limited extraction at Stevenston.

The strata between the Kilwinning Ell Coal and the Ladyha' Coal are 8 to 16 in thick. They are largely sandy, with thin beds of more argillaceous strata and the local development of coals with rooty beds and ironstone nodules. The Southhook Ell Coal is up to 0.40 m thick and is principally developed towards the eastern end of the Lower Coal Measures outcrop. The Plann Coal is not always identifiable and is commonly represented by a seatearth with Carbonicola sp. in the overlying mudstone. Locally it is developed as a thin blackband ironstone.

The Ladyha' Coal is more extensive, is about 0.30 m thick and commonly occurs interleaved with beds of carbonaceous seatclay. The roof mudstones carry a fauna including Carbonicola sp., Naiadites sp. and ostracods. The Ladyha' Coal was extensively worked south-east of Kilwinning, at Annicklodge and east of Perceton. More limited areas of working occur between Stevenston and Saltcoats, south of Irvine and north of Dreghorn.

The strata from the Ladyha' Coal to the Shale Coal are mostly mudstones and siltstones with some thin sandstone beds and are about 10 m thick.

The Shale Coal is a rather poor-quality coal or bituminous mudstone, usually no more than 0.25 m thick.

Middle Coal Measures

The Middle Coal Measures closely resemble the Lower Coal Measures, but upward fining channel-fill sandstones, dominated by fine- to coarse-grained sandstone, are more common. Many of the coal seams have been worked.

The base of the Middle Coal Measures is drawn regionally at the Vanderbeckei (Queenslie) Marine Band. Where this horizon cannot be established the closest approximation, based on nonmarine bivalve faunas, is taken. In the Irvine district this is at the top of the Shale Coal. The top of the formation is drawn at Skipsey's Marine Band.

Description of strata

The strata between the Shale Coal and the Upper Wee Coal are between 20 and 35 m thick. In places the proportion of sandstone is low and the sequence is mostly mudstone and siltstone, but reddened sandstones up to 20 in thick are locally developed. The roof mudstones of the Shale Coal usually contain Naiadites sp., ostracods and fish remains.

The Upper Wee Coal is the lowest of three closely associated seams that further east come together to form the Hurlford Main Coal. The Upper Wee Coal is usually about 0.6 m thick but may be thicker in places. The Kilmarnock Turf Coal, the middle seam, is usually about 0.5 m thick but may occur in leaves, so that it is difficult to correlate this part of the sequence on a bed-for-bed basis. The Kilmarnock Parrot Coal, the highest of the three, is up to 1.0 m thick and may also split, making correlation difficult. Seatbeds are common in the strata between the coals. The seams are usually only separated by a few metres of strata but where sandstones are included the separation is usually greater.

From the Kilmarnock Parrot Coal to the Linn Bed Coal the strata are usually sandy, fining upwards to a seatclay beneath the Linn Bed. The interval is about 6 m thick. The Linn Bed Coal is usually about 0.50 m thick, and has been worked locally to the north of Dreghorn and in the Sourlie area. The strata between the Linn Bed Coal and the Kilmarnock Five-quarter Coal are mostly sandstones with interbedded siltstones and mudstones, and are about 15 to 20 m thick.

The Kilmarnock Five-quarter Coal commonly exceeds 1.0 m in thickness, and is widely developed in the Irvine district. It has been worked extensively, principally in the Ardeer area, to the south and east of Dreghorn, and south-east of Irvine.

The roof beds of the Five-quarter Coal are characteristically argillaceous and carry a fauna including Carbonicola sp. and Naiadites sp. The remaining strata between the Kilmarnock Five-quarter and the Kilmarnock Major coals, 45 m thick, are mostly siltstone with some sandstone beds. The Ardeer Sandstone occupies up to 15 in of this strati-graphical interval, but it is a channel-fill sandstone and is not extensive laterally.

The Kilmarnock Major Coal, known in the Stevenston area as the Crawford Coal, is usually fairly thick, up to 0.7 rn. Thinner seams may be present locally where the seam has split, and in these cases the intervening strata are usually seatclays. The Kilmarnock Major Coal has been worked extensively to the south-east of Irvine, extending eastwards to the south and east of Dreghorn.

The strata between the Kilmarnock Major and Kilmarnock Tourha' coals comprise about 22 m of mixed strata, mostly light grey siltstone, but with sandstones, mudstones, seatclays and thin coals.

The Kilmarnock Tourha' Coal is a thick seam consisting usually of closely spaced leaves of coal with intervening carbonaceous seatclay. The total thickness of coal is usually around 1.0 in but this may be spread over 2.0 m of strata. The Kilmarnock Tourha' Coal, like the Kilmarnock Major, has been extensively worked to the south-east of Irvine, extending eastwards to the south and east of Dreghorn.

Between the Kilmarnock Tourha' and Kilmarnock McNaught coals the strata generally consist of 3 to 4 m of siltstone with some sandstone beds. Rooty beds are common. The argillaceous roof beds of the Kilmarnock Tourha' Coal commonly contain Naiadites sp. In the Stevenston area these beds are pale grey in colour.

The Kilmarnock McNaught Coal is also a split seam, usually in two leaves, the total thickness of coal being about 2.0 m. It has been extensively worked to the southeast of Irvine, extending eastwards to the south and east of Dreghorn, and at Ardeer.

There are about 13 m of mixed strata consisting of sandstones, siltstones and mudstones between the Kilmarnock McNaught and Kilmarnock Jewel coals. The roof beds of the McNaught Coal include a fauna consisting of Anthracosia sp., A nthraconaia sp. and Euestheria sp.

The Kilmarnock Jewel Coal is thin in the Irvine district, up to 0.2 m, and locally may only be represented by coaly streaks. Euestheria sp. and Anthracosia sp. have been recorded from the mudstones in the roof.

Between the Kilmarnock Jewel Coal and Skipsey's Marine Band is a mixed sequence, about 28 m thick, consisting of sandstones, siltstones and mudstones with thin coals and seatclays. The Kilmarnock Diamond Coal, 0.4 m thick, lies about 8 m below the marine band. Both the Kilmarnock Jewel and Kilmarnock Diamond coals are of limited lateral extent.

Upper Coal Measures

The Upper Coal Measures consist of sandstone, siltstone and mudstone in repeated cycles. Upward-fining trends predominate. The mudstones and silty mudstones most commonly occur as structureless clayrocks (including some calcareous marls) and seatrock. The strata are usually reddish brown and purplish grey in colour. The colouration is considered to be due mainly to oxidation of originally grey strata beneath the Permian unconformity, but some of it may be due to alteration in soil profiles soon after deposition. The reddening has destroyed much of the fossil record in this part of the sequence. Coal seams are not common, are usually less than 0.3 m thick, and may be replaced, totally or in part, by red (haematitic) and dark grey carbonaceous diagenetic limestone (Mykura, 1960). Brecciation textures occur.

Regionally the base of the formation is taken at the base of the Aegiranum (Skipsey's) Marine Band, which also corresponds to the base of the Bolsovian (Westphalian C) Stage. This bed has been identified in the Ardeer Borehole. The top is drawn at the unconformable base of the Mauchline Volcanic Formation, which is of Permian age (Chapter 11). There is no evidence within the district for the existence of Westphalian D or younger Carboniferous strata.

Outcrop areas

The Upper Coal Measures crop out in three main areas: one around Ardeer, one in an east–west-trending zone from Crosshouse to just south of Irvine, and one south of the Inchgotrick Fault. Smaller fault-bounded areas occur around Troon.

Ardeer area

The Ardeer Borehole provided a key section of the Upper Coal Measures in this area. The base of the formation, Skipsey's Marine Band, is present at 113.3 m depth, with a fauna containing Lingula sp. and conodonts. Medium-to coarse-grained sandstones are common in the sequence and are generally around 15 in thick. They are usually well sorted and mature by comparison with sandstones in the Lower and Middle Coal Measures. Rip-up clasts are common towards the tops of the sandstones. A group of thin coals occurs at 88 m depth, with Euestheria sp. in the roof mudstones. Boreholes nearby suggest that other coal seams were present but have been largely eroded by channels now filled by thick sandstones. Towards the top of the borehole there is a sequence of mottled siltstones and mudstones with thin sandstone layers. The equivalent of the Bothwell Bridge Marine Band is recorded at 40.8 m depth with a fauna consisting of Naiadites sp., Anthraconaia sp. and Myalina sp.

The area from Crosshouse to Irvine

The rocks in this area are highly faulted, hence it was known to miners as the 'Red Trouble'. The structural complexity of the area is recorded in mine plans, which show many minor folds developed within the fault zone. The strata are extensively reddened, the fault structures having provided pathways for oxidising meteoric waters. There is no exposure in this area and the borehole records are old and unreliable. Those that do exist suggest that thick channel-fill sandstones occur among silts-tones and mudstones. Richey et al. (1930) drew the base of what is now termed the Upper Coal Measures at an arbitrary 20 m (60 ft) above the Kilmarnock McNaught Coal. In the absence of any new data this boundary has been retained.

South of the Inchgotrick Fault

Here, there is a greater number of borehole records through the Upper Coal Measures. Thick channel-fill sandstones are common in the lower part of the sequence but higher in the succession the strata become more argillaceous and include palaeosols. Massive mudstones, often termed 'marls' in the early literature, also become more common. No coals are preserved in this area but it has been established by Mykura (1960) that the process of oxidation can replace a coal by limestone. Primary limestones of freshwater origin also occur. The colour of the Upper Coal Measures in this area is highly variable, often variegated in shades of red, green and lilac.

Chapter 10 Carboniferous: biostratigraphy

The biostratigraphy of the principal named fossil horizons in the Irvine district is reviewed here in ascending stratigraphical sequence. A full list of the taxa recorded from each of these beds is given in (Table 13). The palaeontological data refer to both borehole and field collections, and include some localities which may no longer be accessible. The collections relating to the resurvey were identified by R B Wilson and P J Brand, who have also re-identified most of the earlier collections of fossil material. The multielement taxonomy and bio-stratigraphy of late Brigantian to mid-Arnsbergian conodonts from the principal limestones within the district were described by Dean (1987). Authors of all species mentioned in the text are cited in the fossil inventory (p.131).

Viséan

Lower Old Mill Limestone

The earliest Carboniferous marine band known in the Irvine district, the Lower Old Mill Limestone, is of Brigantian (P1) age. It has been found in the Old Mill Borehole (the type locality) and the Kirkwood Borehole. Taxa recovered from this horizon include algae, Fenestella sp., Beecheria sp., Latiproductus cf. latissimus, Pleuropugnoides sp. Spirifer sp., Edmondia sp., Pernopecten sowerbii, Solemya primaeva, orthocones, ostracods, crinoid columnals and fish scales.

On faunal evidence this limestone is correlated to the east (Wilson, 1989, fig. 3) with the Hollybush Limestone of Paisley and with the Craigenglen beds of Milngavie, in the Glasgow district.

Upper Old Mill Limestone

The Upper Old Mill Limestone (Brigantian, P1) lies about 10 m above the Lower Old Mill Limestone in the Old Mill Borehole. It has also been sampled at the Barr-mill railway cutting [NS 3744 5131] and in the Kirkwood Borehole. The faunas are characterised by the presence of Pleuropugnoides cf. pleurodon, Productus cf. concinnus and Spirifer cf. crassus. Otherwise the composite fauna from these localities includes Serpuloides sp., Fenestella sp., Rhabdomeson?, trepostomatous bryozoa, Beecheria cf. hastata, Buxtonia?, Composita cf. ambigua, Echinoconchus cf. punctatus, orthotetoids, Productus cf. concinnus, Productus sp., Pugilis?, Schizophoria cf. resupinata, Actinopteria persulcata, Aviculopinna cf. mutica, Dunbarella sp., Leiopleria sp., Limipecten dissimilis, Lithophaga lingualis, Polidevcia attenuata, Pterinopectinella? Sulcatopinna cf. flabelliformis and crinoid columnals.

In an earlier memoir (Richey et al., 1930, p.142) this fauna was doubtfully referred to the Dykehar Limestone, but Wilson (1979, fig. 2; 1989, fig. 3) correlated the Upper Old Mill Limestone with the Blackbyre Limestone of Paisley.

Broadstone Limestone

The name Broadstone Limestone is applied to a thick fossiliferous sequence of strata which on faunal evidence reflects both brackish and fully marine environments. The bed is of Brigantian (P2) age. In an early account Craig (1869) described the sequence above the underlying coal at the type locality [NS 3615 5297] to [NS 3660 5334] as commencing with a mudstone rich in fish, above which Lingula and Orbiculoidea are present. This gives way to a more fully marine fauna with Productus sp., which completes the basal metre of the sequence. Above this the main post of limestone, about 1–1.5 m thick, is poorly fossiliferous, but is in turn overlain by a similar thickness of limestone and mudstone containing abundant large corals. Overlying this coral-bearing layer is 1–1.5 m of bluish limestone and limy shale containing a rich marine fauna. At the top of the succession is a white crinoidal limestone varying in thickness between about 1 and 3 m and described by Craig (1869) as poor in shells but rich in fish remains. The Broadstone Limestone is best developed between Beith and Stewarton to the north and south of the Dusk Water Fault.

Among other localities from which fossils have been recovered are Langside Quarry [NS 369 537], Lyonshields Quarry [NS 373 538], various quarries, outcrops and bore-holes in the Beith area, Kirkwood Borehole, Montgreenan Borehole, Old Mill Borehole and Roughwood Quarry [NS 347 523].

Typical elements of the Broadstone Limestone fauna include the coral Aulophyllum cf. fungites, fenestellid bryozoa, the brachiopods Cleiothyridina cf. fimbriata, Composita cf. ambigua, Krotovia spinulosa and Schizophoria resupinata, bivalves such as Actinopteria persulcata, Aviculopecten sp., Limipecten dissimilis, Lithophaga lingualis and Streblochondria sp. Crinoid columnals and trilobite fragments are also common. No species appears to be either diagnostic of this limestone or ubiquitous, however. Mudstones above the limestone contain a poor fauna consisting of Lingula sp. and calcareous brachiopods, although frequently the beds are leached and without recorded fossils.

On faunal evidence the Broadstone Limestone is correlated (Wilson, 1979, p.317) with the Hurlet Limestone of Paisley.

Wee Post Limestone

Where present, the Wee Post Limestone (Brigantian, P₂ age) is essentially a thin limestone or limy parting overlying a coal seam in the otherwise non-calcareous mudstones between the Broadstone and Dockra limestones. Generally it is a crinoidal limestone with occasional brachiopods present and other fossils very scarce. The Wee Post Limestone is exposed in the Lugton Water below Lugton Bridge, where it occurs as a hard limy bed, apparently unfossiliferous. In the Ardrossan Borehole algae were doubtfully identified, together with the coral Siphonodendron junceum, zaphrentid corals and crinoid columnals. Generally the roof consists of sandstone although in the Dusk Water at Hessilhead [NS 5840 5300] a thin bed of calcareous mudstone contains a poor fauna of calcareous brachiopods and bivalves.

Dockra Limestone

Outcrops of the Dockra Limestone (Brigantian, P₂ age) are numerous and fossils have been found at a number of exposures in quarries and stream sections. These are largely in the area between Glengarnock and Hessilhead to the south of Beith, continuing on the south-east side of the Dusk Water Fault to Lugton. Further south, on the south-east side of the Dusk Water Fault, the limestone is again widely exposed around Auchenskeith.

Field localities from which fossils have been identified include, from west to east, Blackstone [NS 2575 4918], Birk-head [NS 2578 4964], Hunterston to Eaglesham Gas Trench [NS 262 501] (temporary section), Thirdpart e.g. [NS 2642 5059], Cunningham Baidland [NS 278 511], Hindog Glen (Rye Water) [NS 279 511] to [NS 279 509], Hourat [NS 2885 5365]; [NS 2850 5410], Thornyside [NS 2887 5198], Garde Burn [NS 2938 4503], Paduff Burn [NS 3018 5493]; [NS 293 557], Pundeavon [NS 3091 5581], Dusk Water [NS 3192 4747]; [NS 3200 4749], Auchenskeith [NS 3135 4652]; [NS 3145 4655], Auchenmade [NS 3380 4850], Middleton area [NS 392 522]; [NS 391 523]; [NS 3980 5304] to [NS 4025 5267], Old Mill Quarry [NS 393 520]; [NS 392 529] (Plate 9), Lugton [various quarries from [NS 405 522] to [NS 412 530], Gameshill [NS 4066 4772], Bourock [NS 4075 5151], and Inchgotrick [NS 415 335; [NS 416 331].

Fossils have also been recovered from cored boreholes at Ardrossan, Kirkwood, Montgreenan, Coalhill and Old Mill (Figure 11), and from miscellaneous cored boreholes at Beall.

The wide distribution of outcrops of the Dockra Limestone and the extent to which it has been quarried have probably allowed the limestone to be more thoroughly sampled than any other in the district. This may have given an exaggerated impression of its faunal diversity in relation to other marine horizons. Nevertheless, the fauna of the Dockra Limestone is demonstrably more diverse (Table 11) than that of any other marine horizon in the district. The limestone varies lithologically and ranges in thickness from over 18 m east-south-east of Broadstone to less than 2 m at Pollick [NS 431 548]. These differences are reflected, to some extent, in the faunal abundance and diversity between localities: at Pollick and in the Annick Water, west of Broadland, all that is present is about a metre of calcareous shale containing the brachiopod Spirifer duplicicosta. Faunal differences between the Trearne and Lugton facies of the Dockra Limestone are small, although relative abundances of species vary.

In the Beith and Dusk Water areas the limestone tends to be massive and creamy white, with brachiopods and abundant large crinoid fragments characterising the upper part. Lower down the limestone becomes dark in colour, with shaly partings, and is poorly fossiliferous. At its base thick beds of mudstone alternate with thin beds of limestone In this respect the Dockra Limestone differs from the Broadstone, which tends to be massive at the base and more shaly higher up.

In general corals are more abundant in the Dockra Limestone succession than in the Broadstone and its associated mudstones. Characteristic forms include Siphonodendron junceum and zaphrentids. Aulophyllum cf. fungites, and Caninia sp. are relatively common, as are the sponges Chaetetes sp. and Hyalostelia parallels. Also typical of this horizon are fenestellid and trepostomatous bryozoa. A diverse brachiopod assemblage has been identified, in which the most commonly occurring taxa are Angiospirifer trigonalis, Antiquatonia hindi, A. insculpta, Avonia youngiana, Brachythyris sp., Buxtonia sp., Composita cf. ambigua, Echinoconchus defensus, E. elegans, E. punclalus, Krotovia spinulosa, Latiproductus cf. latissimus, Lingula mytilloides, orthotetoids, Productus concinnus, Pugilis spp., Schizophoria cf. resupinala and Spirifer cf. bisulcatus. Common, but tending to be more patchily distributed, are Alitaria cf. panderi, Beecheria cf. hastata, Fomarginifera cf. longispina, Gigantoproduclus cf. giganteus, Phricodothyris cf. lineata, Pleuropugnoides cf. pleurodon, Promarginifera trearnensis and Schellwienella sp. Bivalves regularly present include Aviculopecten sp., Edmondia .sulcata, E. cf. senilis, Leiopteria thompsoni, Limipecten, dissimilis, Lithophaga lingualis, Palaeolima cf. simplex, Pernopecten sowerbii, Plerinopectinella sp., Schizodus sp., Streblochondria sp. and Sulcatopinna cf. flabelliformis. In addition to the corals and molluscs listed, crinoid columnals are widely distributed, whilst spines of the echinoid Archaeocidaris sp., trilobite fragments and fish scales or teeth all occur commonly, although not ubiquitously.

In the Dusk Water below Hessilhead, and in the railway cutting east of Barrmill, the upper portion of the Dockra consists of calcareous mudstone with abundant zaphrentid corals. At Auchenmade and Auchenskeith the Dockra and Broadstone limestones are exposed in the same face, with the intervening strata reduced to about 2 m, only half the thickness present at Crawfield Quarry east of Glengarnock. The roof of the Dockra Limestone comprises mudstone and ironstone layers containing fish scales and ostracods.

Wilson (1979, p.318) correlated the Dockra Limestone with the Blackhall Limestone of Paisley on the grounds that the calcareous mudstones in the upper part of the Dockra contain diagnostic elements of the Neilson Shell Bed, the marine mudstone which overlies the Blackhall Limestone. These diagnostic fossils, which came from a number of different localities, are Crurithyris urii, Eomarginifera sp., Tornquistia youngi, Straparollus (Euomphalus) carbonarius, Pernopecten sp., goniatite and orthocone nautiloid fragments, Archaeocidaris sp. and trilobite fragments.

Hosie limestones

The Hosie limestones can include up to four separate limestones, referred to elsewhere in the Midland Valley of Scotland as the Main, Mid, Second and Top Hosie limestones. The three lowest are of Bingantian (P₂) age but the Top Hosie is regarded as Namurian (Pendleian, E₁) in age (Ramsbottom et al., 1978, p.31). It is impossible in Ayrshire to distinguish the four limestones palaeontologically but, whilst the background assemblages are broadly similar, there are recognisable differences between the combined fauna of the Mid and Main Hosie limestones and that of the Top and Second Hosie limestones. The fact that these faunas are recognisably different may reflect the presence of an intervening root bed, the likely equivalent of Lillie's Coal of the Glasgow district.

Faunas ascribed to the Mid and Main Hosie limestones have been recovered from an unnamed tributary of Garnock Water at Auchenskeith [NS 3145 4665], from the Dusk Water [NS 3181 4750] and from two localities in the Paduff Burn [NS 2956 5553] and [NS 3020 5490], whilst fossils more doubtfully referred to this horizon were collected at Jameston Quarry [NS 3155 4646].

Assemblages of the Top and Second Hosie limestones were identified in collections from a pylon foundation at Gateside [NS 2870 4545], from two localities in the Paduff Burn [NS 3024 5486]; [NS 3030 5477], from the Lugton Water [NS 4121 5298] and from the Dunniflat Burn [NS 4235 5355]; [NS 4232 5354]. More doubtfully referrable to these limestones is material from Whitestones Railway Cutting [NS 3332 5250] to [NS 3343 5239].

Faunas from the full Hosie sequence were recognised in material from a number of localities including the Ardrossan Borehole, a temporary section exposed in the Hunterston to Eaglesham Gas Trench [NS 262 501], Pitcon Burn [NS 293 536]; [NS 294 533], Kirkwood Borehole, Montgreenan Borehole, the Lora Burn Borehole, boreholes at Dairy and Beith, the Dusk Water [NS 382 525], the East Burn near Stewarton, and Wheatrig Colliery Blind Borehole.

The Hosie limestones contain rich marine faunas, diversity being greatest in the Mid and Main limestones. Many of the taxa seem to be irregular in their occurrence, but Lingula squamiformis, Pleuropugnoides sp., Productus sp., Schizophoria cf. resupinata, Aviculopecten sp., Nuculopsis gibbosa, Streblochondria sp., orthocone fragments and crinoid columnals occur commonly in all four Hosie limestones.

Taxa occurring much more commonly in the Mid and Main Hosie limestones, although also found in the Top and Second, are Crurithyris urii, Orbiculoidea cincla, Spiriferellina sp., Euphemites urii, Dentalium sp., Anthraconedo laevirostrum, Edmondia sp., Phestia attenuata„Sanguinolites costellatus and Dimorphoceras sp.

A few taxa which occur regularly in the Mid and Main Hosies appear never to be present in the Top and Second. These include Retispira decussata, R. striata, Cypricardella sp. and trepostomatous bryozoa.

Conversely, in the Top and Second Hosies Posidonia corrugate is usually present in large numbers, and is the commonest species, whilst this bivalve has been recorded much less frequently, and usually in smaller numbers, in the Mid and Main. Productus carbonarius, Composita ambigua and ostracods also seem to occur more commonly in the Top and Second Hosie limestones.

The faunal distribution within the Hosie limestones of Ayrshire is essentially similar to that described for the Hosie limestones in the Scottish Dinantian as a whole (Wilson, 1989, p.105), but Avonia youngiana would appear to be less common in Ayrshire than elsewhere. Although insufficient occurrences of the gastropod Hesperiella thompsoni have been recorded in Ayrshire to include it in the above list of typical taxa, it is interesting that such occurrences as have been observed are restricted to the Mid and Main Hosie limestones, in accordance with Wilson's conclusion (1989, p.105) that this species is common elsewhere only in the lowest two limestones of the four.

Namurian

Dairy Clayband Ironstone

The Dairy Clayband Ironstone (Pendleian, E1 age) has been sampled at various localities in the Dairy and Beith areas, in the Paduff Burn and in the Muirlaught Borehole. It is poorly fossiliferous, the fauna consisting essentially of Naiadites sp. and ostracods.

Johnstone Shell Bed

Marine faunas of the Johnstone Shell Bed (Pendleian, age) have been recovered from boreholes at Dairy, Lugton, Maulside, Montgreenan, Muirlaught and Bankend. All of these localities have moderately diverse assemblages, whilst elsewhere, as at Knollhcad No. 1 Borehole, Wheatrig Colliery and the Dunniflat Burn [NS 4215 5354], faunas are much more restricted, consisting principally or entirely of Lingula squamiformis.

In general the Johnstone Shell Bed is characterised by an abundance of Lingula squamiformis with Paracarbonicola pervetusta usually present in significant numbers. Other species which tend to occur commonly at this horizon are Liratingua wilsoni, Euphemites urii, Retispira striata, Anthraconeilo spp., Naiadites sp., Sanguinolites costellatus, Streblochondria sp., Streblopteria ornata and fish scales. Productoid fragments are often present, but no individual species appears to be ubiquitous.

Maich Shell Bed

From the type locality for the Maich Shell Bed (Pendleian, E1 age) in the Maich Water Lingula squamiformis, Aviculopecten inaequalis, Edmondia?, Limipecten dissimilis and a thick shelled, but specifically unidentifiable, specimen of Myalina were obtained. Otherwise fossils have been collected from the Maich Water Borehole and a limited number of borcholes at Beith and Maulside. The essential elements here seem to be restricted to Lingula squamiformis and Naiadites sp., but at Beith R₂ Borehole Aviculopecten inaequalis? was also present. This bivalve is known to occur sporadically in the stratigraphically equivalent Linwood Shell Bed of the Glasgow area, as is Paracarbonicola pervetusta which was identified in Maulside No. 7 Borehole. In this borehole Myalina sp. was recorded, although less positively identified.

Logan's Bands (Black Metals Marine Band)

Logan's Bands (Pendleian, E₁ age) have been recognised in several boreholes in the Beith and Maulside areas where, in all cases, restricted faunal assemblages were recovered. These consist essentially of Lingula squamiformis and fragments of Naiadites sp., but additionally Spirorbis sp. was identified in Beith R₂ Borehole. At Maul-side No. 5 Borehole poorly preserved specimens of Serpuloides? and Trigonoglossa? were also found. Serpuloides carbonarius was identified in strata less confidently attrbuted to the Logan's Bands in boreholes at Moorpark. Here, in two boreholes, fragments of Pleuropugnoides sp. augmented the characteristic association of Lingula and Naiadites. In Moorpark Borehole D and in a borehole at Glengarnock ostracods were also found. Elsewhere this horizon has been proved at Wheatrig Farm Borehole and possibly in the Paduff Burn, both localities producing only L. squamiformis.

Borestone Coal

The Borestone Coal (Pendleian, E₁ age) outcrops in the Lochside railway cutting [NS 3699 5906] but only fragments of Naiadites sp. have been recovered from this locality. Otherwise the horizon is best known from borehole evidence. Samples from boreholes at Glengarnock, Kilbirnie arid Maulside all produced faunas comprising only Naiadites sp. This bivalve was found beneath and/or above the coal. A similar fauna was recovered from beneath the coal in the Roche Products No. 7 and No. 12 boreholes, near Dalry. In a series of boreholes at Glengarnock, strata between the Borestone and Stone Coals were found to include bands containing Lingula squamiformis.

Stone Coal

To the north-west of the Dusk Water Fault, mudstones forming the roof of the Stone Coal (Pendleian, E₁ age) have been found in the Glengarnock series of boreholes to contain very restricted faunas confined usually to Lingula squamiformis, although fragments of Naiadites sp. were doubtfully identified in one instance.

Smithy Coal

Overlying the Smithy Coal (Pendleian, E₁ age) is a marine mudstone which has been sampled in a number of boreholes in the Glengarnock area. In every case the fauna was restricted to Lingula squamiformis. In the Roche Products series of boreholes at Dalry L. squamiformis again characterised the roof beds, but in some instances the brachiopod Orbiculoidea cincta was also identified, possibly indicating a slightly more marine environment.

Index Limestone

The Index Limestone is of Pendleian (E₁) age. Fossils have been collected at Highfield Quarry [NS 318 500] and at other field exposures including, from west to east, Girthill [NS 269 468]; [NS 276 469] , Swinlees Row [NS 2873 5324] , Gooseloan Quarry [NS 3200 4616], Crofthead Quarry (= Goldcraig Quarry) near [NS 320 499], Wheatyfauld Quarry [NS 3235 5058], Monkredding Quarry [NS 3267 4545], Lyle-stone Quarry [NS 3321 4661], Clonbeith Old Quarry [NS 3411 4555], Lugton Water [NS 3421 4534]; [NS 3457 4520], Mosside Quarry [NS 3616 4553] and Dernshaw, Stewarton [NS 3657 4518]. Faunal data from these field localities have been supplemented by occurrences of the limestone in the boreholes at Bankend and Dalry and from the Hullerhirst Borehole. To the north-east of these, boreholes at Knollhead, Montgreenan, Castleton [NS 3145 4433]; [NS 3150 4430], Glengarnock and Beith have all provided faunas indicative of this horizon.

The fauna of the Index Limestone is characterised by the presence of algae, the brachiopods Composita cf. ambigua, Latiproductus cf. latissimus, orthotetoids and Pleuropugnoides sp., the bivalve Nuculopsis gibbosa and crinoid columnals. Also common, although not ubiquitous, are trepostomatous bryozoa, the brachiopods Beecheria sp. and Rugosochonetes celticus, the gastropod Euphemites urii, the bivalves Phestia attenuata and Posidonia corrugata, and orthocones. Other species are more sporadic in their distribution.

Third Post Limestone

The Third Post Limestone (Pendleian, El age) is poorly exposed and relatively few collections have been confidently referred to this horizon. One of the best exposures is in the Putyan Burn at Broadlie Glen [NS 285 497] to [NS 278 499], from which a rather restricted fauna comprises Composita sp., Echinoconchus cf. punctatus, Eomarginifera cf. lobata, Latiproductus cf. latissimus, Leptagonia .smithi„Schizophoria cf. resupinata, Spirifer sp. [crassus group] and Edmondia sulcata. One of the most diverse assemblages recovered came from the Diddup Borehole. This contains Buxtonia sp., Eomarginifera sp., Lingula mytilloides, Pleuropugnoides sp., Promarginifera sp., Donaldina sp., Euphemites sp., Naticopsis?, Retispira?, Aviculopecten sp., Edmondia sp., Leiopteria sp., Palaeolima sp., Sanguinolites cf. plicatus., Streblochondria cf. elliptica, trilobite fragments, crinoid columnals and fish scales.

A similar though less-diverse assemblage was recovered from the Montgreenan Borehole. This comprises Beecheria sp., Composita cf. ambigua, Eomarginifera?, Gigantoproductus sp., Lingula mytilloides, Pleuropugnoides sp., Promarginifera sp., Retispira?, Sanguinolites? and crinoid columnals. From Knollhead No. 1 Borehole only unidentifiable brachiopod and bivalve fragments were recovered. In the Kilmaurs Borehole the fauna identified comprises trepostomatous bryozoa, Composita ambigua, Euphemites sp., Retispira cf. striata and crinoid columnals. The only other boreholes from which fossils of the Third Post Limestone have been identified are at those at Bankend which yielded only Eomarginifera cf. lobata, Lingula mytilloides, Cypricardella?, Posidoniella sp., Streblopteria ornala and crinoid columnals.

Lower Linn Limestone

Fossiliferous outcrops of the Lower Linn Limestone (Arnsbergian, E₂ age) are well exposed in the Caaf Water above Linn Brig, in the Broadlie Burn, Glen Burn and the Rye Water. Fossils of this horizon have also been recovered from the Diddup and Montgreenan boreholes, from Knollhead No. 1 Borehole, Mayfield No. 12 Borehole, Whitehurst Park No. 2 Borehole and from various boreholes at Bankend and Beith.

Although a diverse faunal assemblage has been identified from the Lower Linn Limestone, no species can be described as ubiquitous. The brachiopod Latiproductus cf. latissimus is generally present. Composita cf. ambigua, Productus sp., Schizophoria cf. resupinata and Spirifer cf. bisulcatus are all common components of the fauna. Bivalves tend to be scarce and sporadically distributed, but trilobite fragments and crinoid columnals are not uncommon.

Upper Linn Limestone

The Upper Linn Limestone (Arnsbergian, E₂ age) is best developed at Linn Spout [NS 2828 4853] in the Caaf Water, It has also been quarried at Diddup and is exposed in stream sections in the same area. Fossils have been collected from these localities and from others in the Bornbo Burn [NS 3110 4860; 3100 4933], at Bowertrapping [NS 326 495], Broadlic Glen (Putyan Burn) [NS 2852 4975 to 2785 4993], Flashwood [NS 278 502], Girthill Farm [NS 269 468] and Outer Smithstone [NS 2755 4615]. Boreholes at Beith, Diddup and Glencart have provided further faunal data.

The limestone and its roof are more fossiliferous than the Lower Linn. Where the underlying mudstones are present these are usually characterised by the presence of a bed containing the bivalve Edmondia punciatella in abundance. In the Douglas area, the Central Coalfield and the Kincardine Basin the presence of this bed is characteristic (Wilson, 1958, pp.25–26) of the Calmy Limestone, of which the Upper Linn is the Ayrshire equivalent. Corals occur, but not in abundance. Brachiopods are common, the most frequently encountered taxa being Buxtonia sp., Composita cf. ambigua, Latiproductus cf. latissimus, orthotetoids, Productus sp., Pugilis cf. pugilis, Pugnax cf. pugnus, Rugosochonetes cf. celticus and Schizophoria cf. resupinata. A moderate diversity of gastropods was recovered from the type locality, but elsewhere relatively few gastropods have been found. Bivalves are quite common, and the scaphopod Dentalium sp. was collected at several localities including Linn Spout. Otherwise crinoid columnals, trilobite fragments and, less commonly, nautiloid remains are characteristic elements of the fauna.

Diddup Limestone

The Diddup Limestone has been recognised in the Diddup Burn [NS 2657 4440] and in the Diddup Borehole where it is clearly recognisable as a distinct marine horizon separated by 10 m of seatclay, sandstone and silty mudstone from the underlying Upper Linn Limestone. The composite fauna recovered from these localities comprises Seipuloides carbonarius, Buxtonia sp., Composita cf. ambigua, Eomarginifera lobata?, Latiproductus cf. latissimus, Lingula mytilloides, Linoproductus?, Orbiculoidea cf. cincta, orthotetoids, Pleuropugnoides sp., Productus coninnus?, P. sp., Promarginifera cf. trearnensis, Pugilis cf. pugilis, Spirifer cf. bisulcatus, Schizophoria sp., Sinuatella?, Euchondria?, Limipecten sp., Palaeolima sp., Streblochondnia?, Tylonautilus nodiferus?, orthocone fragments, crinoid columnals, and fish teeth and scales.

This limestone has been tentatively correlated with the Castlecary Limestone (Wilson in Richey et al., 1930, p.177), but it seems more probable that it is equivalent to one of the Plean limestones. Either correlation implies an Arnsbergian (E₂) age.

Corsankell Limestone

The uppermost limestone of the Upper Limestone Formation in the Irvine district, the thin Corsankell Limestone, is poorly known and fossils definitely attributable to this horizon have been collected only from the type locality, the Diddup Borehole. Here the restricted fauna of brachiopods comprised Gigantoproductus?, together with athyrid, orthotetoid and productoid fragments.

The faunal evidence is inconclusive, as in the case of the Diddup Limestone, but it is possible that the Corsankell Limestone is also equivalent to one of the Plean limestones, which are of Arnsbergian (E₂) age.

Passage Formation

Plant remains have been found in the lowest (sandstone-dominated) member of the Passage Formation in the Caaf Water [NS 280 481] near Drumastle Mill, in mudstoncs overlying a sandstone about 10 m thick. These plants were identified by Kidston (Wilson in Richey et al., 1930, p.204) as Calamites sp., Calynimatotheca stangeri, C. stangeri forma schlehani, Lepidodendron rhodeanum, Lepidophloios scoticus, Rhodea goepperti, Sphenophyllum tenerrimum, Sphenopleridium cf. crassum, Sphenopteris sp. and Stigmaria ficoides. Wilson (in Richey et al. 1930, p.204) indicated that in this section three marine horizons are separated by 'coaly shale' and 'fireclay', with Productus sp. present in the upper two bands. In material collected more recently from the Caaf Water at Drumastle Mill [NS 278 470] Productus sp. and Schellwienella sp. are both present. In the same stream, below Drumastle Mill, two fossiliferous bands occur, the upper yielding Lingula mytilloides, orthotetoids, Pugilis pugilis and Schizophoria cf. resupinata. Fossils from the lower hand include Pugilis pugilis?, spirifer bisulcatus? and Temnocheilus sp.

A typical fauna is found in outcrops in the Fenwick Water [NS 449 413]. This comprises Lingula mytilloides, Schizophoria cf. resupinata, Donaldina. sp., Edmondia sp., Sanguinoliles costellatus?, Schizodus sp. and Wilkingia elliptica.

Material recovered from the Glen Burn [NS 2610 4392] contained Productus sp. and gastropod fragments, and probably belongs to this part of the Passage Formation, although it could be interpreted as representing one of the Plean limestones.

The Hullerhirst Borehole yielded only fragments of Productus sp. Even poorer was the fauna from the Diddup Borehole which comprised only unidentifiable shell fragments, but spore samples collected between 7.20 and 8.20 m confirmed a Namurian R1–R2 (Kinderscoutian to early Marsdenian) age for that depth range. No evidence is available to prove the age of the strata between this horizon and the Corsankell Limestone at 56 m, however, and the presence or absence of beds of Chokierien (H1) and Alportian (H2) age has not been established here.

Boreholes in the Smithstone area have afforded megaspores and marine faunas including Lingula mytilloides, Orbiculoidea cincta, orthotetoid and productoid fragments, Schizophoria cf. resupinata, Retispira sp. and unidentified gastropod fragments, Edmondia sp., Parallelodon sp., Schizodus sp. and unidentified bivalve fragments. Of these forms Lingula, orthotetoids and Schizophoria are the most consistently present.

In the Smithstone boreholes, as in the Caaf Water, at least three marine bands occur within about 15 m of strata, but elsewhere the thickness of strata and number of marine bands present in the sequence vary considerably across the Irvine district. It is not possible to identify specific bands on palaeontological evidence as has been done in equivalent Passage Formation successions elsewhere in Central Scotland. None of the faunas identified to date has included anything other than background elements typical of the Passage Formation of Scotland as a whole.

Ayrshire Bauxitic Clay Member

Plants collected south of Laigh Smithstone were identified by Kidston (Wilson, in Richey et al., 1930, p.217) as Calamites cf. undulatus and Sigillaria sp. Specimens of Neuropteris gigantea were recovered from strata immediately overlying the Ayrshire Bauxitic Clay in a stream near Kirktown Church, south east of Kilmaurs [NS 4148 4069]. The age of the member lies in the range late Marsdenian (Namurian, R2) to early Langsettian (Westphalian A).

Westphalian

Langsettian: Lower Coal Measures

Nowhere in the district has it been possible to recognise the basal faunas of the stage, and the local base of the Lower Coal Measures has been taken at the top of the Ayrshire Bauxitic Clay.

The Raise Coal overlies the Ayrshire Bauxitic Clay but may with certainty he recognised only in the area around Kilwinning. Plant remains are the only fossils recorded from the overlying mudstones. The first faunal horizon is that in the roof of the overlying Kilwinning Main Coal. Brand (1983, p.183) has discussed the regional distribution of the fauna and its principal elements. It seems likely that the fauna, which includes Carbonicola pseudorobusta amongst other species of the genus, belongs to an horizon in the Communis Chronozone and that faunas of the Lenisulcata Chronozone are absent. Musselbands above the succeeding Kilwinning Stone, Kilwinning Ell and Southhook Ell horizons also contain faunas characteristic of the Communis Chronozone. Carbonicola of the pseudorobusta group is the most commonly occurring species at all these horizons, although as a general rule preservation of individuals is poor. New faunal elements appear at the Plann Blackband Ironstone horizon, suggesting that this may be regarded as the base of the Modiolaris Chronozone. Here Anthraconaia modiolaris is found associated with Carbonicola of the oslancis group and C. subconstricta. The ironstone dies out westwards and the fauna is less varied around Stevenston [NS 265 421] and Ardeer [NS 275 415]. The fauna from the roof of the overlying Ladyha' Coal is very varied and includes Anthraconaia of the modiolaris group, Anthracosia of the ovum group, A. regularis, Anthracosphaerium affine and Carboni-cola oslancis. The ostracod Geisina arcuata often occurs in great numbers, and the combination of this feature with the bivalve fauna creates a distinctive marker bed. A comparison may be drawn between the Ladyha' Coal fauna and that recently recognised in the Govanhill area [NS 582 628] of the adjacent Glasgow district at the horizon of the Ladygrange Coal.

Duckmantian: Middle Coal Measures

So far there has been no record of marine fossils from this part of the sequence and thus the position of the Vanderbeckei Marine Band is only inferred from comparisons made with sequences to the south. Brand (1983, p.185) concluded that the roof of the Shale Coal is the most likely correlative. The fauna at this horizon is dominated by species of Naiadites, principally N. quadratus, although other species do occur.

The interval between the Shale Coal and the Upper Wee Coal contains at least three musselbands, in which a diverse fauna includes Anthraconaia modiolaris, A. salteri and numerous species of Anthracosia including A. beaniana, A. disjuncta and A. phrygiana. In a restricted area around Irvine [NS 320 390] Anthraconaia iroinensis is present. Brand (1991, pp.27–31) illustrated the unusually limited geographical and stratigraphical ranges of the species, and related these to the sediment type.

A poorly preserved fauna of species of Anthracosia occurs above the Kilmarnock Turf Coal in a limited area in the eastern part of the Irvine district, whilst a more diverse fauna characterises the roof of the Linn Bed Coal. The latter includes the last appearance in North Ayrshire of large deep-bodied Anthraconaia of the salteri group.

The overlying Kilmarnock Five-quarter Coal may be taken as the base of the succeeding Lower Similis-Pulchra Chronozone on account of the presence of abundant Anthraconaia pulchella in the overlying mudstones. Euestheria sp. is also present at this horizon over much of the northern part of the district, but occurrences die out to the south, and also to the east of Kilmarnock in the adjacent district. The association of Anthraconaia pulchella and Euestheria sp. is similar to that recorded above the Maudlin Coal (Blackball Estheria Band) of the Durham Coalfield and in other Pennine coalfields at the base of the Lower Similis-Pulchra Chronozone (Calver in Mykura, 1967, p.64). Forsyth and Brand (1986, fig. 1) correlated this horizon (in its eastern development) with the Cambuslang Marble of the Central Coalfield where a persistent hand with Euesthetia sp. is present over a wide area and is associated with a marked faunal change. Anthraconaia pulchella is not normally present at that level there, however, but is found in higher strata, between the Glasgow Ell and Glasgow Upper coals. Nevertheless it is argued that the correlation holds because it has been shown that Anthraconaia pulchella is present at higher levels also in north Ayrshire (Brand, 1983, pp.186–187), thus presaging the persistent and higher occurrences in the Central Coalfield.

Faunas collected above the Kilmarnock Major Coal are poor, but those above the Kilmarnock Tourha' Coal are more varied and include Anthracosia atra, Naiadites alatus and N. obliquus. In the more north-westerly outcrops around Stevenston, Anthraconaia pulchella is present in the predominant pale grey sediments at this position.

The Kilmarnock McNaught Coal is distinctive in that a musselband is found within the leaves of the coal, contained in a rooty siltstone of varying thickness. The fauna in this bed includes Anthraconaia cymbula, Anthracosia acutella, A. atraand A. fulva, whilst in strata overlying the coal Anthracosia atra and Naiadites spp. occur. The coal has been equated with the Ayr Ell and Sub-Ell seams combined, but the intervening Sub-Ell Marine Band has not been recognised as such. Brand (1983, fig. 12c) showed the limited distribution of Lingula sp. at this horizon, which may be represented here by the mussel-band.

Strata above the level of the Kilmarnock McNaught Coal are imperfectly known except in the Ardeer Borehole. Other boreholes suggest that the sequence is very variable, however, and that the Ardeer Borehole may not provide a reliable guide for the whole of the Irvine district. In that borehole a sequence of mudstones, thin coals and sandstones up to 25 m thick overlying the Kilmarnock McNaught Coal are characterised by faunas with indeterminate species of Anthraconaia, Anthracosia and Naiadites associated with Euestheria sp. Elsewhere, silty strata with Anthracosia atra occupy this interval.The beds in the Ardeer Borehole may correspond in part to those in the Oldhall Borehole where a marine band with Orbiculoidea cf. nitida occurs some 23 m above the Kilmarnock McNaught Coal. A Lingula band 23 m above the McNaught in the Dreghorn Borehole appears to be a higher horizon, possibly represented at about 30 m above the McNaught in the Oldhall Borehole. None of these bands was recognised in the Ardeer Borehole.

Bolsovian: Upper Coal Measures

A marine band containing Planolites ophthalmoides, Lingula mytilloides and conodonts 40 m above the Kilmarnock McNaught Coal in the Ardeer Borehole represents the Aegiranum Marine Band. The band has not been recognised in other boreholes in the district. Strata above a group of coals at about 85 m in the Ardeer Borehole contain Euestheria sp., whilst near rockhead (at c.40 m depth) a marine band with Planolites sp. and Myalina sp. is associated with specimens of Anthraconaia spathulata and Naiadites cf. daviesi. Other boreholes in the area struck the same marine band and in these Lingula myliltoides, Curvirimula sp. and estheriids are additional elements in the fauna. This hand represents the Bogton Marine Band of south Ayrshire and the Bothwell Bridge Marine Band of the Central Coalfield which contain similar associations. These in turn have been correlated with the Shafton Marine Band of the Pennine area (Mykura, 1967).

A limestone containing Spirorbis sp. and ostracods occurring offshore in Stevenston Point Bore B may represent a high horizon in the local sequence above the Bogton Marine Band. No occurrences of fauna characteristic of the Phillipsii-tenuis Chronozone have been recorded.

Chapter 11 Permian

Permian rocks crop out within the Irvine district only in a small tract of country south of the Inchgotrick Fault. These rocks are at the edge of the Mauchline Basin, most of which lies to the south, within the Cumnock district (Sheet 14E).

The rocks of the Mauchline Basin are subdivided into two formations: the Mauchline Volcanic Formation, below, and the Mauchline Sandstone Formation, above. The Mauchline Volcanic Formation consists mostly of bedded tuffs and lava flows, generally of olivine-basalt and basanite. The Mauchline Sandstone Formation comprises brick-red sandstones characterised by the presence of wind-rounded grains and high-angle, large-scale cross-bedding. Only the Mauchline Volcanic Formation occurs in the Irvine district.

Fossil evidence for the age of these strata is scant but a probable early Permian (Autunian) age has been suggested on the basis of plant material from sedimentary rocks intercalated with lava flows near the base of the Mauchline Volcanic Formation at localities to the south of the district (Wagner, 1983). Potassium/argon whole rock dates of 281 to 278 Ma have been given for lavas near Mauchline (De Souza, 1979; recalculated with new constants). The lowest volcanic strata rest disconformably on Upper Coal Measures but, although there is a distinct lithological break, there is no apparent angular discordance.

Mauchline Volcanic Formation

The Mauchline Volcanic Formation comprises tuffs, largely at the base of the sequence, and basaltic lavas. The basal tuffs and associated volcanically derived sedimentary rocks are generally poorly exposed in the Irvine district, the best exposure being in Underwood Glen [NS 395 288]. The lavas are best exposed in quarries to the east of Barnweill Hill [NS 4089 2968]; [NS 4088 2985]. At the more northerly quarry red sandstone is seen infilling fractures in the lava.

The lavas of the Mauchline Volcanic Formation as a whole (within and outwith the Irvine district) are mostly olivine-basalts comparable to the Carboniferous Dalmeny and Hillhouse types, but a significant number of analcimebasanites and nepheline-basanites has been recorded (Eyles et al., 1949). Both olivine and, less abundantly, augite occur as idiomorphic microphenocrysts (up to 2 mm) in a groundmass consisting of granular augite and small feldspar laths in variable proportions. The basanites have analcime or nepheline in the groundmass but characteristically there is little interstitial glassy material. Detailed petrographical descriptions were given by Tyrrell (1928a).

The rocks commonly appear to be altered and, like those of the Troon Volcanic Member, many have a red-speckled appearance owing to the alteration of olivine to Iddingsite'. However, except for the olivine the rocks are generally quite fresh in thin section.

Analyses of lavas from the Mauchline Volcanic Formation have been published by Macdonald et al. (1977) and Wallis (1989). (Table 14) shows analyses by Wallis of four lavas from the district, together with nine analyses of blocks included within pyroclastic vents and seven from minor intrusions associated with the vents (Chapter 12). An analysis of a dyke which cuts the Black Rock vent was given by Alexander et al. (1986). The rocks of the whole suite are predominantly silica-undersaturated (nephelinenormative) alkali-basalts and basanites, but there are also a large number of silica-saturated (hypersthene-normative) 'transitional' basalts. More differentiated rocks (Figure 22) are represented by a few basaltic hawaiites and hawaiites. More detailed geochemical studies, and discussion in the context of Carboniferous to Permian igneous activity in southern Scotland as a whole, are to be found in Macdonald et al. (1977), Macdonald (1980) and Wallis (1989).

Environment of deposition

The lavas of the Mauchline Basin were erupted under subaerial conditions after a relatively long period of non-deposition covering much of Stephanian time. Reddening of the underlying Carboniferous strata is probably due mainly to atmospheric oxidation of originally grey strata during this period. The topography at the start of Permian times was subdued, and the lavas were erupted on to a floodplain. The district lay within a large continental land mass about 8° north of the equator, with an arid to semi-arid climate. These tropical conditions led to strong contemporaneous weathering of both the lavas and tuffs, and account for the present paucity of fresh material at outcrop.

Chapter 12 Intrusive igneous rocks

The intrusive igneous rocks of the Irvine district range in composition from picrite to rhyolite. The oldest is a dyke of possible early Devonian age, while the youngest are dykes intruded during the Palaeogene Period. The intrusions are variable in size and form. Near vertical dykes are mostly only a few metres wide, though some range up to ten metres and extend across country for several kilometres. Sills, gently inclined sheet-like masses approximately concordant with the bedding in the country rock, are particularly widespread in the south of the district. Volcanic vents and plugs are roughly cylindrical in shape, with many of the vents containing fragmental material of both igneous and sedimentary origin.

Plugs and vents

Over 40 pyroclastic vents, and a few plugs, are known in the Irvine district. Many of these stand out from the surrounding country as small hills or grassy knolls; others have been encountered in underground workings where they may cut through a whole series of coal seams. The underground exposures show that the vents are typically steep and that they commonly narrow downwards, although a gently inclined flat sheet of vent material has been traced at one place.

The vents and plugs intrude strata ranging from the Stratheden Group (late Devonian) to the Upper Coal Measures (late Westphalian), and outwith the district they cut the Mauchline Volcanic Formation (Autunian) but not the overlying Mauchline Sandstone. However, the age of the youngest country rock cut by an intrusion can only give a maximum age. Further evidence must come from the petrography, either of igneous material incorporated in the vents, or of minor intrusions that cut them. Previous accounts (e.g. Richey et al., 1930) attributed great significance to the presence of well-rounded, frosted quartz pebbles and grains in many of the vents. These were thought to have been derived exclusively from the aeolian Mauchline Sandstone, of Autunian age. However, aeolian sandstones of late Devonian age have now been identified in the Fairlie Sandstone Formation (Chapter 4), through which some of the vents could have passed, so the Autunian age must be viewed with caution.

The majority of vents in the Irvine and adjoining districts cut Coal Measures or younger rocks. Many of these contain distinctive igneous rock fragments or are cut by minor intrusions which have clear affinities with lavas of the Mauchline Volcanic Formation, in that they are basanitic, monchiquitic or nephelinitic. There is, therefore, little doubt that most of the vents and plugs are contemporaneous with the Autunian volcanicity. Vents and plugs outwith the outcrop of the Coal Measures could possibly be older, although many of these also contain characteristic Mauchline lava rock types. The vent at Holmbyre [NS 264 483], however, contains pebbles of limestone and isolated specimens of marine fossils such as Productus, Euphemites urei and Orthoceras, which are no younger than Namurian (Smith, 1891; Richey et al., 1930). The fact that the fossils are free of a sedimentary rock matrix and that ash has been found inside them suggests that they were not derived from pre-existing sedimentary rocks. It follows that this vent, at least, is of Carboniferous age, contemporaneous either with the Troon Volcanic Member or with beds of tuff in the Lower Limestone Formation. Vents at Law Hill [NS 21 48], West Kilbride, which are cut by quartz-dolerite dykes of the late Carboniferous regional swarm, must also be older than Autunian. Several vents in the Kilbirnie Hills contain predominantly trachytic material; these, together with trachytic plugs in the Kilbirnie and 'kith hills, can be confidently assigned a Visean age, since all post-Visean igneous activity was entirely basic in composition. Other basaltic vents and plugs in the Kilbirnie Hills may belong to the same period, but their age cannot be proved.

Most of the vent material consists of unstratified greenish grey basaltic volcaniclastic material ranging from tuff to agglomerate or pyroclastic breccia. Included blocks and lapilli are mostly of locally derived sedimentary and igneous rocks, including some that have subsided into the vents from higher stratigraphical levels no longer extant in the area (e.g. blocks of Mauchline Sandstone). Olivine-basalts and basarrites from the Troon and Mauch-line lavas are common and alkali dolerite from the sills occurs locally. Plugs, thin sills and dykes associated with the vents are predominantly of highly undersaturated olivine-analcimite or monchiquite but alkali dolerite and camptonite are also known. Analyses by Wallis (1989) of included blocks and intrusions associated with some of these vents are listed in (Table 14) and plotted on (Figure 22), which shows their link with the Mauchline lavas.

Exotic lithologies

A notable feature of many of the vents is the presence of fragments of middle crustal, lower crustal and upper mantle material, which are a valuable source of information on the deep geology of the region. The presence of exotic minerals and rock fragments in several vents was first noted by Wilson (1916) and Richey et al. (1930). Details of those known in the Irvine district are summarised in (Table 15). Of particular importance are the vents at Black Rock, Campbelton Farm, Baidland Hill and Holmbyre. Descriptions of material from these vents have been included in several syntheses of xenoliths from volcanic rocks of the Midland Valley (Graham and Upton, 1978; Upton et al., 1984) and of Scotland or the British Isles in general (Upton et al., 1983; Hunter and Upton, 1987; Halliday et al., 1993).

The middle and lower crustal xenoliths comprise a variety of equigranular to gneissose rocks, commonly banded, with compositional and textural features indicative of granulite facies metamorphism. Quartzofeldspathic felsic gneisses and/or granulites are found in all four of the vents mentioned above and some, from Baidland Hill, are garnetiferous. Hydrous minerals such as biotite and amphibole are rare in these gneisses. The Baidland Hill and Holmbyre vents have also yielded more basic pyroxene-granulites consisting essentially of varying proportions of plagioclase and clinopyroxenc, usually with spinel, altered orthopyroxene and, rarely, garnet. These assemblages represent the pre-Palacozoic basement of the Midland Valley, which is evaluated in a detailed isotopic and geochemical study by Halliday et al. (1993). The LISPB deep seismic profile (Bamford, 1979) shows a clear distinction between the middle crust with a seismic velocity of more than 6.4 km per second and the lower crust with a seismic velocity of about 7 km per second (Chapter 13). In broad terms the quartzofeldspathic granulites and gneisses probably represent the middle crust, whereas the pyroxene-granulites and metaanorthosites represent stratiform igneous complexes from the lower crust, which have been subjected to various degrees of recrystallisation, partial melting and possible local metasomatism. The felsic and mafic granulites may also be intermixed in coarsely banded gneiss formations, particularly at lower crustal levels (Upton et al., 1984).

The upper mantle is represented by a variety of ultramafic rocks within a range of magnesian peridotites and pyroxenites. The most abundant mantle-derived xenoliths are spinel-lherzolites, but wchrlites, clinopyroxenites and garnet-pyroxenites also occur in the district and some biotite- and kaersutite-bearing hydrous pyroxenites have been recorded. Separate crystals, commonly large megacrysts derived from pegmatitic facies in the upper mantle or lower crustal rocks, are a feature of many vents. Some of these are clearly derived from ultramafic or ultrabasic rocks but others, most notably garnet and anorthoclase, are not (e.g. at Holmbyre). Many are clearly out of equilibrium with the host magmas which carried them to the surface. Many of the ultramafic xenoliths, in particular the magnesian peridotites, show a variety of textures which reveal a history of deformation and recrystallisation. These may represent upper mantle material depleted by partial melting episodes. Other ultramafic xenoliths are undeformed and may represent slightly younger intrusions within the upper mantle. The hydrous pyroxenites suggest localised potash metasomatism within the mantle. Higher level crustal or subcrustal intrusions are a more likely source for xenoliths of layered gabbro and picrite, commonly with cumulus textures, and for some of the megacrysts, in particular the anorthoclase (Upton et al., 1984; Hunter and Upton, 1987).

Trachytic plugs and vents of supposed Viséan age

Plugs and irregular bodies of pyroclastic breccia within the outcrop of the Misty Law Trachytic Centre are almost certainly related to this Visean phase of extrusive activity and acted as feeders for the trachytic and rhyolitic lava flows. Many occur in the Greenock district (Paterson et al., 1990) but those of the Knockside Hills [NS 25 58] in the Irvine district are among the most prominent, forming steep-sided hills, sub-circular in plan. The hills are formed from plugs of porphyritic trachyte which is commonly brecciated, especially close to the plug margins. Phenocrysts of perthitic sanidine, microperthitic orthoclase and less commonly albite, are set in a fine-grained ground-mass of alkali feldspar, oxidised pseudomorphs after mafic minerals and secondary quartz.

Vent agglomerates are recognised in the Misty Law Trachytic Centre, but much of the widespread outcrop of trachytic breccias around the Knockside Hills consists of bedded deposits, faulted against lava flows in places (as at Guillic Burn [NS 25 57] ) to give the impression of steep-sided contacts (cf. Wilson, 1916). It is likely that vent infills do occur close to the plugs, but these cannot be distinguished from the extrusive deposits.

Another vent of trachytic composition occurs at Castle Hill [NS 285 536], 3 kin west-south-west of Kilbirnie. It is associated with three small plugs of basalt. A large out, crop of trachyte centred upon Lochlands Hill [NS 37 55] and Brownmuir Plantation [NS 37 56], to the north-east of Beith, has been interpreted as two large plugs, 3 X 2 km and 2 X 1 km in size (Richey et al., 1930). The rock is typically porphyritic with abundant phenocrysts of sodic orthoclase, anorthoclase and more rarely albite. The highly feldspathic groundmass consists of alkali feldspar laths with interstitial quartz, some of which may be secondary. Pseudomorphs after augite or olivine are present in some sections and iron oxide occurs as a fine dusting throughout the groundmass. A small plug east-north-east of Wardlaw [NS 249 514] in the Kilbirnie Hills consists of similar rock types.

Other plugs and vents

Fairlie–West Kilbride–Dairy

Some 20 vents are recognised in this area, mostly cutting late Devonian sedimentary rocks and Clyde Plateau lavas. This group was described by Wilson (1916).

At Fairlie, two vents cut the Seamill Sandstone Formation and trachyte sills. The larger vent, forming Diamond Hill [NS 211 540], contains tuffs and agglomerates and has yielded some exotic xenoliths. Disturbed strata at the margin of the other, more easterly, vent are exposed in a stream bed [NS 2165 5382]. An important suite of xenoliths has been obtained from the coarse agglomerates of the Black Rock in the intertidal zone south-west of Fence Bay [NS 20 53] (Young, 1895; Smellie, 1916; Alexander et al., 1986). Basalt and sandstone fragments are abundant but the vent has also yielded spinel-lherzolites, wehrlites, websterites, hydrous ultramafic rocks, quartzofeldspathic granulites, small garnets and various megacrysts, including pyroxenes up to 10 cm across. The local red sandstone is bleached and brecciated at the eastern margin of the vent (Patterson, 1949), but this exposure is now obscured by the Hunterston iron-ore terminal.

Around Hunterston Power Station, small vents occur near Stoney Port [NS 180 513] and at Campbelton Farm [NS 191 511]. Boreholes at the latter site have yielded xenoliths of spinel-lherzolite and lower crustal rocks. Inland from here small vents cut the western escarpment of the Biglees Hill Sill [NS 207 518]. Vents also occur on Whiteside Hill [NS 225 524] and at two grassy mounds known as Green-side [NS 205 506] and Lairdside [NS 530 504] hills, to the south of Caldron Hill. All three are close to the margins of the Glentane Hill Sill, and that at Greenside Hill contains blocks of porphyritic rhyolite similar to that of the sill. A small exposure of calcite-veined tuff, cut by a northeast-trending basic dyke, occurs in the Crosbie Burn [NS 222 507].

Three large vents cut Clyde Plateau lavas near Baidlandhill Farm [NS 265 519]. The margins of the most northerly vent are well exposed; in the Millour Burn [NS 265 523] the western margin is steeply inclined inwards, the tuffs inside dip inwards at about 60° and the surrounding lavas are baked. All three vents consist of dark green tuffs which weather red, with a spheroidal structure. Inclusions are of sandstone, various types of basalt and some felsic rocks, the latter probably derived from underlying sills. Well rounded quartz-grains are common, and several fragments of charred wood have been found. Large slabs and masses of sandstone, with rootlets and coaly streaks, occur close to the margins of the most northerly vent and it is recorded that a coal about 1.5 m thick was once worked from a level in the field north of the farmhouse (Richey et al., 1930). It seems probable that the coal was part of a large mass which had subsided into the vent and that the vent therefore postdates the earliest coal-bearing strata of the area. If so, it must be significantly younger than the Clyde Plateau Volcanic Formation. The vents have yielded a wide range of xenoliths, including spinel-lherzolite, garnet-pyroxenite and hornblende- and biotite-bearing hydrous pyroxenite, from the mantle; and pyroxene-granulite, garnet-granulite, quartzofeldspathic granulite and garnetiferous quartzofeldspathic granulite from the lower and middle crust. Megacrysts of biotite, clinopyroxene, amphibole and brown spinel are common.

A line of vents and plugs extends east-south-east from Carlung [NS 194 492], through West Kilbride to Holmbyre. A small vent west of Carlung House contains abundant fragments of felsic rock and between here and Carlung Farm, a slightly larger vent is cut by a small basaltic plug. A vent of similar size forms Drummilling Hill [NS 205 492] to the east. Law Hill [NS 21 48], West Kilbride consists of coarse agglomerate containing red sandstones and black shales. Both this, and a small vent 150 m to the east, are cut by east-south-east-trending quartz-dolerite dykes. Blackshaw Hill [NS 226 485] consists of tuffs from which fragments of black shale with Spirobis have been recorded. A large mass of baked sandstone occurs at the northwestern margin and the top of the hill is formed by a small plug of Dalmeny-type basalt, cut by a north-west-trending basalt dyke. The fragments of black shale found in these vents suggest that they post-date the Clyde Plateau lavas. The large vent at North Hill of Knockewart [NS 238 481] is intruded into lavas at the base of the Clyde Plateau Volcanic Formation; green tuffs close to the western margin of the vent are seen to dip inwards at 40°. The Holm-byre vent cuts strata just below the Index Limestone and contains fossils of Namurian or earlier age which are apparently contemporaneous with the formation of the vent. The dark grey tuff of the vent contains a great mixture of rock fragments including various basalts, some trachyte possibly from the underlying lavas and, in one exposure, pebbles of limestone. The xenolith suite from this vent includes garnet-pyroxenites, pyroxene-granulites and quartzofeldspathic granulites, but it is probably best known for its megacrysts. These include biotites (up to 40 X 12 mm), large crystals of anorthoclase and small crystals of garnet and brown spinel.

Tarbert Hill [NS 208 473], to the south of West Kilbride, is a prominent green hill composed mainly of yellowish or green fine-grained sandy tuff with scattered quartz pebbles. Traces of vertical stratification are seen in places.

Around the margins the vent material is a coarse agglomerate composed of rounded blocks and lapilli of various basalts, with some large blocks of red sandstone.

In this area basaltic plugs are obvious where they cut through sedimentary rocks below the Clyde Plateau Volcanic Formation but are difficult to identify within the lavas themselves owing to the similarity in rock type. Some or most of the plugs may be contemporaneous with the lavas, but this cannot be proved. Others, such as the small plug which cuts the Blackshaw Hill vent (containing fragments of black shale), probably postdate these lavas.

The largest basic plugs occur on Lairdside Hill [NS 23 54], where the rock is an ankaramite of Craiglockhart type, and to the north of Munnoch Burn [NS 236 489], where the rock is similar to the Markle-type lavas. Three small plugs which cut the trachytic vent agglomerate at Castle Hill [NS 285 536] consist of olivine-basalt of Dalmeny or Hillhouse type. On Castle Hill itself, columnar jointing at right angles to the plug margin changes from near horizontal, in a small quarry on the flank, to near vertical on top of the hill, suggesting an upper termination of the plug near the present erosion level. Pectolite and prehnite occur here in large amygdales. Two small plugs of olivine-basalt occur west of the vent at Knockewart Hill and others cut basaltic vent agglomerates at Blackshaw Hill [NS 226 485] and Garbing Farm [NS 194 492].

Stevenston

Two small mounds near Ashgrove House [NS 282 446], 3 km north-west of Kilwinning, consist of dark decomposed tuff. They presumably represent two small vents which cut the Kilbirnie Mudstones. Fellie Hill near Kerelaw [NS 268 429] consists mainly of a plug of coarse-grained olivine-basalt of Dalmeny type with a little tuff at its northern margin, cutting the Limestone Coal Formation. Two small vents of basaltic tuff cutting the lowest part of the Passage Formation and the Troon Volcanic Member form Castle Hill [NS 282 431] and Little Hill [NS 283 431], 1.5 km north-east of Stevenston.

Dunlop

A small vent is inferred from fragments of tuff and basalt on the knoll of Kirk Hill near South Kilbride Farm [NS 396 462], within the outcrop of the Limestone Coal Formation.

Irvine–Dreghorn

Sixteen vents are recorded in the area around Irvine and Dreghorn. Most of these are known only from underground workings or boreholes.

South-west of Irvine, an exposure of basaltic agglomerate in the river bed [NS 3125 3840] marks the site of a vent cutting Middle Coal Measures. Upstream from the town an exposure of agglomerate in the River Irvine [NS 344 385] north of Greenwood contains fragments of Mauchline Sandstone. Just to the south of here, at Oldhall [NS 331 369], a vent is marked at the surface  by a small ridge. Underground, strata of the Upper Coal Measures were seen to turn down sharply towards the highly irregular western margin of this vent. The outer 10 m of the vent consist of coarse agglomerate made up entirely of locally derived sedimentary rocks. This rim grades into a finer-grained central part consisting entirely of igneous clasts. The vent is cut by two small dykes, one a Dalmeny-type basalt and the other a fresh augite-rich monchiquite.

Two large vents occur near Springside [NS 37 38], both cutting Upper Coal Measures. The western vent is exposed in the Carrier Burn [NS 371 398] and has been proved in underground workings to consist of two vents which widen and coalesce upwards. The rock is agglomerate with some sandy tuff containing rounded quartz grains. The eastern vent, around Thorntoun [NS 38 38], consists of basaltic tuff with monchiquite fragments. It was shown underground to have an inward sloping western margin.

An outcrop of tuff in the River Irvine near Cockhill [NS 378 367] is shown in workings to be part of a vent extending eastwards. A little farther to the north-east a near-horizontal intrusion of fragmental material at the level of the Parrot Coal has been traced underground over an area of 300 X 125 m. The intrusion contains fragments of 'white trap' and locally derived sedimentary rocks, including sandstone slabs up to 1 m long. Adjacent strata are not much disturbed, but the Parrot Coal is burnt in places.

Other vents in the area are known only from boreholes or underground workings.

Dundonald

Temporary exposures in Dundonald village [NS 36 34] revealed brecciated fine-grained basalt which is interpreted as being the margin of a plug cutting the Dundonald Sill. South-east of the village, at Broomhill [NS 377 338], a plug is marked by several exposures of xenolithic fine-grained basalt. This plug probably cuts leaves of the Hillhouse Sill and the Middle Coal Measures. At Templeton [NS 395 345], east of Dundonald, a small plug of xenolithic fine-grained monchiquite cuts the Middle Coal Measures.

Symington

A large outcrop of basalt at Townend [NS 435 310], west of Symington, is interpreted as a plug that cuts the junction between the Troon Volcanic Member and the Lower Coal Measures. South of the village, three basaltic vents in an east–west line cut the Upper Coal Measures and bedded pyroclastic rocks at the base of the Mauchline Volcanic Formation. The western vent was exposed formerly atleanfield [NS 385 306] and temporarily in a pipeline trench. The eastern vent, at Helenton Hill [NS 39 30], consists of coarse pyroclastic breccia containing fragments of various basalts and alkali dolerites, including some with conspicuous large augites that resemble parts of the Hillhouse Sill. This vent is cut by a small plug and a dyke of basalt.

A group of seven small vents and a plug in the area around Heughmill [NS 402 305] all cut bedded sandy tuffs or lavas of the Mauchline Volcanic Formation. The 101 m hill to the north-east of Heughmill consists mainly of olivine-basalt of Dalmeny type, but pyroclastic vent deposits occur on the eastern margin. Agglomerate in the vent at Heughmill contains blocks of augite-rich monchiquite and xenocrysts of anorthoclase and diopside.

Sills

Sills are present in several parts of the district, most extensively in the south (Figure 1b). Three groups are recognised, one of probable Visean age, one of Silesian to early Permian age and one of Palaeogene age.

Sills of Viséan age

Extensive thick sheets of felsite cut late Devonian sedimentary rocks on the west side of the Kilbirnie Hills. Their felsic nature suggests affinity with the Visean volcanicity rather than with later, more basic activity. There are four main areas of outcrop and numerous smaller ones. The larger ones occur on Goldenberry Hill [NS 18 50], south of Hunterston; from Biglees Hill [NS 20 51] northwards as two separate leaves to Fairlie Glen, and possibly farther north to Kelburn Glen; on Glentane Hill [NS 22 51] north to Whiteside Hill [NS 22 52]; and from Caldron Hill [NS 22 51] northwards to the headwaters of the Caaf Water. The geometry of these intrusions is difficult to ascertain but the outcrop pattern suggests that some of the disparate outcrops belong to a single sheeted complex with rafts of sandstone included. The relationship with the adjacent sedimentary rocks appears to be conformable in the east at Caldron Hill, the intrusion dipping eastward. To the west, however, at Glentane and Biglees hills the intrusion is disconformable and dips to the west.

The sills are composed, for the most part, of highly altered rock and many are identified merely as felsite or unclassed felsic rock. However, the large sills at Caldron hill and Glentane Hill have sufficient primary quartz to be termed rhyolite, whereas the Biglees–Fairlie Sill is a trachyte. Almost all of the rocks have feldspar phenocrysts (0.5 to 2 mm), commonly in glomeroporphyritic aggregates. They are mostly altered to calcite and clay minerals, but anorthoclase has been identified, rimmed by orthoclase in one case. Rare pseudomorphs after mafic minerals are present. The groundmass consists of alkali feldspar laths and quartz with minor iron oxides.

Several sills of silicified feldsparphyric rock cutting trachytic agglomerates at Rye Water Head are of similar lithology to the adjacent plugs of the Knockside Hills, to which they are probably related. Thin sills of albitetrachyte on Great Cumbrae are associated with the earliest dyke phase recognised on the island.

Sills of Silesian to early Permian age

In the southern half of the Irvine district, Namurian and younger strata are intruded by numerous doleritic sills ranging in thickness from less than a metre to over 30 m. Mining activities and associated boreholes have shown that many of them have considerable lateral extent, commonly departing little from one stratigraphical level. Many thin sills follow planes of weakness created by coal seams, which has resulted locally in productive seams being destroyed ('burnt' or coked) or converted to a higher grade of coal (anthracite). In such localities, and also where the sills cut bituminous shales, the dolerite is usually altered to 'white trap' by volatiles released when the sedimentary host rocks were heated up (Mykura, 1965). Where the dolerites are altered in this way the primary igneous texture is usually preserved but the constituent minerals are pseudomorphed by kaolinite, chlorite, leucoxene, amorphous silica and carbonate to give the characteristic pale cream to yellowish brown 'white trap' material.

The sills are varieties of alkali dolerite. They have been the subject of many detailed petrological studies (Falconer, 1907; Tyrrell, 1909; 1912; 1923; 1928a; 1928b; Eyles et al., 1930; 1931; 1949; Richey et al., 1930; Patterson, 1945; 1946) in which many different names have been used to describe the more distinctive rock types. Most of these names have now been abandoned, except for historical reference. The grouping into petrographical types adopted by Richey et al. (1930) and Eyles et al. (1949), although useful for descriptive purposes, has little or no petrogenetic significance (Henderson et al., 1987) and has been simplified considerably. Only three types are regarded as significant in the Irvine district and only one of these (Type 2) is volumetrically important.

• Type 1 Alkali olivine-dolerites and basalts similar to the Dalmeny-type lavas of the Troon Volcanic Member.
• Type 2 Alkali dolerites, gabbros, monzogabbros and picritic variants, commonly analcime- and/or nepheline-bearing (teschenites, theralites, essexites); also some fine-grained amphibole-bearing camptonitic rocks (former 'teschenitic' and 'kylitic' types).
• Type 3 Fine-grained analchne-basanites (former 'monchiquitic' types).

Details of the whole-rock geochemistry and mineral chemistry of the sills are discussed in relation to Silesian-Permian igneous activity as a whole by Wallis (1989), who gives 12 whole-rock analyses from the Irvine district ((Table 16); (Figure 22)). The analyses show a restricted compositional range of silica-undersaturated rocks at the basic end of an alkali basalt series. Normative nepheline ranges from 3.26 to 13.90 per cent, but most values are greater than 5 per cent, indicating basanitic compositions. Three analyses indicate slightly more fractionated rocks; these are basaltic hawaiites from the Saltcoats Main Sill and the Crosshouse Sill, and a hawaiite from the Caprington Sill.

Age relationships of the sills are difficult to determine with precision. Type 1 sills have a petrographical and spatial association with the Troon Volcanic Member and only cut rocks of that unit and older. They are probably contemporaneous with this Namurian volcanic activity, but could be younger.

Type 2 sills cut Coal Measures throughout the Ayrshire Coalfield and are therefore of late Westphalian age or younger. They are cut by vents and dykes having petrological affinities with the Mauchline Volcanic Formation and occur as included blocks in some of the vents. Hence they are assumed to be older than, or possibly broadly contemporaneous with, this Autunian volcanicity. Relationships with major fault sets suggest that some significant age differences do exist (Eyles et al., 1949; Mykura, 1967); many of the sills are cut by north-west- to west-north-west-trending faults, whereas others postdate all major faults. Palaeomagnetic measurements support a Permian age (Armstrong, 1957) and K/Ar ages from separated minerals range from 303 to 282 Ma (De Souza, 1979). A more accurate 40Ar/39Ar age of 288 ± 6 Ma has been obtained from the related Lugar Sill in the Cumnock district (Henderson ct al., 1987).

The Type 3 sills, rare in the Irvine district, have obvious petrological affinities with the Mauchline lavas and similar rock types are found in associated vents, both as included blocks and as intrusions. None has been found cutting Permian sedimentary rocks above the volcanic formation (Eyles et al., 1949) and there is little doubt that these sills are part of the Autunian volcanic activity.

Type 1 Olivine-basalts

A sill of coarse-grained olivine-basalt of Dalmeny type, petrographically similar to the Troon lavas, occurs at the top of the Kilbirnie Mudstone at Coalhill [NS 245 468], near Meikle Busbie. Two sills of similar rock, too small to be shown on the 1:50 000 map, occur at Ardrossan, on the Eagle rock [NS 2220 4175] and on the foreshore at the west end of South Bay [NS 2317 4191].

Type 2 Alkali olivine-dolerites, garbros and related rocks

Two major sills are well exposed on the foreshore at Saltcoats, conformable with the regional dip to the southeast. To the west of the harbour [NS 24 41], the Saltcoats Main Sill is about 18.5 m thick and composite, with three main units believed to have been intruded successively (Patterson, 1945; 1946). A marginal unit of flow-banded analcime-dolerite (teschenite) up to 2.7 m thick contains xenoliths of hornfelsed sedimentary rock. Upper and lower zones (2.7 and 3.7 m) of biotite-analcime-dolerite (biotite-teschenite) contain pink segregation veins and patches rich in analcime, titanaugite and alkali amphibole. A central zone of alkali amphibole-, biotite- and nepheline-hearing picrite (thcralitic picrite) is 9 m thick. Overlying sandstones are altered to a spotted hornfels and, at the base of the sill, the Kilwinning Main Coal has been transformed into a columnar-jointed coke. The basal 1.5 m of the sill has been altered to 'white trap'.

The promontory of Inner Nebbock [NS 2445 4085] to the east of the harbour is formed by a sill of analcime-dolerite (teschenite) intruded between the Upper Wee and Shale coals in the Middle Coal Measures. Hornfelsed sedimentary rocks can be seen above the sill. In a railway cutting about 1 kin inland, this sill consists of three layers with sharp margins; an upper and lower analcime-dolerite and a central coarse-grained picrite, each 3 to 4 m thick. The picrite was probably worked at the former Parkend Quarry [NS 253 415], which supplied 'Osmond Stone' for bakers' ovens. Farther inland the sill can be followed as a topographical feature to Caponcraig Quarry [NS 262 420], Stevenston, where the picrite layer is absent. The sill has been traced farther east in boreholes.

The headland of Castle Craigs [NS 227 415] at Ardrossan is formed by a composite banded sill described by Falconer (1907). It is considered to be part of the Saltcoats

Main Sill, displaced by the Ardrossan Harbour Fault. A lower, marginal layer of 'olivine-feldspar rock' is overlain by a thick layer of coarse-grained amphibole-bearing picrite. The upper part of the sill, considered by Falconer to have been intruded later, consists of a thin layer of amphibole-bearing dolerite overlain by fine-grained banded dolerite. The banded rock is biotite-analcime-dolerite (biotite-teschenite) which becomes less olivine rich upwards and develops alkali amphibole as it passes up into the topmost metre, a fine-grained camptonitic analcime-basalt.

Numerous sills, all of analcime-dolerite (teschenite), occur in the area between Kilwinning and Dunlop. Several outcrops, possibly of the same sill, occur within the Kilbirnie Mudstone between the Lugton Water east of Auchentibcr [NS 365 471] and Magbiehill [NS 409 472]. The sill exposed in the Lugton Water is a subophitic dolerite with a little biotite. The most extensive sills occur in several leaves within the Limestone Coal Formation between Fergushill Hall [NS 350 461], Auchentiber and Kennox [NS 384 450]. All are analcime-dolerites with petrographical similarities to parts of the Inner Nebbock Sill. Farther south a sill cuts the Lower Coal Measures between Montgreenan [NS 344 443] and Torranyard [NS 357 440].

Much of Irvine is sited on the outcrop of a sill which intrudes the lower part of the Middle Coal Measures. The sill dies out abruptly to the east and splits into thinner sills to the north and north-east. Petrographically it is similar to the analcime-dolerites of the Inner Nebbock Sill and the picrites of the Saltcoats Main Sill. The analcime-dolerites at the former Duntonknoll Quarry [NS 324 395] were an important source of 'Osmond Stone' for bakers' ovens at one time. In the area of Sourlie, north-east of Irvine, one or more sills are known from boreholes and mining between the Ladyha' and Upper Wee coals at approximately the same stratigraphical level as the Inner Nebbock Sill.

Two east–west-elongated outcrops of analcime-dolerite around Crosshouse are believed to be parts of a single sill. The northern crop has no surface exposure and is known only from mining information, whereas the southern crop is exposed in the Carmel Water and has been quarried near Craig House [NS 378 372]. The sill is intruded into Middle Coal Measures at the level of the Kilmarnock Major Coal, which is locally destroyed by it. Some samples are biotite-bearing analcime-doleritcs (teschenites) but others are very coarse grained with large sub-ophitic to ophitic plates of augite and resemble parts of the Hillhouse Sill.

The largest outcrop of dolerite in the Irvine district occupies some 16 km² centred upon Dundonald [NS 36 34]. It is probable that two sills are present. The area between Dundonald, Ploughland [NS 364 360] and New-field House [NS 379 348], together with an outlier to the east at Caprington Castle [NS 406 362], is occupied by a sill of analcimedolerite (teschenite) intruded into the Middle Coal Measures at the level of the Kilmarnock Parrot Coal. This is termed the Caprington Sill. The Kilmarnock Parrot, Kilmarnock Turf and Upper Wee coals have been removed or rendered unworkable and the overlying Finnie's Main or Caprington Blind Coal has been converted to anthracite. The area of the Dundonald hills to the west and south of Dundonald is occupied by the Hillhouse Sill (formerly the Hillhouse Quarry Sill, Richey et al., 1930). This sill is over 58 m thick and forms an almost horizontal capping to the surrounding Middle-Coal Measures, with prominent scarp features on several sides. Good columnar jointing is seen in Hillhouse Quarry. The sill is composed of a variety of rock types, all containing nepheline in addition to analcime and hence could be classed as theralitic. The main types are a medium-grained olivine-rich non-ophitic nepheline-dolerite (theralite); and a coarse-grained, less olivine-rich, ophitic nepheline-gabbro or monzogabbro (essexite) with large plates of green-rimmed titanaugite up to 5 mm across. Some fine-grained basanites also occur. Outcrops around Dundonald Castle [NS 364 345] are shown on the map as continuous with the Caprington Sill. However, petrographically they resemble the Hillhouse Sill and they may represent a faulted outlier of this higher sill. The Hillhouse Sill extends eastwards through Dankeith [NS 383 333] and a small sill southeast of Rowanhill [NS 387 343] resembles the coarse-grained ophitic parts of the larger sill. However, other small sills at Ditton [NS 408 342] and Spittalhill [NS 404 334] are more like the analcime-dolerites of the Caprington Sill.

Two sills are intruded close to the base of the Upper Coal Measures at Troon where they dip gently to the south-west. The upper, and larger, forms the Troon promontory and consists of coarse-grained ophitic alkali olivine-dolerite containing analcime, green-rimmed ophitic pyroxencs and some alkali amphibole. A few thin coarse-grained syenitic segregation veins occur and a chilled top is seen in contact with disturbed sedimentary rocks on the south foreshore. A smaller sill of fine-grained analcime-dolerite (teschenite), is intruded a little below the large sill.

Type 3 Analcime-basanites

The only example of this type in the Irvine district forms the Castle Hill [NS 232 423] at Ardrossan and extends southwestwards as far as the Ardrossan IIarbour Fault on the foreshore. It shows good columnar jointing, with low-angled curving columns in one place. The rock has been described as a very fine-grained olivine-basalt of transitional Dalmeny/Hillhouse type, but with monchiquitic affinities (Richey et al., 1930), presumably because it has a high proportion of glassy groundmass and little visible feldspar. It is probably best described as an analcimebasanite. A coarse-grained rock with similar mineralogy crops out on an islet south of the limestone outcrop on the foreshore south-west of Ardrossan Harbour [NS 225 416].

Sills of Palaeogene age

Several extensive outcrops of coarse-grained, ophitic, analcime-bearing olivine-dolerite (formerly known as crinanite) in the Mauchline Basin (Eyles et al., 1949) have been shown to be part of a continuous Prestwick–Mauchline Sill Complex (Mykura, 1967). The complex has a maximum thickness of 60 m and is believed to consist of between one and three leaves, up to 500 m apart, connected by dykes or steeply inclined sheets. The petrographical similarity to sills of Palaeogene age on Arran suggests a similar age for the complex and this has been supported by palaeomagnetic measurements (Armstrong, 1957) and a K/Ar mineral date of 58.4 ± 1.4 Ma (De Souza, 1979).

Only the north-western part of the complex crops out in the Irvine district, in the Rumbling Burn area southeast of Troon. The nearest surface exposure is at Monktonhill [NS 34 29], just to the south of the district. Boreholes and underground information show that in this area the sill is between 13 and 34 in thick and that it is emplaced some 6 in above the Ayr Ell Coal, near the top of the Middle Coal Measures. Farther east, outwith the district, the complex cuts the Upper Coal Measures, the Mauch-line Volcanic Formation and the Mauchline Sandstone Formation. The nearby Troon sill has very similar petrographical features, but is not considered to be part of this complex.

Dykes

The diversity of composition and trend among the dykes is such that a simple age-based subdivision is impossible to adopt. However, it is likely that they were emplaced during Visean, Silesian to early Permian, and Palaeogene times.

Dyke at Farland Head

A single dyke of carbonated basalt is intruded into the Sandy's Creek Formation on the Hunterston peninsula (Chapter 3). It is contorted, like the host sediment. It may be of early Devonian age.

Dykes of Viséan age

Impersistent dykes from 0.5 to 3 m wide, with compositions corresponding to the full range of rock types in the Clyde Plateau Volcanic Formation, form a north-east-trending swarm passing through the Misty Law Trachytic Centre and the thickest part of the Renfrewshire Hills lava sequence (Tyrrell, 1917a; Richey, 1928; Johnstone, 1965; Paterson et al., 1990). They are particularly concentrated in the lower parts of the lava succession and the underlying sedimentary rocks. Most of the dykes trend between north-east and east, but other trends are also common. This swarm extends into the Irvine district around Largs and on Great Cumbrae, where it is particularly well seen (Tyrrell, 1917a; Caldwell, 1973; Lawson and Weedon, 1992).

Characteristic of the Great Cumbrae swarm as a whole are porphyritic albite-trachytes (formerly known as 'bostonites') and finer-grained felsites, both commonly exhibiting well-developed fluxion banding. However, macroporphyritic olivine-basalt dykes are probably the most numerous. Cross-cutting relationships of dykes are commonly exposed which, together with trends and distinctive petrological characteristics, have enabled classification into several intrusive phases (Tyrrell, 1917a; Caldwell, 1973). Starting with the oldest, these are:

i.        basaltic plugs, all to the west of the Irvine district

ii.      dykes, sills and small bosses of albite-trachyte, generally trending east-north-east

iii.    dykes of macroporphyritic olivine-basalt of Markle and Dunsapie type, generally trending east-northeast

iv.    dykes of albite-trachyte and felsite, generally trending east-north-east

v.      a few dykes of microporphyritic basalt of Jedburgh type trending east-north-east to north-east.

The dykes are particularly well exposed in a section on the east coast of Great Cumbrae. At the north end of the section over sixty dykes are seen in a little over 1 km. Detailed accounts of the field relationships and petrography were given by Gunn et al. (1903) and Tyrrell (1917a). Two analyses of trachyte dykes from Smedley (1986a) are given in (Table 8) and are included in (Figure 8).

To the south-east of the Great Cumbrae swarm, Visean dykes are thinly scattered. Notable examples are a northnorth-west-trending dyke transitional between Markle and Jedburgh types intruded into the base of the lavas just east of Bradshaw [NS 24 53]; an east-north-east-trending dyke of Dunsapie type in Routdane Burn [NS 26 57]; a northwest-trending dyke of Craiglockhart type beneath the lavas at Fairlieward [NS 22 56], which is cut by a north-east-trending felsite dyke; an east-north-east-trending mugearite dyke, 500 in south-east of the last locality; and two dykes of hornblende-phyric trachybasalt trending northeast and east-north-east at Barcraigs Reservoir [NS 38 57] and north-east of Shutterflat [NS 39 54] in the Beith Hills. Felsite dykes generally have a north-easterly trend, but a few in the Burnt Hill and Brown Hill areas trend northwest. They seem to be later than the basic dykes (Richey et al., 1930), and are most numerous close to the Misty Law Trachytic Centre.

Quartz-dolerite dykes of late Carboniferous age

The northern part of the Irvine district lies on the southern edge of the major swarm of east- to east-north-easttrending quartz-dolerite dykes which extends across the south-west Highlands and central Scotland into the North Sea (Walker, 1934; Macdonald et al., 1981; Russell and Smythe, 1983). In the offshore part of the Fife Coalfield, members of the swarm cut Duckmantian (Westphalian B) strata and, in the west Highlands, quartz-doleritc dykes are cut by dykes of the Permian camptonitemonchiquite suite. Whole-rock K/Ar dates range from 302 to 297 Ma (Fitch et al., 1970) giving a late Westphalian to early Stephanian age for the swarm.

Only two major dykes of this swarm crop out in the Irvine district. The northern one extends east-north-east from the hairpin bends above I.args on the A760 road [NS 2126 5852] to the southern slopes of the Gogo Water valley in the adjacent Greenock district. It cuts the Fairlie Sandstone Formation and the lower part of the Clyde Plateau Volcanic Formation. The southern dyke is called Jenny's Dyke and extends east-south-east from the coast at Portencross [NS 1755 4922] to West Kilbride, and then eastwards as a series of disconnected, possibly en echelon, sectors to the Knockewart Hills. It cuts the early Devonian Portencross Formation, the late Devonian Seamill Sandstone Formation, the Law Hill vent and a basalt plug. Thinner dykes of finer-grained tholeiitic dolerite/basalt to the north and south of Jenny's Dyke at Farland Head may he part of the same swarm.

The major dykes are typical coarse-grained quartzdolerites, consisting of labradorite laths, pscudomorphs after hypersthene and iron-titanium oxides, all partially enclosed by sub-ophitic augite. The groundmass is a rnesostasis of quartz and alkali feldspar, commonly inter-grown as micropegmatite. The smaller, finer-grained dykes have a glassy groundmass and would formerly have been classed as 'tholeiites'.

Alkali basalt and alkali dolerite dykes of post Vise= age

Numerous dykes representative of alkali-basaltic magma types occur in the Irvine district, in addition to those of Visean age. They are exposed in coast sections, and are particularly well recorded in the Ayrshire Coalfield, where they were commonly encountered in underground workings. They are clearly younger than the Coal Measures. Some dykes are probably contemporaneous with the Silesian to Autunian sills of the area; some have undoubted petrographical affinities with the Autunian lavas and vent intrusions of the Mauchline Volcanic Formation; and many must be part of the extensive Palaeogene dyke swarms, either the major regional swarm centred upon Mull or part of a separate sub-swarm related to the central intrusive complexes of Arran (Speight et al., 1982). Dykes of all the groups can be very fresh but petrography is not a reliable indicator of age, with the possible exception of the camptonitic and monchiquitic types which can be assigned to the Autunian volcanic activity. Some attempt has been made to divide the dykes on the basis of their trends and observed cross-cutting relationships with other dykes and with faults (Eyles et al., 1949; Mykura, 1967). However, such classifications only give general groupings and individual dykes cannot usually be assigned with confidence to any particular swarm or age. The dykes have been divided for descriptive purposes into petrographical types (Richey et al., 1930; Eyles et al., 1949) and this approach is followed here.

Olivine-dolerites

Most of the post-Visean dykes of olivine-dolerite in the Irvine district are analcime-hearing and the majority are olivine-rich 'crinanitic' types with coarsely ophitic titanaugite. However, none has sufficient analcime to warrant classification as analcime-dolerite (teschenite). Hence there are no dykes directly comparable to the 'teschenitic' suite of sills, which forms a conspicuous feature of the Midland Valley of Scotland as a whole (Eyles et al., 1949). Richey et al. (1930), however, described two east- to eastsouth-east-trending dykes which appear to rise from a teschenitic sill intruded between the Ladyha' and Upper Wee coals in the Lower to Middle Coal Measures between Eglinton Park and Sourlie. In underground workings, both dykes cut coals above the sill, but neither was encountered below the sill. At Ardrossan a north-west-trending olivine-dolerite dyke cuts a teschenitic sill on the north-west shore of South Bay [NS 2317 4191]. North-east-trending dykes of this type are quite common; one crops out on the coast 1 km north-west of Ardrossan and another is exposed in an old railway cutting at Whitespot [NS 345 509], 3 km south of Beith. In the coalfield east of Irvine, three prominent north-north-east-trending dykes have been traced for up to 5 km in underground workings. Unfortunately, their relationship to equally prominent east–west dykes in the area, and to the sills, is not known.

Olivine-dolerite dykes range from 2 to 6 m in width and, as indicated above, occupy a wide range of fracture directions. Many, however, have a north-westerly trend and lie within the projected continuation of the north-west-trending Palaeogene dyke swarm from Mull. In adjacent districts to the south, north-west-trending olivine-dolerites cut the Mauchline Sandstone and this evidence, coupled with their 'crinanitic' petrography, led Tyrrell (1923), Richey et al. (1930) and Eyles et al. (1949) to suggest that the majority of north-west-trending olivinedolerites are of Palaeogene age. However, some northwest-trending dykes, as for example on the coast near West Kilbride, are cut by dykes with an east-west trend. This suggests that these north-west-trending dykes must be of late Carboniferous or Permian age. Conversely, not all Palaeogene dykes are north-west-trending; some that converge on the central complexes of Arran probably have trends between north-east and south-southeast.

Camptonitic basanites

A few basic dykes have camptonitic affinities as shown by large idiomorphic titanaugite and small crystals of brown amphibole. The rocks have interstitial analcime and are probably basanites. All trend west-north-west or east–west. Notable examples occur south of Ardrossan Harbour and as two separate dykes between Dreghorn and Dry-bridge, these two being traceable for some distance underground. Camptonitic rocks are known from the Autunian Mauchline volcanic rocks of south Ayrshire (Richey et al., 1930) and east–west-trending camptonitic dykes of broadly similar age occur in the Scottish Highlands. Hence the camptonitic dykes of the Irvine district are also probably Autunian.

Monchiquitic analcime-basanites

Dykes of this type consist essentially of idiomorphic titanaugite and lesser amounts of olivine set in a groundmass of analcime with little or no feldspar. They are probably common in a group of east–west-trending dykes which cut Middle and Upper Coal Measures in the Irvine valley. Most notable are a dyke which crops out on the right bank of the River Irvine, west of Caprington Castle [NS 406 362], and one which cuts a teschenitic sill in an old quarry at Craig [NS 37 37], south of Springside. Their close petrographical affinities to the lavas and associated vent intrusions of the Mauchline Volcanic Formation leaves an Autunian age in little doubt.

Tholeiitic basalt, dolerite and andesite dykes of Palaeogene age

Igneous rocks of Carboniferous and Permian age in the Midland Valley of Scotland are almost all members of various alkali basalt series. This dominantly alkaline activity was interrupted by only one episode of tholeiitic activity in the late Carboniferous, which is represented in the Irvine district by a few persistent and petrographically distinctive east-trending dykes of quartz-dolerite described above. All other dykes of tholeiitic affinity in the district are assumed to be of Palaeogene age. (Table 17) gives six analyses of dykes belonging to this group.

The tholeiitic dykes trend predominantly between west-north-west and north-west although significant numbers have trends between north-north-east and east. They all lie within the regional linear dyke swarm which extends from the Palaeogene central complex of Mull, across the south-west Highlands and into west central Scotland. Persistent members of the swarm have also been traced south-eastwards across southern Scotland and into north-east England. Most of the dykes of this swarm are tholeiitic, although alkali basalts also occur. Some of those described above, in particular those that trend north-west, are probably members of this regional swarm. The tholeiitic (and alkali basalt) dykes with trends other than west-north-west or north-west are more likely to be members of a separate sub-swarm centred upon the central complex of Arran. In the Stevenston–Kilwinning area, Richey et al. (1930) noted that the north-west-trending tholeiitic dykes commonly carry phenocrysts of calcic plagioclase, but that those with other trends are non-porphyritic, suggesting separate periods of intrusion.

Direct evidence for the age of the tholeiitic dykes is limited. Within the Irvine district and adjacent areas, tholeiitic dykes cut Coal Measures strata, east-trending quartz-dolerite dykes, teschenitic sills, vents and plugs related to the Mauchline Volcanic Formation, and the Mauchline Sandstone above the volcanic rocks. On Arran they cut Triassic strata, and in north Yorkshire the Cleveland Dyke, which is almost certainly a member of the Mull swarm, cuts Lower Jurassic strata. Evidence from Arran indicates that alkali dolerite dykes (and sills) are earlier than the Northern Granite and Central Ring Complex, all of which are cut by tholeiitic dykes. Age relationships between the Arran sub-swarm and the Mull regional linear swarm cannot be established with certainty; the two probably overlapped in time, although it is likely that the Mull swarm continued to be active until much later.

The dykes of this group range in composition from tholeiitic basalts and dolerites to quartz-dolerites, basaltic andesites and andesites, with the thicker more persistent dykes having, in general, the more fractionated compositions. The basic rocks are composed essentially of labradorite laths and augite, with varying amounts of glassy mesostasis which is commonly devitrified and darkened by finely disseminated iron-titanium oxides. Both olivine-bearing and olivine-free rocks are represented amongst the more basic rocks and the more siliceous rocks commonly contain orthopyroxene or pigeonite. Some contain phenocrysts of anorthite and/or groundmass labradorite whilst still retaining an overall andesitic composition, due to the presence of a siliceous glassy groundmass.

The petrography of the dykes in Ayrshire has been described in detail by Tyrrell (1917b) and by MacGregor (in Richey et at, 1930; and in Eyles et al., 1949), whilst the proposed continuation of the more persistent dykes into northern England has been described by Holmes and Harwood (1929). All of these authors used a plethora of local names, many of them derived from Mull, to describe the wide variety of rocks (Bailey et al., 1924). Published modern studies of the dykes in central and southern Scotland are lacking, although some details of their petrochemistry based on incomplete thesis work are given by Thompson (1982, p.470), and detailed studies of the Cumbrae-Stevenston and Cleveland dykes throughout their length are described by Hornung et al. (1966) and Macdonald et al. (1988).

A striking feature of the Mull regional linear swarm as it crosses west central Scotland is its sharply defined north-eastern edge. This is marked in the Greenock district (Paterson et al., 1990) and in the Irvine district, from Westhills Hole [NS 344 578] to Lugton [NS 41 52], by a group of thick regionally persistent dykes in a zone 3 km wide. To the north-east of this zone, dykes of Palaeogene age are virtually absent, but to the south-west, numerous mainly short lengths of dyke occur. The south-western edge of the swarm probably coincides more or less with the coast between Troon and Portencross, making the swarm some 20 km wide in this area. Dykes of the Arran sub-swarm undoubtedly occur within the area of the Mull swarm, hut also extend farther to the south-west outwith the Irvine district. The more persistent dykes are commonly 6 to 15 m wide and the large regionally persistent ones reach up to 37 m wide in places. Smaller dykes are up to 3 m wide. Most of the dykes have well-developed columnar jointing perpendicular to obvious chilled margins. The rock is commonly fresh, hard, and resistant to erosion.

Regionally persistent dykes

The Moneyacres Dyke, at the north-eastern edge of the Mull swarm, has been traced from Wemyss Bay, through the Misty Law Trachytic Centre and across the Irvine district (with gaps) from Cockston Farm [NS 320 585] to Bardarroch Craig [NS 333 578]. It is then exposed in two lengths across the Beith-Barrhead Hills, one passing through Loanhead Quarry [NS 365 554] (Gribble, 1992) and one near Over Hessilhead, and it has been quarried in Middleton Quarry [NS 403 526], west of Lugton. To the south-east, in the Kilmarnock district, it is exposed around Moneyacres [NS 42 50] and north of Darnel. From there, lengths of dyke with a similar petrography have been identified at. intervals to Acklington in Northumberland. The dyke is from 15 to 24 in wide and is a basaltic andesite containing labradorite, sub-ophitic augite and hypersthene in variable amounts of glassy mesostasis with small quartz patches ((Table 17), analysis 2).

Some 3 km to the south-west of the Moneyacres Dyke, the Barrmill Dyke crops out in several lengths in the area north of Pundeavon Reservoir, in road metal quarries between Beith [NS 34 53] and Barrmill [NS 36 51], and in the area south of Dunlop [NS 40 49]. In the adjacent Kilmarnock district the dyke can be traced with only minor breaks in continuity to Darnel. Dykes farther to the southeast which have been correlated with the Barrmill Dyke occur at various places as far as Seaton Sluice in Northumberland. The dyke is from 15 to 37 m wide and is an andesite with labradorite, granular augite and pigeonite or hypersthene in a quartzofeldspathic groundmass, Micropegmatite intergrowths and biotite are common in the groundmass.

Between the Moneyacres and Barrmill dykes in the Irvine district are several locally persistent dykes, mostly of tholeiitic basalt and dolerite, up to 1 km long. Between Cockston [NS 320 585] and Ladyland [NS 323 579], to the southwest of Loanhead Quarry and south-east of Bottoms [NS 373 544] are lengths of dyke, 9 to 15 m wide, of olivine-free tholeiitic basalt. Other dykes are mostly olivine-bearing, such as the tholeiitic olivine-dolerite of the Middleton Dyke ((Table 17), analysis 1). These dykes have been equated by Holmes and Harwood (1929) with a group of olivine-bearing tholeiitic dykes close to the north-eastern edge of the swarm in the Keilder Head and Morpeth areas.

Close to the south-western edge of the Mull swarm, the Cumbrae-Stevenston Dyke is the most siliceous and also the most petrographically distinctive dyke of the whole swarm (Tyrrell, 1917a). A dyke of this type occurs at Toward Point on the Cowal Peninsula and several, with a general north-north-westerly trend, traverse Great Cumbrae, notably that which forms the prominent Lion Rock [NS 180 549] (Tyrrell, 1917b). On the mainland at Stevenston a single north-north-west- to north-west-trending dyke has been traced, mainly from mining information, from Drummilling Hill to the shore between Saltcoats and Irvine. In Caponcraig Quarry [NS 262 420] the dyke, known to local miners as the 'Caponcraig Gaw', is seen to cut two teschenitic sills. Farther south-east the dyke is exposed at the Stinking Rocks [NS 32 33], Barassie, and on the same trend has been traced by mining information beyond the present district as far as Coylton. From there the dyke appears to swing eastwards, through two isolated outcrops, to the vicinity of New Cumnock, where it is deflected north-eastwards by the Southern Upland Fault before resuming a south-easterly course for a short distance. An undeflected south-easterly projection of the Stevenston–Coylton sector would pass through two outcrops in the Southern Uplands and then align with the Cleveland Dyke, which extends from Armathwaite in the Vale of Eden, across the Pennines and the Cleveland Hills to near Whitby, a distance of some 300 km from Cumbrae. In the Irvine district the dyke varies from 12 to 18 in in width. It is essentially andesitic in composition, but it is characterised by abundant phenocrysts, up to 3 mm long, of anorthite set in a groundmass of labradorite laths and granular augite. Some glass and patches of quartz are present and the rock is best described as a glass-bearing, anorthite-phyric augite-andesite. A westnorth-west-trending andesite dyke of similar appearance in the Kirkland Glen, 400 m west of Haupland Farm [NS 224 466], has labradorite phenocrysts and dominant orthopyroxene.

Although intermittent lengths of dyke, aligned over a considerable distance, can be correlated by means of distinctive petrographical features, it has for long been recognised that many also show considerable variations in tex ture and/or composition (Hornung et al., 1966). A study by Macdonald et al. (1988) of the Cumbrae–Stevenston and Armathwaite–Cleveland dykes has shown that they are not comagmatic and hence that they must have originated from the Mull Centre as separate pulses of different magma, which were emplaced in separate sectors of an aligned fracture system.

Other tholeiitic dykes

Smaller tholeiitic dykes within the Irvine district are almost all of tholeiitic basalt. They are rarely more than 3 m wide and cannot he traced for any distance in surface outcrops, although dykes have been traced for up to 3 km in underground workings. Dykes with north-west to west-north-west trends are by far the most abundant (Richey et al., 1930, p. 296), but many other trends are known. Most notable are the abundant east- to east-north-easttrending dykes, up to 6 m wide and commonly following fault lines, which have been encountered in workings of the northern Ayrshire Coalfield. These are the only tholeiitic dykes for which a Palaeogene age could be questioned. In the absence of analyses they could be fine-grained equivalents of the late Carboniferous quartzdolerite swarm, which are common in some parts of central Scotland. However, they are not associated with any coarse-grained quartz-dolerites of this type and are well to the south of the accepted limit of the swarm. Since no meaningful cross-cutting relationships have been observed (for example with sills or dykes of undoubted Permian age) their age is currently taken as Palaeogene.

A wide range of petrographical types has been described, from olivine-free or olivine-poor augite-plagioclase rocks with considerable glassy or very fine-grained mesostasis, to coarser-grained ophitic dolerites with fairly abundant olivine. Porphyritic varieties, with anorthitebytownite or bytownite-labradorite phenocrysts, occur throughout the range, but they are a minority.

Chapter 13 Geophysics

The Midland Valley of Scotland is characterised by strong geophysical anomalies in the gravity and magnetic domains and by a distinctive upper crustal velocity structure defined by numerous refraction experiments. Geophysical anomalies generally reflect a combination of effects, some related to the exposed sequences and some to deeper crustal structure. The most useful features of the geophysical data are those that relate to the structure and extent of Lower Carboniferous volcanic activity and the development of the Upper Devonian and Carboniferous sedimentary sequences.

The seismic model of the upper crust in the Midland Valley consists of three layers:

• Layer 1 Carboniferous and Upper Devonian rocks up to 4 km thick
• Layer 2 Lower Devonian rocks and Lower Palaeozoic strata
• Layer 3 Strata with a velocity of about 6.0 km s⁻¹, not exposed but considered to be crystalline basement.

The crystalline basement is thought to change to a higher grade of metamorphism at about 8 km depth.

Within the Irvine district, aeromagnetic data relate mainly to the depth and extent of the magnetic Clyde Plateau Volcanic Formation. Strong local anomalies in the total field coincide with local volcanic centres and can he interpreted in terms of relatively thin (1 km) sheets of basaltic lava, with basic intrusions beneath the main volcanic centres, especially the Hill of Stake.

Across much of the Midland Valley, the main outcrops of Upper Carboniferous strata are associated with Bouguer gravity anomaly lows, the anomaly contour pattern frequently showing a close correlation with the faults that controlled Carboniferous sedimentation. To a large extent, however, this general pattern is less clear in the western part of the Midland Valley, including the Irvine district.

Within the district, and north of the Inchgotrick Fault, Bouguer gravity anomalies increase generally to the south-west. Local gravity highs occur over known volcanic centres at Hill of Stake and Dundonald, and over minor intermediate intrusive rocks in the Clyde Plateau Volcanic Formation. The main gravity low at Kaim Hill occurs over the Upper Devonian Stratheden Group, of unknown thickness.

Gravity data

McLean (1966) collected and interpreted gravity data from Ayrshire, concluding that south of the Inchgotrick Fault the gravity data reflected the known structure of the Upper Devonian and Carboniferous sequence but that north of the fault the correlation was less clear. In north Ayrshire, he suggested the existence of marked west-north-westerly changes in the Stratheden Group to Clyde Plateau Volcanic Formation succession, with thickening of the Stratheden Group or thinning of the lavas or both, superimposed on an observed thickening of the lava pile north-north-eastward from Ardrossan. A marked change in thickness of Carboniferous strata across the Dusk Water Fault was also inferred. The significant gravity low to the north of West Kilbride showed good correlation with the surface geology and was considered to reflect thinning of the Clyde Plateau Volcanic Formation to the north-northwest with a complementary rapid thickening of the Upper Devonian sequence. The low was subsequently re-interpreted by McLean and Walker (1978) as a basin of Devonian sedimentary rocks up to 1600 m thick. The lack of a significant gravity anomaly low over the Kilmarnock Basin (in the eastern part of the district and beyond) was attributed to a higher density for the Carboniferous sequence there, possibly related to the presence of igneous material in the sequence. A local gravity high near Troon and a linear high trending east-north-eastnorth of the Inchgotrick Fault were attributed to dolerite intrusions (Hillhouse Sill) or a buried ridge of Namurian lavas.

Bouguer gravity anomalies across the western Midland Valley reduced to OD at a density of 2.75 Mg ITO are shown in (Figure 25). The most significant gravity low is that north of West Kilbride, where a closed minimum with residual anomalies up to 10 mGal below the local background occurs around Kaim Hill [NS 227 535]. The broad shape of the anomaly and local inflexions in the contours are directly related to significant faults, especially the Dusk Water Fault, the Ardrossan Harbour Fault, the Largs–Huntcrston Fault Zone and the south-west extension of the Paisley Ruck. A regional analysis of the Bouguer gravity data suggests that the Ardrossan I Iarbour Fault is part of a significant basement structure extending to the south-east across Ayrshire and the Southern Uplands to the west side of the Permian Dumfric basin.

A closed Bouguer gravity anomaly low occurs over the small outcrop of Namurian strata south of the Annick Water Fault and north of Kilmarnock but in general the Bouguer gravity increases to the south-west across the Westphalian outcrop to local maxima at Troon and directly over the Dundonald vent. New data over the Renfrew Hills east of Largs indicate a significant closed positive anomaly centred on the Hill of Stake (Figure 25). This has been interpreted as a buried basic intrusion with a depth extent of about 3 km.

There is a general increase in Bouguer gravity anomalies across the west side of the Midland Valley. A strong gradient in the Bouguer gravity anomaly field, locally associated with the Dechmont Fault, can be traced northwestwards and then westwards across the Renfrew Hills to the Clyde coast at Wemyss Bay. A basement high, or a more dense basement lithology, is presumed to underlie all the Midland Valley west and south-west of the Dechmont feature. The structure is considered to have had important control on the development of Carboniferous sedimentation and volcanicity.

Aeromagnetic data

Analogue total field acromagnetic data were collected over the Midland Valley at a mean terrain clearance of 305 m. These data show a series of high amplitude, high frequency anomalies superimposed on several long wavelength (about 30 km) anomalies from deeper sources and long wavelength regional trends (Institute of Geological Sciences, 1967).

(Figure 26) shows a map of the aeromagnetic data across part of the western Midland Valley based on the observed data gridded at 0.5 kin, reduced to the pole assuming induced magnetisation and analytically upwardly continued to 500 m above observation level to simplify the forms of the anomalies. Strong positive anomalies are clearly associated with the sites of the major central vent eruptions and the zones with local exposures of intermediate intrusive rocks. North of West Kilbride the anomaly pattern reflects the outcrop of the Clyde Plateau Volcanic Formation. Significant anomalies occur to the south-west of Beith just north of the Dusk Water Fault and horizontal gradients in the field trend east–west through Irvine and approximately parallel to the Inchgotrick Fault. A linear east–west anomaly offshore from Troon is associated with the intrusive rocks of Lady Isle [NS 275 294].

Maximum anomalies are sited above the Hill of Stake centre to the north of the district and near Hartfield Moss [NS 420 560] and Neilston Pad [NS 475 551], to the east. Close to these sites there are mapped exposures of intermediate and basic intrusive rocks within the Clyde Plateau Volcanic Formation, and the basic rocks are considered to be more extensive at depth.

Simple modelling of the polar magnetic field across the western Midland Valley, using large prismatic shapes and assuming a mean susceptibility of 0.06 SI units for the lavas, with thicknesses in the range 500–1000 in, fails to generate the local maxima observed above the volcanic centres, and these areas must therefore have thicker igneous sequences or greater than average magnetisation.

The form of the aeromagnetic anomalies indicates that the Clyde Plateau lavas extend south from the Renfrew Hills at least as far as the Inchgotrick Fault, with clear indications of dislocation of the upper surface of the lavas at the Dusk Water and Annick Water faults. Away from the exposed Clyde Plateau Volcanic Formation, maximum anomalies occur in the zone north of the Dusk Water Fault, between Dairy and Beith.

Seismic data

A velocity model for the upper crust of the Midland Valley was first established by the LOWNET seismometer array (Crampin et al., 1970) using quarry blasts as sources. A layer with velocities in the range 3.0–5.7 km s⁻¹ was attributed to Palaeozoic rocks lying above a refractor at about 8 km depth. This model was refined and confirmed by the LISPB experiment (Bamford et al., 1978; sec also below) across the eastern part of the Midland Valley from Dunkeld through Edinburgh.

The results of three east to west refraction lines (Davidson et al., 1984) using quarry blast sources suggest that crystalline basement is present beneath the southern Midland Valley at depths of only 3 to 4 km. An 80 km line (GLA–L3) in the south-west of the Midland Valley, across the Irvine district and the exposed Lower Palaeozoic inliers, identified a 6 km s⁻¹ refractor at about 3 km depth, taken to represent the base of the Lower Palaeozoic sequence. It is postulated that this refractor extends across the Southern Upland Fault at about 2.5 km depth and matches that defined as the top of a seismically fast layer (F₃, Hall et al., 1983) on the Southern Uplands Seismic Refraction Profile. West of the Inchgotrick Fault this refractor is mapped at a depth of about 2 km.

The occurrence of a 6.0 km s⁻¹ refractor on two refraction lines some 30 km apart was used to infer the presence of amphibolite gneiss basement lithologies at depths of about 4 km across much of the central Midland Valley. The crucial arguments supporting shallow crystalline basement are, firstly, the presumed hiatus in the observed velocity spectrum between 5.8 and 6.2 km s⁻¹ and, secondly, the presence of a refractor beneath the Lower Palaeozoic inliers.

The MAVIS seismic refraction experiments (Conway et al., 1987; Dentith and Hall, 1989) have supported the overall crustal model described by Davidson et al. (1984). Three refractors were identified along two east–west lines. The uppermost refractor at depths of between 1.7 and 3.8 km defined the base of the Carboniferous and Upper Devonian sequence (layer 1). Layer 1 velocities ranged from 3.5 to over 5 km s⁻¹, the higher velocities corresponding to known occurrences of thick Carboniferous lavas. Layer 2, with velocities of 5.3 to 6 km s⁻¹, extends to depths of between 3.5 and 5 km on the south line and is presumed to consist of Lower Devonian and Lower Palaeozoic rocks (Conway et al., 1987). The MAVIS results suggest that the base of layer 2 is essentially uniform at about 4 km depth over much of the south and west Midland Valley, deepening to about 6 km north of the Campsie Fells and the Ochil Fault.

The most reliably observable velocity discontinuity in the upper crust of the Midland Valley appears to be that at about 7 to 8 km depth. This has been identified on all the longer refraction lines and is most likely to represent an important lithological break in the crust, probably the top of quartz-feldspathic granulite gneiss. This refractor, at such relatively shallow depths, is peculiar to the crust beneath the Midland Valley and the southern Highlands and is part of the geophysical signature of this region. The distinctive crustal structure, coupled with the absence of clasts of Dalradian material in Lower Palaeozoic rocks of the Midland Valley (Bluck, 1984), support the hypothesis of the Midland Valley as a 'suspect terrain', tectonically emplaced against the Grampian Highlands.

Physical properties

Rock densities

Exposed Lower Palaeozoic rocks in the southern Midland Valley can be expected to have mean saturated densities close to 2.72 Mg m⁻³ (McLean, 1961; Bott and Masson-Smith, 1960). Lower Devonian sandstones, conglomerates and shales exhibit densities in the range 2.52–2.64 Mg m⁻³ (McLean and Walker, 1978), with a mean density quoted by McLean (1961) of 2.60 Mg m⁻³. Lower Devonian lavas have a mean measured density of about 2.66 Mg m⁻³ while Upper Devonian sandstones in Ayrshire have a characteristic low density of about 2.42 Mg m⁻³ (McLean, 1961). Similar sandstones in the Glenrothes heat flow borehole, in Fife, have measured densities in the range 2.34–2.41 Mg m⁻³ with a mean value close to 2.40 Mg m⁻³ (Brereton et al., 1988). These sandstones generally produce a marked density inversion in the geological succession, especially where they occur beneath Dinantian lavas with a mean density of 2.72 Mg m⁻³ (Cotton, 1968; McLean and Walker, 1978). The Carboniferous sedimentary succession is generally assumed to have a mean density close to 2.50 Mg m⁻³ (Bott and Masson-Smith, 1960; McLean, 1961). The only significant exposed Devonian granodiorite in the Midland Valley, at Distinkhorn, is associated with a local gravity high and can be expected to have densities in the range 2.68–2.83 Mg m⁻³, on the basis of bulk chemistry.

Magnetic susceptibilities

The mean susceptibility of the macroporphyritic Markle-type lavas in the Clyde Plateau Volcanic Formation (0.056 SI units) is not significantly different from that of the microporphyritic Jedburgh flows (0.059 SI units), although the range of susceptibilities for the Markle flows is greater and the highest susceptibilities at outcrop (> 0.12 SI units) are generally observed on clearly macroporphyritic Markle flows. Laboratory measurement of lava susceptibility from 20 flows in the Campsie and Kilpatrick Hills (Cotton, 1968) gave a mean of 0.062 SI units with a range of 0.024 to 0.139 SI units. For the Clyde area, McLean and Wren (1978) give a range of 0.035 to 0.09 SI units.

The mean susceptibility of 11 sites along the Midland Valley Sill (Milton, 1972) was 0.022 SI units and the mean of two olivine dolerites in vents was 0.028 SI units. From field measurements made at Dungoil in the Campsie Hills, the mean susceptibility of four sites on the basic intrusive mass was 0.040 SI units with a small standard deviation (0.005 SI units). In contrast a much reduced mean field susceptibility (0.004 ± 0.002 SI units) was observed for the intermediate (andesitic) intrusion at Lochlands Hill [NS 380 550] in Renfrewshire. Similarly the intrusion at Neilston Pad has a field susceptibility of 0.007 ± 0.002 SI units, much lower than adjacent exposed (Markle-type) lavas at 0.063 ± 0.006 SI units.

Milton (1972) studied the palaeomagnetism of the Carboniferous rocks of the Midland Valley at 15 sites. The natural remanent magnetism (NRM) intensity at five sites on lavas was mostly less than 2 A m⁻¹, and a stable normal NRM (Declination= 21°, Inclination= −5°) was observed at only one site from the Limestone Coal Formation. This is one of the few normal polarity determinations from the British Carboniferous, most results indicating a reversed polarity in the Dinantian and Westphalian (Palmer et al., 1985). At 15 sites on the Midland Valley Sill, Milton (1972) observed a stable reversed NRM with D= 175° and I= 15°. Maximum intensities reached 7 A m⁻¹ implying a Q ratio above 6.

Sonic velocities

The LISPB experiment (Bamford et al., 1978) identified two important refractors in the upper crust of the Midland Valley. The top refractor separated Devonian and Carboniferous sedimentary rocks with velocities in the range 4.5–5.0 km s⁻¹ from assumed Lower Palaeozoic rocks with velocities of 5.8–6.0 km s⁻¹. At about 8 km depth, a lower refractor was interpreted as the top of a basement with velocity above 6.4 km s⁻¹. This refractor appears to extend north of the Highland Boundary Fault.

LISPB was not designed to examine the velocity structure of the top few kilometres of crust, but subsequent more detailed refraction experiments have examined the upper sedimentary sequence. Davidson et al. (1984) suggest velocities of 3.0–3.7 km s⁻¹ for the Carboniferous and Upper Devonian sedimentary rock and about 4.5 km s⁻¹ for the Carboniferous lavas. According to Davidson et al., velocities in the range 4.0–5.5 km s⁻¹ are characteristic of the Lower Devonian and Lower Palaeozoic sedimentary rocks but velocities in the range 5.7–5.9 km s⁻¹ are not recognised at surface in the Midland Valley. On this basis, refraction velocities above 6.0 km s⁻¹ were attributed to widespread crystalline basement at depths of about 3 to 4 km (see above).

Early measurements of seismic velocities for the Carboniferous rocks in south-west Scotland range from 2.66–3.48 km s⁻¹ (Hall, 1970), the maximum velocities representing the lavas. Subsequent work by Hall (1974) suggested a mean Carboniferous sedimentary rock velocity of 2.90–3.04 kin s⁻¹ with a velocity of about 3.9 km s⁻¹ for the Clyde Plateau lavas; a short reflection profile adjacent to the Rashiehill borehole was interpreted using a Carboniferous sedimentary rock velocity of 2.84 km s⁻¹ and a lava velocity of 4.0 kin s1 (Hall, 1971).

Velocities from sonic logs, well velocity surveys and moveout interval velocities calculated from seismic surveys are all generally higher than those measured by shallow refraction surveys. Carboniferous velocities are generally in the range 3.4–3.9 km s⁻¹. In the Glenrothes Borehole (Brereton et al., 1988), velocities in the Strathclyde and Inverclyde groups are close to 3.2 km s⁻¹, increasing to 3.9–4.0 km s⁻¹ in the Upper Devonian Stratheden Group sandstones.

Shear wave velocities can sometimes he useful indicators of rock composition especially when combined with the compressional velocity as a means of calculating Poisson's ratio. For most of the LISPB line Poisson's ratio has been calculated as close to 0.25, typical for upper crustal rocks (Assumpcao and Bamford, 1978). The main exceptions are in the upper crust of the Southern Uplands (0.231) and beneath the 6.4 km s⁻¹ refractor in the Midland Valley (0.224). In the Midland Valley, the low ratio below about 8 km depth implies quartz-feldspathic crystalline basement rather than a mafic lithology such as amphibolite gneiss.

Interpretation of bouguer gravity and aeromagnetic anomalies

Deconvolution

Deconvolution is a semi-automatic procedure to obtain estimates of the depth and position of steep discontinuities in the boundary surfaces of magnetic or gravity sources. In the western Midland Valley the aeromagnetic data have been analysed using focused Euler deconvolution (Rollin, 1995) of the observed data after upward continuation by 500 m (Figure 26).

The main features of the results are:

1.      Most of the shallow solutions less than 1 km deep are associated with extrusive Carboniferous rocks and known exposed intrusions.

2.      Many of the deeper solutions are related to significant faults, especially the Dusk Water, Annick Water and lnchgotrick faults. Near Irvine the analysis suggests that magnetic interfaces occur at depths of about 1 to 2 km adjacent to the Annick Water Fault.

3.      Some deeper solutions directly correlate with mapped small exposures of trachyte, trachyandesite and gabbro intrusions.

2.5D gravity and magnetic modelling

The gravity and magnetic data along two lines which cross the western Midland Valley have been modelled using an interactive 2.5D computer program (GRAVMAG, Pedley, 1991). This uses models of the crustal structure which generate theoretical profiles to match the observed geophysical profiles as closely as possible, using known rock densities and magnetic susceptibilities. The models shown (Figure 27) and (Figure 28) are possible solutions that fit the known regional geology as well as the geophysical data. Aeromagnetic data were modelled at the observation level, 0.305 km above the topography. The gravity models were calculated against a mean crustal density of 2.95 Mg m⁻³ and all structures to a depth of 40 km were modelled. The effects of lower crust and Moho topography were therefore included. The main density interfaces in the upper crust were set at about 5 km and 8 km to correspond with the evidence from the MAVIS seismic refraction experiments. The interface at a depth of about 5 km is associated with a density contrast of 2.74 to 2.80 Mg m⁻³, and that at 8 km with a density contrast of 2.80 to 2.85 Mg m⁻³.

The main results of upper crustal 2.5D integrated gravity and magnetic modelling in the western Midland Valley are:

1.      The closed gravity minimum at Kaim Hill north of West Kilbride can be explained as a wedge of Upper Devonian strata extending to about 1.5 km depth.

2.      The Inchgotrick Fault marks a local basement high with maximum Bouguer gravity anomalies on the northern downthrown side. Lower Palaeozoic rocks are modelled at depths of about 1 km at the fault. The main mass of Clyde Plateau lavas terminates against the fault and is modelled as about 500 m thick, perhaps resting directly on Lower Palaeozoic rocks.

3.      The Bouguer gravity anomaly profile shows minima close to the Dusk Water and Annick Water faults which have been interpreted as normal faults downthrown to the south in late Devonian times and to the north in late Carboniferous times.

4.      The Hill of Stake volcanic centre in the north Renfrew Hills is associated with a local positive residual gravity anomaly of about 5 mGal which can be interpreted as a gabbroic or basic intrusion to depths of about 3 km.

Ayr section [NS 140 800]–[NS 580 000]

The Ayr section extends from near Dunoon south-east across the Ayrshire coalfield to near the Cairnsmore of Cairsphairn intrusion in the Southern Uplands. It forms part of a crustal-scale section across northern Britain from the Southern Uplands to northern Skye. Within the Irvine area (Section A: (Figure 27)) the main features of the gravity profile are a strong low in the Clyde estuary (just north of the district), a positive anomaly over the north Renfrew Hills, a local low north of the Dusk Water Fault, a local maximum just north of the Inchgotrick Fault and a broad low across the coal basin south of this fault. The main features of the aeromagnetic profile are a series of strong anomalies with amplitudes above 150 mT, over much of the section between The North Renfrew Hills and the Inchgotrick Fault.

The Inchgotrick Fault has a significant downthrow to the north juxtaposing Namurian age sedimentary rocks against Upper Devonian sandstones which have a top surface at about sea level on the south side of the fault. The local maximum in the Bouguer gravity profile north of the fault is associated with doleritic intrusive rocks and agglomeratic vent material at Dundonald. The sharply defined nature of the local positive anomaly suggests a shallow source. The Dundonald anomaly could be caused by a largely unexposed volcanic centre and basic intrusion. However, there is a regional positive anomaly just north of the Inchgotrick Fault as far east as Kilmarnock and the model shown in (Figure 27) suggests that the fault forms the southern boundary to a structural high. The model indicates that Upper Devonian rocks on the south side might overlie a few hundred metres of Lower Devonian strata above Lower Palaeozoic rocks at about 1 km depth. North of the fault, the Clyde Plateau lavas are modelled as about 400 m thick over Lower Devonian strata which are thin, or even absent locally along the fault.

The details of the Bouguer gravity anomaly profile across the Inchgotrick Fault, just south of the Dusk Water Fault and at the Annick Water Fault, can be interpreted to suggest several phases of movement on these and related structures. Associated with these structures are local gravity lows which cannot be satisfactorily modelled assuming a simple normal fault structure with downthrow to the north. The northern downthrow is well established in the field and confirmed by the modelling of the upper surface of the lavas in accord with the aeromagnetic profile. In order to generate the observed gravity minima the Upper Devonian strata are assumed to thicken southwards approximately across these faults, possibly suggesting an earlier phase of movement. Locally Upper Devonian strata are modelled as absent beneath the Clyde Plateau Volcanic Formation.

North of the Inchgotrick Fault the aeromagnetic profile relates mainly to the depth to the upper surface of the Clyde Plateau Volcanic Formation but the thickness of the volcanic sequence is less well constrained. The modelled lava thickness is greatest just south of the Dusk Water Fault, where it is about 1400 m thick. This thickness is required if the aeromagnetic anomaly is to be satisfied even assuming a high mean magnetic susceptibility of 0.075 SI. However, part of the anomaly could be due to a magnetic basement lithology beneath the Clyde Plateau lavas or basic intrusions within this basement. Intrusions are modelled as partly responsible for local anomaly maxima in the profile further north-east.

North of the Dusk Water Fault a basin of Upper Devonian sedimentary rocks with a low mean density of 2.45 Mg m⁻³ and a thickness of about 600 m below the Clyde Plateau lavas can explain the observed gravity low north-east of west Kilbride. The aeromagnetic anomaly maximum over the north Renfrew Hills cannot be easily explained by the wedge of lavas which is assumed to thicken only gradually to the south. The coincident gravity high is explained by a composite intrusion through the Devonian sequence and into the base of the lavas. This intrusion is part of the Hill of Stake volcanic centre.

The gravity low in the Firth of Clyde is explained by 700 m of Upper Devonian strata (2.45 Mg m⁻³) over about 800 m of Lower Devonian strata. This modelled thickness of Upper Devonian strata is substantially less than indicated from summation of outcrop thickness but would be increased if a higher density was used for the Upper Devonian strata in this area.

Arran section [NR 700 300]–[NR 924 392]–[NS 566 653]–[NS 720 730]

The Arran section extends east-north-east from Argyll through Arran to Glasgow, based on the geological section shown on the 1:250 000 Solid Geology Sheet 55N 06W (Clyde). The section crosses the Bouguer gravity low north-east of West Kilbride. The eastern part of the section is close to the deep seismic reflection line collected as part of the BGS Urban Hot Dry Rock Programme (Penn et al., 1984).

For the Irvine area (Section B: (Figure 28)) the main features of the gravity profile are a gravity low in the Firth of Clyde, a local high around Farland Head, a low north of West Kilbride and broad high over the main Clyde Plateau lava outcrop north of the Dusk Water Fault. The aeromagnetic profile indicates significant maxima over the outcrop of the lavas.

The Carboniferous and Upper Devonian sequence beneath the Firth of Clyde (in the East Arran Trough) are modelled at least 2 km thick but thin rapidly eastwards across faults and are absent at the coast. Clyde Plateau lavas can explain most of the aeromagnetic anomaly west of the Ardrossan Harbour Fault but not the general level (50 nT) of the magnetic anomaly between here and the western limit of the lavas in the Renfrew Hills. The pre-Lower Devonian strata are therefore considered to be magnetic in this region.

The Upper Devonian sequence around Kaim Hill is modelled to depths of about 1.5 km beneath the Clyde Plateau lavas with a depth of about 1 km at the western edge of the lava outcrop. North-east of Beith at about 73 km along Section B, the aeromagnetic anomaly maximum of about 350 nT is interpreted as a basic intrusion into the base of the Clyde Plateau Volcanic Formation, which is about 1.2 km thick in this zone. The gravity profile is also at a maximum in this locality and suggests a local high in the Lower Palaeozoic basement.

Chapter 14 Structure

The origins of the Midland Valley can be traced back to the Caledonian Orogeny, and many of its structural lineaments probably date from that episode. These lineaments have since been re-activated at various times to accommodate changing stress regimes. The structural development of the Midland Valley has had a fundamental effect on the pattern of sedimentation, and the role of the major fault lineaments in determining the distribution and thickness of the sedimentary rocks deposited in the Midland Valley during the Upper Palaeozoic has been widely recognised (Goodlet, 1957; Bluck, 1978; 1980; 1984). In the Irvine district, the relationship between major north-easterly faults and the thickness and lithological character of the Carboniferous sedimentary succession was identified by Richey et al. (1930) and has been described further by Monro (1982a; 1984; 1985; 1986) and Monro et al. (1983). Volcanic episodes, such as those that formed the Clyde Plateau, Troon and Mauchline volcanic rocks, are also related to the stress pattern and provide important evidence about the structural development of the region.

Regional unconformities beneath the Lower Old Red Sandstone, Stratheden Group and Permian beds of the Midland Valley indicate major episodes of deformation in the late Silurian, mid-Devonian and early Permian respectively. The first two belong to the Caledonian Orogeny, and the third to the Variscan; most of the structures observed in the Irvine district belong to the last. Although the main Variscan movements took place after the end of the Carboniferous, movements on a lesser scale took place at intervals all through that period. The stress regime varied from time to time, and was resolved along pre-existing Caledonian structures, so that the orientation of the resulting structures is rarely related directly to the north-south compression that emanated from the Variscan Front.

Plate tectonic models

The evolution of the Midland Valley was discussed by Kennedy (1958) and George (1960) before the theory of plate tectonics became widely accepted. Kennedy postulated a compressional rift model while George considered that the Highland Boundary Fault was initiated as a pre-Arenig normal fault downthrowing to the north, with the Southern Upland Fault having originated no earlier than late Silurian.

Models for the evolution of the Scottish Caledonides and the early initiation of the Midland Valley structure in plate tectonic terms have been proposed by several authors (Dewey, 1969; 1971; Fitton and Hughes, 1970; Phillips et al., 1976; Stone, 1980). The post-orogenic history has been reconstructed using evidence from the Devonian rocks (Bluck, 1978; 1980), the Carboniferous volcanism (Macdonald, 1975; Macdonald et al., 1977; Francis, 1979), and the evolution of the adjacent Northumberland Basin (Leeder, 1971; 1974; 1976). Periods of east–west extension have been suggested by Leeder (1971), Haszeldine (1988) and Stedman (1988), while periods of north-south extension have been proposed by Leeder (1982) and Leeder and McMahon (1988). Dewey (1982) and Read (1988) suggested early Carboniferous stretching with right-lateral strike-slip movement followed by thermal recovery in the mid and late Carboniferous. Aspects of these models were applied to the evolution of the coal basins of central Scotland by Browne and Monro (1989).

Structural development in relation to volcanism

The Irvine district experienced three major volcanic events during the Carboniferous and early Permian. The eruption of the voluminous lavas of the Clyde Plateau Volcanic Formation may have caused a tectonic response in the crust through which the lavas emerged. This is evidenced by changes in thickness of the lavas across major faults, and may account for some of the thickness changes in the sedimentary rocks which overlie the lavas.

An association between volcanism and faulting in the Midland Valley is suggested by the line of necks along the Campsie Fault (Craig et al., 1975), by the complex history of the Ardross Fault (Francis and Hopgood, 1970) and by the thickness variations of the Clyde Plateau lavas (Hall, 1974). In the Irvine district, petrochemical studies of the Carboniferous lavas indicate two cycles of magma generation (Macdonald, 1975; Macdonald et al., 1977). The first, occurring from early Carboniferous to mid-Namurian times, relates to a tensional tectonic regime with normal faulting, while the second, from late Namurian to Permian times, is ascribed to a compressional environment with intermittent tension and continued normal faulting.

Francis (1979) attempted to put the Carboniferous volcanism into a plate tectonic setting. By early Carboniferous times the Iapetus Ocean had long been closed and the two flanking continents were welded into one. The Carboniferous volcanism is therefore considered as a mid-plate phenomenon. Volcanism in this tectonic environment is generally considered to relate to mantle plumes, which are stationary; as the crustal plate moves, the volcanic centres appear to move in the opposite direction. There is no evidence to suggest systematic migration of volcanic centres during the Carboniferous, however, and the processes connecting the volcanism with the structural evolution of the Midland Valley are still not fully understood. A simple tectonic regime can be suggested for the first of Macdonald's cycles, but the second cycle, with compression intermittently relieved, demands a more complex model.

Episodes of structural development

Devonian

Lower Devonian rocks crop out in two strips within the Midland Valley, one adjacent to the Highland Boundary Fault and the other adjacent to the Southern Upland Fault. Bluck (1978) concluded that during the early Devonian the margins of the Midland Valley sedimentary basin were not far from the present positions of these two bounding faults and envisaged sediment derived from a Highland source with a palaeoslope clipping to the southwest. Within the Irvine district the Portencross Formation, which is probably of early Devonian age (Chapter 3), contains a predominance of clasts derived from the Highlands. The geographical association of these lithologies with the Largs–Hunterston Fault Zone, a splay from the Highland Boundary Fault, suggests deposition in alluvial fans adjacent to active faults.

During mid-Devonian times, compressive earth movements produced marked changes in the palaeogeography of central Scotland. The Lower Devonian rocks were folded into north-east-trending structures and continuing uplift and erosion prevented the deposition of any Middle Devonian strata. The direction of the palaeoslope within the Midland Valley was reversed during this period, so that predominantly south-west flowing early Devonian rivers were replaced during the late Devonian by an axial drainage system directed generally towards the east-north-east.

Upper Devonian rocks rest unconformably on Lower Devonian strata over most of the Midland Valley but in the Irvine district the exposed contacts are faulted. Bluck (1980) considered that Upper Devonian deposition occurred in two sub-basins, one in the Midland Valley and one in Kintyre, separated by a north-north-west-trending upland ridge. He suggested that the two embayments developed different dispersal patterns, to the south-west in Kintyre and to the north-east in the Midland Valley proper, and envisaged eastward extension of the depositional basin by sinistral strike-slip faulting along the Highland Boundary Fault and step faulting within the basin. He considered that '... basin development in the Clyde is only a local expression of a wider structural event which encompasses most of Midland Scotland ...'. An earlier plate tectonic model (Phillips et al., 1976) also involves strike-slip faulting. Strike-slip faulting as a control of sedimentation may have had a continuing role into the Carboniferous.

Dinantian

Over much of the western part of the Midland Valley, the Clyde Plateau Volcanic Formation rests unconformably on a variety of rocks belonging to the Stratheden and Inverelyde groups. The phase of uplift which produced this unconformity has been recognised over a wide area (Paterson et al., 1990; Forsyth et al., 1996) and it has been suggested that the uplift may have resulted from magmatic updoming prior to the Clyde Plateau eruptions (Monro, 1982a; 1986).

The gravity data (Chapter 13) suggest that the thickness of some lithostratigraphical units changes across major faults, and that these variations may be attributable to contemporaneous tectonic activity. Hall (1974), using seismic techniques, showed that the lavas thin southeastwards towards the Dusk Water Fault, and thicken abruptly on its south-east side. His seismic line 69/3 suggested that the fault plane dips gently to the south-east. Hall concluded that in the Irvine district the lavas originally formed a north-west-elongated outcrop, thinning to the north-east and south-west and truncated by major north-east-trending Faults.

Thickness variations of the Lower Limestone Formation across the Dusk Water Fault (Figure 11) suggest that there was very little tectonic activity on the north-easterly faults in late Dinantian times.

Namurian

Richey et al. (1930) noted the influence of tectonism on Namurian sedimentation in the Irvine district in relation to the Inchgotrick Fault and to a lesser extent the Dusk Water Fault (Figure 15), (Figure 17). Thickness variations of Namurian strata (Figure 17) suggest that the Annick Water Fault was also active (Monro, 1985; 1986). All are north-east-trending structures (Figure 29).

The Limestone Coal Formation is condensed or absent south of the Inchgotrick Fault, and the Upper Limestone Formation is condensed and characterised by channel-fill deposition (Chapter 8). The Inchgotrick Fault therefore defines the northern margin of a block area, the southern margin of which is marked by the Kerse Loch Fault and the 'Line of Steep Measures' in central Ayrshire (Eyles et al., 1949), outwith the present district.

The Dusk Water Fault had a similar history during the Namurian. Downthrow to the north is suggested by thickness variations of early Namurian strata (Figure 17) but it was during the latter part of the Namurian that the fault had its greatest effect on sediment distribution. In the post-Lower Linn Limestone period the Dusk Water Fault marked the northern margin of an uplifted block area characterised by channel sandstone deposition, and the marine transgressions associated with the Upper Linn, Diddup and Corsankcll limestones were restricted to the area north of the fault (Figure 19).

Westphalian

Brand (1983) has shown that in the Irvine district lateral variation in the thickness of Westphalian strata is not dramatic. To the south of the district, however, between the Kerse Loch Fault and the Southern Upland Fault, a much thicker sequence accumulated, showing that tectonic movements were still affecting the Midland Valley at that time.

Fold and fault structures

The strata of the Irvine district were deformed into broad, low-amplitude folds during post-Westphalian times (Figure 29). The style of folding is slightly different on either side of the Dusk Water Fault. To the north, the folds have a wavelength of about 2 km and fold axes trend between north and north-east. Some small-scale folding can be seen on the foreshore at Ardrossan [NS 2318 4182] where a syncline/anticline couplet has fold axes trending at 050° and plunging at 28° to the south-west. These small-scale folds are associated with the line of the Dusk Water Fault and, as their existence implies compression, they may have been formed during a period of transcurrent or reversed movement on this fault. South of the Dusk Water Fault the fold structures are generally more open in style with a wavelength of the order of 10 km and axial planes orientated east to west. A syncline occurs just north of Irvine and an anticline occurs north of Barassie. The major fold structures are truncated by the major north-east-trending faults, so some periods of faulting clearly post-date the folding.

The dominant faults trend north-east: these are the Paisley Ruck, Dusk Water, Annick Water and Inchgotrick faults (Figure 29). Further east, in the Kilmarnock district, the trend of the Inchgotrick Fault becomes more east–west, however. Faults trending east–west occur in the southern part of the Irvine district, the most important being the zone of intense disturbance, known to miners as the 'Red Trouble', immediately to the south of Irvine. From the evidence discussed in the earlier part of the chapter, it appears that the northeast-trending faults have a long and complex history, with different amounts and direction of throw at different times, but that most of the movements took place in extensional regimes dominated by normal faulting. However, if any of the Carboniferous stress patterns predicted from regional studies are applied (e.g. east–west extension or north–south extension, see above), then an element of transcurrent movement must have occurred along the north-east-trending faults, with normal faulting on accommodation faults oblique to the main fault trends. For the wider region, sinistral movement has been suggested during the Devonian (Bluck, 1980) with the pattern changing in Carboniferous times to give dextral movement (Dewey, 1982; Read, 1988).

Main Late Devensian glaciation

Directions of ice flow

At least two directions of ice flow are recognised in the Irvine district (Figure 30). One is represented by strong north–south directions recorded in the crag-and-tail features on the coastal strip north of Ardrossan, and by north-east to south-west trends in the drumlin features of the Dalry–Kilbirnie area and the southern part of the district. The other is seen in the east–west trend of drumlins of the Irvine and Annick Water valleys. The geomorphological relationship between the two flow directions is preserved at Kilmaurs. Both trends can be identified as far south as central Ayrshire (Richey et al., 1930; Eyles et al., 1949). The flow directions in the Irvine district probably were deflected by the topography to only a minor extent; evidence from the adjacent Greenock district (Paterson et al., 1990) indicates that the ice was commonly constrained to flow up steep hillsides. This in turn suggests that ice thicknesses at the time of formation of these directional features was very great.

Evidence of the relative age of these ice-flow directions is limited, but Richey et al. (1930) considered that glacial features with north-east to south-west and north–south trends are later than features with an east–west trend. Tills with marine shells (shelly tills) are common in the Irvine valley and it was suggested that these resulted from the west-to-east movement of ice. It is likely that the features with an east–west trend, and other sets with an easterly component, were formed while a major ice sheet lay to the west of the Clyde coast, having formed perhaps formed by the coalescence of ice caps centred on the western Highlands, the Southern Uplands and Northern Ireland. This episode may relate to the pre-late-Devensian glaciation, and so to the lower till at Sourlie, or to an early phase of the late Devensian glaciation. The lineations with a north-east to south-west and north–south trend must have formed later, as Highland ice encroached farther into the Midland Valley, following decline of the more southerly ice centres. In the southern part of the district, the north-east to south-west trend is dominant and may have resulted from the coalescing of ice from the western Highlands and the Southern Uplands.

Glacial deposits

The typical deposit laid down directly by the ice-sheet is till, a compact diamicton composed of angular rock clasts of local and distant origin set in a matrix of sandy silty clay. On the higher ground the till cover is thin and patchy, but it reaches to the top of the highest hills. In the lower ground the till deposit is more continuous except in a narrow strip along the coast, where it has locally been removed by marine erosion.

The character of the till depends largely on the nature of the bedrock from which it is derived. The till derived from the Devonian strata that outcrop along the coastal strip from Largs to Ardrossan is brown or red-brown in colour, sandy in grain-size, and contains angular blocks of Upper Devonian sandstones and conglomerates. Clasts of Carboniferous lava and sandstone are also present. Till derived from mixed sequences of Carboniferous strata covers most of the Irvine district and is typically blue-grey with a clay matrix and a wide range of clast lithologies. Red sandy tills also occur to the south of the Inchgotrick Fault, where Devonian and Permian strata crop out.

Meltwater drainage channels

Channels cut by meltwaters flowing at the base of an ice sheet, or by water spilling from an ice-impounded lake, are present in the Irvine district. Early formation of such channels is apparent from the existence of buried drift-filled channels in the Irvine–Troon area (Figure 30). These are eroded to depths of about 25 m below OD and the irregular character of the valley-bottom topography would indicate that they were eroded in part by ice. Most of these channels contain till and probably owe their origins to earlier phases of glaciation when sea level was low. These early valley features have been the focus of subsequent erosion and most of the channels are filled with silt and sand, with till at the base.

Younger meltwater channels modify the surface till topography, frequently following topographical lows formed by the drumlin landforms. Some of these channels flow downhill obliquely and may have been cut by meltwater flowing beneath the ice. It has been suggested (Paterson et al., 1990) that the valley through Lochwinnoch and Kilbirnie acted as a spillway for water impounded in the upper Clyde valley to the north, but boreholes at Lochwinnoch have failed to detect any evidence of such an event.

Meltwater deposits

Deposits of sand and gravel laid down during the melting of the late Devensian ice sheet are of limited extent in the Irvine district. Moundy deposits of sand and gravel occur in the Hunterston area and inland south of Fairlie and at West Kilbride. These deposits were laid down beneath the ice and are often associated with crag and tail topography, so that the thickness of sand and gravel is limited to a veneer over a rock or till-cored ridge. Similar moundy deposits also occur in a swathe from Kilwinning to Crosshouse [NS 390 380]. These were probably deposited close to the margin of the retreating ice sheet. They have been extensively modified in extent and form by later meltwater and along their western margin they merge with late Devensian and Flandrian raised coastal deposits. Flat or terraced spreads of sand and gravel in the Irvine valley may be relics of a fluvioglacial ridge modified by erosion during the Flandrian episode of raised sea level. Comparable deposits on the south-east of the Hunterston peninsula, west of West Kilbride, may have been formed by similar processes. Small esker-like ridges of sand and gravel occur north of Kilbirnie, at Greenridge [NS 312 566] and at Kaimhill House [NS 330 574]. These would have been formed in a subglacial or englacial environment.

Examples of temporary lakes created by barriers of meltwater debris and wasting ice, and subsequently infilled with deposits of silt and clay, occur south of Crosshouse and east and north-east from Dundonald.

Late Devensian marine deposits

Changes of sea level relative to the land during the decay of an ice sheet are the result of the interaction between eustatic sea level changes, caused by the addition of water to the oceans as the ice sheet melts, and the isostatic recovery of the land from the depression caused by the former ice load. When the ice retreats from a glacially depressed area and the sea gains access, the sea level is initially high but falls rapidly because the rate of isostatic uplift, due to unloading, exceeds the rate of eustatic sea-level rise.

The history of late Devensian sea-level change is well documented in areas nearby to the north (Paterson et al., 1990) and north-east (Browne and McMillan, 1989) but in the Irvine district evidence relating to these movements is limited. Marine sediments relating to the maximum transgression of the late Devensian are poorly preserved, having been considerably modified by subsequent fluvial reworking and recent urban and industrial development.

Raised coastal deposits of late Devensian age occur at intervals along the coastal strip of the Irvine district. At Largs they reach a height of around 38 m above OD, decreasing to 30 m above OD at Ardrossan and 28 m above OD at Drybridge [NS 360 360]. These are the higher raised beaches, sands and gravels deposited in high-energy beach and other coastal environments during the retreat stages of the late Devensian glaciation. The deposit is usually thin and may rest on rock or on till. To the north of Kilwinning a bench feature [NS 313 443] is cut into till but no deposit is present. Within the general area of the Irvine and Garnock valleys, where the environment is more sheltered, the lithologies change laterally to finer-grained sand and silt deposited in what would then have been an estuary.

Further offshore, finer-grained sediments were deposited at lower topographical levels. Late Devensian marine deposits of this type in the Firth of Clyde area have long been referred to as the Clyde Beds (Table 18), a term which includes sediments laid down under high-boreal to low-arctic climatic conditions over a considerable time span. Robertson (1877b) described a section at Misk Pit [?291 407], near Kilwinning, which included about 6 in of 'muddy sand'. This contained a marine fauna comprising foraminifera, ostracods and molluscs. Robertson noted that Macoma calcarea was the predominant bivalve in what he showed to be a relatively diverse assemblage. Other exclusively cold-water molluscs such as Chlamys islandica and Nalica gvoenlandica were also identified and the fauna as a whole is reminiscent of the Linwood Formation (part of the Clyde Beds) of the Glasgow district. Robertson remarked on the abundance of the alga Melobesia polymorpha encrusting boulders which, he observed, was also common, if less abundant, in the Clyde Beds. He also regarded the positioning of balanids on the underside of stones in the clay as evidence that the fauna was to some extent derived and that the stones had been dropped from floating ice. More recent work by BGS (D K Graham, unpublished) has shown this to be characteristic of Linwood Formation deposits at other localities. A shallow, cold-water, marine assemblage from Misk Pit in John Smith's collection provides supporting evidence for Robertson's interpretation.

An assemblage from the Garnock Water was described by Robertson (1877a) as 'in a scam on the south bank about two hundred yards south-east of Kilwinning Ironworks, the bank being about twenty-two feet above the level of the sea'. This location [probably at about 3058 4234] is some distance to the west of the present river course, which has changed significantly since originally mapped. The fauna includes the cold-water bivalves Macoma calcarea and Nuculana pernula in significant numbers in a diverse assemblage including other cold-water species. Although this assemblage is similar to those of the Linwood Formation, no confirmatory radiocarbon dates have been obtained. Robertson remarked on the abundance with which the gastropod Velutina undata occurred and stated that the latter was rare in the Clyde Beds. However, the species has subsequently been found to be quite common in Linwood Formation assemblages.

Further evidence of the depositional. environment of this assemblage was given by Smith (1898, p.7) who cited a fauna of 44 molluscan species including the exclusively cold-water taxa Buccinum groenlandicum [= B. cyaneum], Chlamysislandica and Macoma calcarea.

Material in early BGS collections recovered from the Irvine Water, west of Craig's Pits Nos. 1 and 2 [?360 366] comprises a very small fauna with Macoma calcarea, Nuculana pernula and Yoldiella lenticula. This assemblage suggests a cold, but not necessarily arctic, environment and probably belongs to the Clyde Beds.

At Arthur's Street, Saltcoats [NS 2548 4124], a borehole sunk in 1948 to study Carboniferous strata penetrated a clay layer in the superficial deposits from which a cold water molluscan assemblage including Nuculana pernula, Yoldiella fraterna and Y. lenticulawas recovered. The preservation of these shells is good and some of the valves are still paired, suggesting that this deposit, which is probably similar in age to the above, is in situ.

Material collected between 10.60 and 12.20 m depth in the Hunterston Borrow Pit [NS 1939 5102] provided a diverse molluscan fauna in which the gastropod Onoba semicostata was abundant at 11.30 m. The only gastropods recovered from the upper part of the section were Littorina spp. but cold-water bivalves occurred in all samples, with Macoma calcarea and Thyasira gouldi especially common. Nuculana pernula and Yoldiella lenticula occurred throughout. Again the assemblages suggest that the strata belong to the Linwood Formation.

At Kilwinning, Smith (1898, p.11) identified Arctica islandica and the cold-water indicator Macoma calcarea in `muddy sand' overlying till deposits.

A cold-water marine fauna from Corsehill Clay Pit [NS 3688 3843] occurring in reddish brown silty clay is badly preserved, apparently due to leaching, and most of the microfossils have been largely dissolved away. The presence of significant numbers of Littorina littorea suggests a rocky littoral environment.

Loch Lomond Stadial

The readvance glaciers of the Loch Lomond Stadial did not reach the Irvine district. A periglacial environment is inferred, but no deposits are known.

Main Rock Platform

By far the most prominent topographical feature in the coastal area of the district is the broad wave-cut bench with a steep, commonly cliffed back feature with its foot at around 15 m above OD. The feature is present all along the coast, but is less well developed south of Ardrossan where the underlying rocks are softer and the drift cover is thicker, and where the landscape has been modified by industrial development. It is particularly well developed along the coast between West Kilbride and Ardrossan (cover photograph). The rock platform is usually covered by a thin veneer of beach sand and shingle of Flandrian age.

The age of the rock platform is problematical, but is thought likely to be significantly older than the Flandrian beach deposits that commonly rest on it. Sissons (1974) considered that wave and frost action during the Loch Lomond Stade were largely responsible for its formation. In the district to the north, till has been found on the rock platform at Wester Ardoch [NS 358 764] and in site investigation boreholes east of Gourock Bay [NS 254 775], indicating a pre-late-Devensian age for the rock platform at these localities (Browne and McMillan, 1984) and suggesting that inheritance of older coastal topography is likely to play a significant role throughout the region.

Flandrian

Apart from the formation of the raised beaches, the landscape of the Irvine district has been little modified during the 10 000 years since the start of the Flandrian Stage. The deposits laid down during this period include spreads of alluvium on the floodplains of streams and rivers, peat in lowland basins and as thin but extensive spreads on high ground, and raised beach sand and gravel. Quarries, rock cuttings, tunnels, and deposits of made ground bear witness to human activities.

Alluvium

Alluvium in the form of floodplain deposits and terraces is found along many of the streams and rivers, the most extensive being in the Garnock valley and in the valley of the River Irvine. Less extensive floodplains are associated with the courses of the Dusk Water, Lugton Water, Annick Water and Carmel Water. The alluvium is variable in character, consisting of gravel, sand and silt in varying proportions, with local peat. In many parts of the district isolated patches of alluvium occupying hollows mark the sites of former lakes. The deposits in such places are generally of silt and mud, commonly peaty.

Marine deposits

During the Flandrian there were major changes in the position of the shoreline because of the interplay between eustatic rise in sea level and the continuing isostatic recovery from glacial loading. From evidence outwith the district, sea level is known to have fallen during the early part of the Flandrian. In the Forth and Tay valleys for example, peat developed on abandoned marine flats as the sea fell to a level close to OD during the period from about 9000 to 8000 BP. The peat layer was buried as a result of a major transgression which reached its maximum level, at the Main Postglacial Shoreline, by about 6500 BP (Sissons and Brooks, 1971).

In the Irvine district evidence of peat formation prior to the Flandrian transgression is preserved in Dundonald Burn [NS 337 373] where a woody peat, 0.24 m thick, rests on grey pebbly till. The top of the peat is at 6.3 m above OD and dates of 9620 ± 150 and 9530 ± 130 BP have been obtained from wood (Godwin and Willis, 1962, Q-642). The Flandrian marine transgression eroded part of the compressed peat and deposited 7.0 m of sand and gravel. The peat is also known in outcrops and boreholes to the south near Girvan (Bishop and Coope, 1977).

Elsewhere the Flandrian transgression laid down thin deposits of sand with some fine gravel on the Main Rock Platform or upon its till cover and built up to the contemporaneous sea level, particularly in the many sheltered bay areas. The landward limit of these deposits is at about 17 m above OD. The coastal strip from Ardrossan to Portencross and from Fairlie to Largs provides a good example of raised beach deposits on a wave-cut rock platform with a pronounced cliff feature at about 15 m above OD (see cover photograph). South of Ardrossan the erosion has taken place into till and the back feature is much less pronounced. Around Irvine and south to Kilnford, near Dundonald, Flandrian shorelines are also poorly defined: they are cut into and merge with fluvioglacial deposits and late Devensian raised coastal deposits also at around 15 m above OD.

Flandrian faunas have been recovered from a number of localities. Material collected by John Smith from the Ardeer Pit at Stevenston [NS 2732 4030] comprises a diverse molluscan fauna. The dominant taxa are the gastropod Acteon tornatilis and the bivalve Corbula gibba. The assemblage as a whole indicates a temperate, postglacial marine environment with some estuarine influence. At Heather-house Pit, Irvine [NS 315 375], a temperate Flandrian marine fauna was recovered; this is unusual in being dominated by the gastropod Caecum glabrum and wood-boring bivalves such as Xylophaga dorsalis. Smith (1895a, pp.29–50) described an exposure of '25ft raised beach' in the River Irvine at Shewalton [NS 334 362]; material in the John Smith Collection (now housed at the Scottish office of the Geological Survey in Edinburgh) comprises a diverse marine assemblage of temperate aspect. A section drawn by Smith (1895a, plate II) described the area between Shewalton Moor and Ardeer Sandstone Quarry. The section shows a succession in which till deposits with an included layer of sand and gravel are overlain by 'sandy clay' with a cold water marine fauna. Overlying the till is a layer of sand and gravel, with postglacial shells in places but with the shells dissolved out in others. Glaciated boulders occur at its base. This layer of sand and gravel includes lenticular peat beds and where the Irvine Water had cut into it whale bones were recovered from the river bed, testifying to the estuarine nature of this location at that time.

Peat

During the early part of the Flandrian the climate became progressively warmer. Inland, many of the lakes which had formed in hollows and low-lying areas began to silt up with alluvium and peat. The extensive spread of alluvium from Lochwinnoch to Dairy would have originally accumulated in a shallow lake, of which the Kilbirnie and Barr lochs are the remnants. Any peat accumulation has probably been removed for fuel or to improve drainage and the agricultural use of the land. Large tracts of basin peat formed, as at Bloak Moss and Auchentiber Moss, east of Dairy. Peat mosses also developed in the coastal areas; the largest, Shewalton Moss, covers some 2.5 sq km and has a maximum recorded thickness of 9.0 m. Note also that some of the peat that formed in low lying coastal areas early in the Flandrian was buried during the Flandrian marine transgression (see above).

On the outcrops of volcanic rocks, in the north-west and north-east of the district, hill and basin peat are both developed. The basins are mostly rock basins, in which peat has accumulated along with some alluvium. Extensive tracts of relatively thin hill peat have accumu lated as part of the soil forming processes of a wet upland area.

Blown sand

Blown sand forms a belt of dunes which runs parallel to the coast from Stevenston to Troon, the most continuous and prominent feature being the fore-dune ridge. The dunes are up to 20 m high in the Ardeer area but reduce to around 8 m at Troon. Comparison with the Ordnance Survey map of 1856 shows that in the Misk Knowes area the dunes have migrated eastwards by nearly 300 in. The dunes are now entirely stabilised, either bvegetation or urban development, but disturbance of the surface 'aild removal of vegetation could allow the movement to start again. The movement of blown sand eastwards has influenced the course of the River Garnock above its confluence with the River Irvine, the sand dune complex forming a barrier across the shortest route to the sea. Elsewhere, minor areas of blown sand are associated with the sandy embayments between Saltcoats and Hunterston.

Made ground

Made ground marks the site of the former steelworks at Glengarnock and that of the explosives and chemicals factory at Ardeer. Reclamation of the intertidal zone has produced areas of made ground at Ardrossan and Hunterston. A further account is given in Chapter 2.

Information sources

Further geological information held by the British Geological Survey relevant to the Irvine district is listed below. It includes published material in the form of maps, memoirs and reports and unpublished maps and reports. Also included are other sources of data held by BGS in a number of collections, including borehole records, mine plans, fossils, rock samples, thin sections, hydrogeological data and photographs.

Searches of indexes to some of the collections can be made on the Geoscience Index System in BGS libraries. This is a developing computer-based system which carries out searches of indexes to collections and digital databases for specified geographical areas. It is based on a geographical information system linked to a relational database management sytem. Results of the searches are displayed on maps on the screen. At the present time (1997) the datasets are limited and not all are complete. The indexes which are available are listed below:

• Index of boreholes
• Topographical backdrop based on 1:250 000 scale maps
• Outlines of BGS maps at 1:50 000 and 1:10 000 scale and 1:10 560 scale County Series
• Chronostratigraphical boundaries and areas from BGS 1:250 000 maps
• Geochemical sample locations on land
• Aeromagnetic and gravity data recording stations
• Land survey records

Details of geological information available from the British Geological Survey can be accessed on the BGS Web Home Page at: http://www.bgs.ac.uk

BGS maps

• Geology maps
• 1:625 000
• United Kingdom (North Sheet) Solid geology, 1979;
• United Kingdom (North Sheet) Quaternary geology, 1977
• 1:250 000
• Clyde (Sheet 55N 06W) Solid geology, 1985
• 1: 63 360
• Sheet 22 (Kilmarnock) Solid & Drift 1928
• Sheet  23 (Hamilton) Solid & Drift 1929
• Sheet 29 (Rothesay) Solid & Drift 1971
• 1:50 000
• Sheet 14W (Ayr) Solid, 1978; Drift, 1978
• Sheet 14E (Cumnock) Solid, 1976; Drift, 1980
• Sheet 22W (Irvine) Solid, in press; Drift, 1987
• Sheet 22E (Kilmarnock) Solid, in press; Solid & Drift, in press
• Sheet 30W (Greenock) Solid, 1990; Drift, 1989
• Sheet 30E (Glasgow) Solid, 1993; Drift, 1993
• Special Sheet (Arran) Solid, 1987; Drift, 1985
• The Irvine district is the western half of the area which was formerly covered by Sheet 22 (Kilmarnock) of the Geological Map of Scotland.
• 1:10 000 and 1:10 560
• Details of the original geological surveys are listed on editions of the 1:63 360 and 1:50 000 geological sheets. Copies of the fair-drawn maps of these earlier surveys may be consulted at the BGS Library, Edinburgh.
• The maps at six-inch or 1:10 000 scale cohering wholly or in part in the 1:50 000 scale Sheet 15W are listed below, tofether with the surveyors' initials and the dates of the survey. The surveyors were: M Armstrong, I B Cameron, A Davies, J Dawson,
• V A Eyles, S K Monro, W Mykura and D Stephenson. Separate sheets cover Solid and Drift geology. The maps are not published but are available for consultation in the BGS Library, Edinburgh, and also at Keyworth and the London Information Office, in the Natural History Museum, South Kensington, London. Dye-line copies may he purchased from the Sales Desk.
• Geophysical maps
• 1:625 000
• United Kingdom (North Sheet) Aeromagnetic anomaly, 1972;
• United Kingdom (North Sheet) Bouguer anomaly, 1981;
• United Kingdom (North Sheet) Regional gravity, 1981
• 1:250 000
• Clyde (Sheet 55N 06W) Aeromagnetic anomaly, 1980;
• Clyde (Sheet 55N 06W) Bouguer gravity anomaly, 1985

• 1:50 000
• Geophysical information maps; these are plot-on-demand maps which summarise graphically the publicly available geophysical information held for the sheet in the BGS databases. Features include:

◦      Regional gravity data: Bouguer anomaly contours and location of observations.

◦      Regional aeromagnetic data: total field anomaly contours and location of digitised data points along flight lines.

◦      Gravity and magnetic fields plotted on the same base map at 1:50 000 scale to show correlation between anomalies.

◦      Separate colour contour plots of gravity and magnetic fields at 1:125 000 scale for easy visualisation of important anomalies.

◦      Location of local geophysical surveys.

◦      Location of public domain seismic reflection and refraction surveys.

◦      Location of deep boreholes and those with geophysical logs. Geochemical atlases

• 1:250 000
• Point-source geochemical data processed to generate a smooth continuous surface presented as an atlas of small-scale colour-classified digital maps.
• Southern Scotland and part of Northern England, 1993.
• Geochemical Survey Programme data are also available in other forms including hard copy and digital data.
• Hydrogeological maps
• 1:625 000
• Sheet 18 (Scotland) 1988.
• Groundwater vulnerability (Scotland) 1995.

BGS books

Memoirs, reports and papers relevant to the Irvine district arranged by topic. Most are either out of print or are not widely available, but may be consulted at BGS and other libraries.

• General geology
• British regional geology
• The Midland Valley of Scotland, 3rd Edition, 1985 Memoirs
• The geology of Central Ayrshire (Sheet 14) 2nd edition 1949, reprinted 1985
• The geology of North, Arran, South and the Cumbraes (Sheet 21) 1903
• The geology of North Ayrshire (Sheet 22) 2nd edition 1930 Geology of the Irvine district (Sheet 22W) (this book) Geology of the Hamilton district (Sheet 23W) 1998 Geology of the Greenock district (Sheet 30W) 1990 Geology of the Glasgow district (Sheet 30E) 1998
• Reports and Bulletins
• Lithological variation in the Lower Limestone Group of the Midland Valley of Scotland. Bulletin No. 12, pp.52–65. 1957.
• IGS Boreholes 1976 Report, No. 77/10. IGS Boreholes 1977 Report, No. 78/21.
• Late-Glacial and post-Glacial marine environments at Ardyne, Scotland, and their significance in the interpretation of the Clyde sea area. Report, No. 78/17.
• The lower part of the Limestone Coal Group in the Glasgow district. Report, No. 78/29.
• IGS Boreholes 1978 Report, No. 79/12. IGS Boreholes 1979 Report, No. 80/11. IGS Boreholes 1980 Report, No. 81/11.
• Lithostratigraphy of the late Devonian and early Carboniferous rocks in the Midland valley of Scotland. Report, *01. 18, No.3, 1986.
• A lithostratigraphic framework for the Carboniferous rocks of the Midland Valley of Scotland. Report, WA/96/29/R
• The geology of the Clyde Plateau volcanic Formation of the Kilmarnock district (Sheet 22E), Central Scotland. Report, WA/97/88.
• A review of the lithostratigraphic framework for the Devonian rocks of Scotland south of the Great Glen Fault. Report in press.
• Economic geology: coal, ironstone and limestone
• Memoirs
• The economic geology of the Ayrshire coalfield, Area I, Kilbirnie, Dairy and Kilmaurs. 1925.
• The economic geology of the Ayrshire coalfield, Area II, Kilmarnock Basin. 1925.
• The economic geology of the Ayrshire coalfields, Area III, Ayr, Preswick, Mauchline, Cumnock and Muirkirk. 1930.
• The economic geology of the Ayrshire coalfields, Area TV, Dailly, Patna, Rankinston, Dalmellington and New Cumnock. 1932.
• The Ayrshire bauxitic clay. 1922. The iron ores of Scotland. 1920. The limestones of Scotland. 1949.
• Biostratigraphy
• The fauna and distribution of the Queenslie Marine Band (Westphalian) in Scotland. IGS Report No. 77/18.
• There is also a collection of internal biostratigraphical reports.
• Bulk minerals
• Sand and gravel resources of the Strathclyde Region of Scotland. Report, No. 77/8.
• Central Scotland mineral portfolio: resources of clay and mudstone for brickmaking. Open-file Report. 1985.
• Central Scotland mineral portfolio: limestone resources. Open-file Report. 1986.
• Central Scotland mineral portfolio: special sand resources. Open-file Report. 1986.
• Central Scotland mineral portfolio: fireclay resources. Open-file Report. 1985.
• Central Scotland Mineral portfolio: hard rock aggregate resources. Open-file Report. 1985.
• Metalliferous minerals
• Barytes in Central Scotland. Wartime Pamphlet No38.
• Baryte and copper mineralisation in the Renfrewshire Hills, Central Scotland. MRP Report, No.67. 1983.
• Engineering geology
• Hydraulic properties of the Upper Old Red Sandstone at Hunterston Peninsula, Ayrshire. Report, WD/ST/74/11.
• Hydrogeology
• Records of wells in the areas of Scottish one-inch Geological Sheets 21–23. Water Supply Papers. 1972.
• Hydrogeology of Scotland. 1990. Geophysics
• An appraisal of geophysical data in the Midland Valley of Scotland. Report, No. RC; 87/12.
• North Britain, Geophysics CD Series. 1997. Geothermal potential
• Interpretation of a deep seismic reflection profile in the Glasgow area. Investigation of the geothermal potential of the UK. 1984.
• The Upper Palaeozoic basins of the Midland Valley of Scotland. Investigation of the geothermal potential of the UK. 1985.
• Offshore geology
• The solid geology of the Clyde Sheet (55N 06W) Report, No. 78/9.

Documentary collections

• Borehole record collection
• BGS holds collections of records of boreholes which can be consulted at BGS, Edinburgh, where copies of most records may be purchased. For the Irvine district the collection (1997) consists of the sites and logs of about 2830 boreholes. Index information, which includes site references, for these bores has been digitised. The logs are either hand-written or typed and many of the older records are drillers logs.
• Boreholes mentioned in the text are listed here in alphabetical order. A few lie in the adjacent districts.

• Site exploration reports
• This collection consists of site exploration reports carried out to investigate foundation conditions prior to construction. There is a digital index and the reports themselves are held on microfiche. For the Irvine district there are presently (1998) about 940 reports with over 10 000 borehole logs.
• Mine plans
• BGS maintains a collection of plans of underground mines for minerals other than coal and oilshale. A large number of plans held fall within the district.
• Hydrogeological data
• Records of water boreholes are held at BGS, Edinburgh. Geochemical data
• Records of stream-sediment and other analyses are held at BGS, Edinburgh.
• Gravity and magnetic data
• Records are held at BGS, Edinburgh.
• Seismic data
• Records of earthquakes are held at BGS, Edinburgh. BGS Lexicon of named rock unit defmitions
• Definitions of the named rock units shown on BGS maps, including those shown on the 1:50 000 Series Sheet 22W Irvine are held in the Lexicon database. This is available on Web Site http://www.bgs.ac.uk. Further information on the database can be obtained from the Lexicon Manager at BGS Keyworth.

Material collections

• Geological Survey photographs
• Photographs illustrating aspects of the geology of the Irvine district are deposited for reference in the libraries at BGS, Edinburgh, and BGS, Keyworth; and in the BGS Information Office, London. The photographs depict details of the various rocks and sediments exposed either naturally or in excavations and also some general views. The photographs can be supplied as black and white or colour prints and 2 X 2 colour transparencies, at a fixed tariff, from the Photographic Department, BGS, Edinburgh. Titles are listed below.

• Petrological collections
• The petrological collections for the Irvine district consist of several hundred hand specimens and thin sections. Most samples and thin sections are of the igneous rocks in the district. The sedimentary rocks are poorly represented. Information on databases of rock samples, thin sections and geochemical analyses can be obtained from the Group Manager, Mineralogy and Petrology Section, BGS, Edinburgh.
• Bore core collection
• Samples have been collected from core taken from boreholes. At present (1998) there are 2660 samples (hand specimens) from 64 bores which are registered in the borehole collection.
• Palaeontological collections
• The collections of biostratigraphical specimens are taken from surface and temporary exposures, and from boreholes throughout the Irvine district. The collections are working collections and are used for reference. They are not at present on a computer database. Databases of fossil samples from the Irvine district can be obtained from the Curator, Palaeontology Section, BGS, Edinburgh.

Collections held outwith BGS

• Coal abandonment plans are held by The Coal Authority, Mining Records Department, Bretby Business Park, Ashby Road, Burton-on-Trent, Staffs, DE15 OQD.
• Sites of Special Scientific Interest are the responsibility of Scottish Natural Heritage, Battleby, Redgorton, Perth, PH1 3EW.

References

Most of the references listed below are held in the Library of the British Geological Survey at Keyworth, Nottingham. Copies of the references can be purchased subject to current copyright legislation.

ALEXANDER, R W S, DAWSON, J B, PATTERSON, E M, and HERVIG, R L. 1986. The megacryst and inclusion assemblage from the Black Rock vent, Ayrshire. Scottish Journal of Geology, Vol. 22, 203–212.

ANDERSON, E M. 1925. The economic geology of the Ayrshire coalfields, Area II, Kilmarnock Basin including Stevenston, Kilwinning and Irvine. Memoir of the Geological Survey, Scotland.

ANDREWS,J E, TURNER, M S, NABI, G, and SPIRO, B. 1991. The anatomy of an early Dinantian terraced floodplain: palaeoenvironment and early diagenesis. Sedimentology, Vol. 38, 271–287.

ARMSTRONG, D. 1957. Dating of some minor intrusions of Ayrshire. Nature, London, Vol. 180, 1277.

ARMSTRONG, M, PATERSON, I B, and BROWNE, M A E. 1985. Geology of the Perth and Dundee district. Memoir of the British Geological Survey, Sheets 48W, 48E and 49 (Scotland).

ASSUMPCAO, M, and BAMFORD, D. 1978. LISPB V. Studies of crustal shear waves. Geophysical Jounial of the Royal Astronomical Society, Vol. 54, 61–73.

BAILEY, E B, CLOUGH, C T, WRIGHT, W B, RITCHEY, J B, and WILSON, G V. 1924. Tertiary and post-Tertiary geology of Mull, Loch Aline and Oban. Memoir of the Geological Survey, Scotland.

BALD, R. 1822. Notices regarding the fossil elephant of Scotland. Memoirs of the Wernerian Natural History Society, Vol. 4, 58–66.

BAMFORD, D. 1979. Seismic constraints on the deep geology of the Caledonides of northern Britain. 93–96 in The Caledonides of the British Isles - reviewed. HARRIS, A C, HOLLAND, C H, and LEAKE, B E (editors). Special Publication of the Geological Society of London, No. 8.

BAMFORD, D, NUNN, K, PRODEHL, C, and JACOB, B. 1978. LISPB-IV. Crustal structure of Northern Britain. Geophysical Journal of the Royal Astronomical Society, Vol. 54, 43–60.

BELL, D. 1885. On the geology of Ardrossan and West Kilbride. Transactions of the Geological Society of Glasgow, Vol. 7, 342–353.

BELT, E S, FRESHNEY, E C, and READ, W A. 1967. Sedimentology of Carboniferous cementstone facies. Journal of Geology, Vol. 75, 711–721.

BENNIE, J. 1885. Note on the content of two bits of clay from the elephant bed at Kilmaurs in 1817. Proceedings of the Royal Physical Society of Edinburgh, Vol. 8, 451–459.

BENNIE, J.1891. On things new and old from the ancient lake of Cowdenglen, Renfrewshire. Transactions of the Geological Society of Glasgow, Vol. 9, 213–215.

BRISTOW, W W, and COOPE, G R. 1977. Stratigraphical and faunal evidence for Lateglacial and early Flandrian environments in south-west Scotland. 61–68 in Studies in the Scottish Lateglacial Environment. GRAY, J M, AND LOWE, J J (editors). (Oxford: Pergamon Press.)

BLUCK, B J. 1967. Deposition of some Upper Old Red Sandstone conglomerates in the Clyde area: a study in the significance of bedding. Scottish Journal of Geology, Vol. 3, 139–167.

BLUCK, B J. 1978. Sedimentation in a late orogenic basin: the. Old Red Sandstone of the Midland Valley of Scotland. 249–278 in Crustal evolution in northwestern Britain and adjacent regions. BOWES, D R, and LEABB, B E (editors). Special Issue of the Geological Journal, No. 10.

BLUCK, B J. 1980. Evolution of a strike-slip fault controlled basin, Upper Old Red Sandstone, Scotland. 63–78 in Sedimentation in oblique-slip mobile zones. READING, H G, and BALLANCE, P F (editors). Special Publication of International Association of Sedimentologists, No. 4.

BLUCK, B J. 1984. Pre-Carboniferous history of the Midland Valley of Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 75, 275–295.

BOSAZZA, V L. 1947. The petrography and petrology of South African clays. (Johannesburg: V L Bosazza.)

BOSWELL, P G H. 1918. A memoir of British resources of sands and rocks used in glass-making, with notes on certain crushed rocks and refractory materials. (London: Longmans, Green and Co.)

BOTT, M H P, and MASSON-SMITH, D. 1960. A gravity survey of the Criffel granodiorite and the New Red Sandstone deposits near Dumfries. Proceedings of the Yorkshire Geological Society, Vol. 32, 317–332.

BRAND, P J. 1983. Stratigraphical palaeontology of the Westphalian of the Ayrshire Coalfield. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 73, 173–190.

BRAND, P J. 1991. Anthraconaia it vinensis sp. nov., a new Westphalian B bivalve with limited geographical and stratigraphical range. Scottish Journal of Geology, Vol. 27, 27–31.

BRERETON, R, and 6 others. 1988. Glenrothes borehole: geological well completion report. Investigation of the geothermal potential of the UK. (Keyworth, Nottingham: British Geological Survey.)

BROWN, S S. 1975. Petrogenesis and depositional environment of the Dockra Limestone (Upper Visean), North Ayrshire. Unpublished PhD thesis, University of Glasgow.

BROWNE, M A E, and 5 others. 1996. A review of the lithostratigraphy of the Carboniferous rocks of the Midland Valley of Scotland. British Geological Survey Technical Report, WA/96/29/R.

BROWNE, M A E, and MCMILLAN, A A. 1984. Shoreline inheritance and coastal history in the Firth of Clyde. Scottish Journal of Geology, Vol. 20, 119–120.

BROWNE, M A E, and, MCMILI.AN, A A. 1989. Quaternary geology of the Clyde Valley. British Geological Survey Research Report, SA/89/1.

BROWNE, M A E, and MONRO, S K. 1989. The evolution of the coal basins of central Scotland. Compte Rendu Ile Congres International de Stratigraphie et de Geologic du Carbonifere, Beijing, 1987, Vol. 5, 1–19.

BRYCE, J. 1865. On the occurrence of beds in the West of Scotland beneath the Boulder Clay. Quarterly Journal of the Geological Society of London, Vol. 21, 213.

BURGESS, I C. 1965. Calcifolium (Codiaceae) from the Upper Visean of Scotland. Palaeontology, Vol. 8, 192–198.

CALDWELL, W G. 1973. The Cumbrae Islands. 141–156 in Excursion guide to the geology of the Glasgow district. BLUCK, B (editor). (Glasgow: Geological Society of Glasgow.)

CAMERON, I B. 1980. Titanium dioxide in the Ayrshire Bauxitic Clay. Mineral Reconnaissance Report, Institute of Geological Sciences, No. 52, 14–18.

CAMERON, I B, and LAWSON, R I. 1979. An occurrence of paratacamite in south-west Scotland. Mineralogical Magazine, Vol. 43, 547.

CAMERON, I B, and STEPHENSON, D. 1985. British regional geology: the Midland Valley of Scotland. (3rd edition). (London: HMSO for British Geological Survey.)

CARRUTHERS, R G, and RICHEY, J E. 1915. The Lower Limestones of Renfrewshire and north Ayrshire: a comparison and correlation. Transactions of the Geological Society of Glasgow, Vol. 15, 249–264.

CHISHOLM, J I, and DEAN, J M. 1974. The Upper Old Red Sandstone of Fife and Kinross: a fluviatile sequence with evidence of marine incursion. Scottish Journal of Geology, Vol. 10, 1–30.

CONWAY, A, DENTITH, M C, DOODYJ J, and HALL, J. 1987. Preliminary interpretation of upper crustal structure across the Midland Valley of Scotland from two east-west seismic refraction profiles. Journal of the Geological Society, London, Vol. 144, 867–870.

COTTON, W R. 1968. A geophysical survey of the Campsie and Kilpatrick Hills. Unpublished PhD thesis, University of Glasgow.

CRAIG, J. 1841. On the Coal Formation of the West of Scotland. 89–92 in Report of the British Association for 1840 (Glasgow). (London: John Murray.)

CRAIG, P M, and HALL, I H S. 1975. The Lower Carboniferous rocks of the Campsie-Kilpatrick area. Scottish Journal of Geology, Vol. 11, 171–174.

CRAIG, R. 1868. Discovery of Bos primigenius in the Lower Boulder Clay of Scotland. Geological Magazine, Vol. 5, 486–487.

CRAIG, R. 1869. Sketch of the Carboniferous basin of Dairy, Ayrshire. Transactions of the Geological Society of Glasgow, Vol. 3, 271–297.

CRAIG, R. 1875. On the first appearance of certain fossils in the Carboniferous strata around Beith and Dairy. Transactions of the Geological Society of Glasgow, Vol. 5, 36–50.

CRAIG, R. 1883. On the fossiliferous strata lying between the Lower and Upper Limestones, in the Beith and Dairy districts. Transactions of the Geological Society of Glasgow, Vol. 7, 86–96.

CRAIG, R. 1885. Volcanic disturbance of the Ironstone Measures in the vicinity of Dairy during the Carboniferous period. Transactions of the Geological Society of Glasgow, Vol. 7, 233–237.

CRAIG, R. 1886. On the Upper Limestones of North Ayrshire as found in the district around Dahl, and elsewhere. Transactions of the Geological Society of Glasgow, Vol. 8, 28–39.

CRAIG, R. 1891. Notes upon a cutting in the new Kilbirnie branch of the Lanarkshire and Ayrshire Railway, on the farm of Gurdy, Beith. Transactions of the Geological Society of Glasgow, Vol. 9, 64–71.

CRAMPIN, S, JACOB, A W B, MILLER, A, and NEILSON, A. 1970. The LOWNET radio-linked seismometer network in Scotland. Geophysical Journal of the Royal Astronomical Society, Vol. 21, 207–216.

DAVIDSON, K A S, SOLA, M A, POWELL, D W, and HAIL, J. 1984. Geophysical model for the Midland Valley of Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 75, 175–181.

DE LAPPARENT J 1936. Boehmite and diaspore in the bauxitic clays of Ayrshire. 1–7 in Summary of Progress of the Geological Survey of Great Britain for 1934. Part 2. (London: His Majesty's Stationery Office.)

DE SOUZA, H A F. 1979. The geochronology of Scottish Carboniferous volcanism. Unpublished PhD thesis, University of Edinburgh.

DE SOUZA, H A F. 1982. Age data from Scotland and the Carboniferous time scale. 456–465 in Numerical dating in stratigraphy. ODIN, G S (editor). (Chichester: John Wiley and Sons Ltd.)

DEAN, M T. 1987. Carboniferous conodonts from the Lower and Upper Limestone groups of the Scottish Midland Valley. Unpublished MPhil thesis, University of Nottingham.

DENTITH, C, and HALL, J. 1989. MAVIS—an upper crustal seismic refraction experiment in the Midland Valley of Scotland. Geophysical Journal International, vol 99, 627–643.

DEWEY, J F. 1969. Evolution of the Appalachian/Caledonian orogen. Nature, London, Vol. 222, 124–129.

DEWEY, J F. 1971. A model for the Lower Palaeozoic evolution of the southern margin of the early Caledonides of Scotland and Ireland. Scottish Journal of Geology, Vol. 7, 220–240.

DEWEY, J F. 1982. Plate tectonics and the evolution of the British Isles. Journal of the Geological Society of London, Vol. 139, 371–412.

DOWNIE, C, and LISTER, T R. 1969. The Sandy's Creek beds (Devonian) of Farland Head, Ayrshire. Scottish Journal of Geology, Vol. 5, 193–206.

EYLES, V A, SIMPSON, J B, and MACGREGOR, A G. 1930. The economic geology of the Ayrshire coalfields, Area III, Ayr, Prestwick, Mauchline, Cumnock and Muirkirk. Memoir of the Geological Survey, Scotland.

EYLES, V A, SIMPSON, J B, and MACGREGOR, A G. 1931. The igneous geology of Central Ayrshire. Transactions of the Geological Society of Glasgow, Vol. 18, 361–387.

EYLES, V A, SIMPSON, J B, and MACGREGOR, A G. 1949 (reprinted 1980). Geology of Central Ayrshire. (2nd edition). Memoir of the Geological Survey, Sheet 14 (Scotland).

FALCONER, J D. 1907. The geology of Ardrossan. Transactions of the Royal Society of Edinburgh, Vol. 45, 601–610.

FITCH, F J, MILLER, J A, and WILLIAMS, S C. 1970. Isotopic ages of British Carboniferous rocks. Compte Rendu 6e Congres International de Stratigraphie et de Geologie du Carbonifere, Sheffield 1967, Vol. 2, 771–790.

FITTON, J G, and HUGHES, D J. 1970. Volcanism and plate tectonics in the British Ordovician. Earth and Planetary Science Letters, Vol. 8, 223–228.

FORSYTH, I H. 1978. The lower part of the Limestone Coal Group in the Glasgow district. Report of the Institute of Geological Sciences, No. 78/29.

FORSYTH, I H, and BRAND, P J. 1986. Stratigraphy and stratigraphical palaeontology of Westphalian B and C in the Central Coalfield of Scotland. Report of the British Geological Survey, Vol. 18, No. 4.

FORSYTH, I H, HALL, I H S, and McMILLAN, A A. 1996. Geology of the Airdrie district. Memoir of the British Geological Survey, Sheet 31W (Scotland).

FRANCIS, E H. 1979. Igneous activity in a fractured craton: Carboniferous volcanism in northern Britain. 279–296 in Crustal evolution in north western Britain and adjacent regions. BOWES, D R, and LEAKE, B E (editors). Geological Journal Special Issue, No. 10.

FRANCIS, E H, FORSYTH, I H, READ, W A, and ARMSTRONG, M. 1970. The geology of the Stirling district. Memoir of the Geological Survey of Great Britain, Sheet 39 (Scotland).

FRANCIS, E H, and HORGOOD, A M. 1970. Volcanism and the Ardross Fault, Fife, Scotland. Scottish Journal of Geology, Vol. 6, 162–185.

FRIEND, P F, HARLAND, W D, and HUDSON, J D. 1963. The Old Red Sandstone and the Highland Boundary in Arran, Scotland. Transactions of the Edinburgh Geological Society, Vol. 19, 363–425.

GALLAGHER, M J 1968. Contribution to discussion of Scott, B. Barytes mineralisation at Gass Water mine, Ayrshire, Scotland. Transactions of the Institution of Mining and Metallurgy (Section 13: Applied earth science), Vol. 77, B44–45.

GEIKIE, A, GEIKIE, J, JACK, R L, and ETHERIDGE, J, junior. 1872. Explanation of Sheet 22: Ayrshire (north part) with parts of Renfrewshire and Lanarkshire. Memoir of the Geological Survey, Scotland.

GEIKIE, J. 1868. Note on the discovery of Bos primigenius in the lower Boulder Clay of Scotland. Geological Magazine, Vol. 5, 393–395.

GEORGE, T N. 1960. The stratigraphic evolution of the Midland Valley. Transactions of the Geological Society of Glasgow, Vol. 24, 32–107.

GEORGE, T N, and 6 others. 1976. A correlation of Dinantian rocks in the British Isles. Geological Society of London Special Report, No. 7.

GODWIN, H, and WILLIS, E H. 1962. Cambridge University Natural Radiocarbon measurements V. Radiocarbon, Vol. 4, 57–70.

GOODLET, G A. 1957. Lithological variation in the Lower Limestone Group of the Midland Valley of Scotland. Bulletin of the Geological Survey of the Great Britain, No. 12, 52–65.

GRAHAM, A M, and UPTON, B G J 1978. Gneisses in diatremes, Scottish Midland Valley: petrology and tectonic implications. Journal of the Geological Society of London, Vol. 135, 219–228.

GREGORY, T W, and CURRIE, E D. 1928. The vertebrate fossils from the Glacial and associated Post-Glacial Beds of Scotland in the Hunterian Museum, University of Glasgow. Monograph of the Geology Department, Hunterian Museum Glasgow University, Vol. 2, 1–21.

GRIBBLE, C D. 1992. Loanhead quarry. 235–239 in Geological excursions around Glasgow and Girvan. LAWSON, J D, and WEEDON, D S (editors). (Glasgow: Geological Society of Glasgow.)

GUNN, W, GEIKIE, A, and PEACH, B N. 1903. The geology of North Arran, South Bute and the Cumbraes with parts of Ayrshire and Kintyre: explanation of Sheet 21. Memoir of the Geological Survey, Scotland.

HALL, I H S, and CHISHOLM, J I. 1987. Aeolian sediments in the late Devonian of the Scottish Midland Valley. Scottish Journal of Geology, Vol. 23, 203–208.

HALL, I H S, FORSYTH, I H, and BROWNE, M A E. 1998. Geology of the Glasgow district. Memoir of the British Geological Survey, Sheet 30E (Scotland).

HALL, J. 1970. The correlation of seismic velocities with formations in the south west of Scotland. Geophysical Prospecting, Vol. 18, 134–138.

HALL, J. 1971. A preliminary seismic survey adjacent to the Rashiehill borehole near Slamannan, Stirlingshire. Scottish Journal of Geology, Vol. 7, 170–174.

HALL, J. 1974. A seismic reflection survey of the Clyde Plateau Lavas in north Ayrshire and Renfrewshire. Scottish Journal of Geology, Vol. 9, 253–279.

HALL, J, and 5 others. 1983. Seismological evidence for shallow crystalline basement in the Southern Uplands of Scotland. Nature, London, Vol. 305, 418–420.

HALLIDAY, A N, and 6 others. 1993. Formation and composition of the lower continental crust: evidence from Scottish xenolith suites. Journal of Geophysical Research, Vol. 98, 581–607.

HAZELDINE, R S. 1988. Crustal lineaments in the British Isles: their relationship to Carboniferous basins. 53–68 in Sedimentation in a synorogenic basin complex: Ilk, Upper Carboniferous of northwest Europe. BESLY, B M, and KELLING, G (editors). (Glasgow: Blackie and Son.)

HENDERSON, C M B, FOLAND, K A, and GIBB, F G F. 1987. The age of the Lugar sill and a discussion of the Late-Carboniferous/Early Permian sill complex of SW Scotland. Geological Magazine, Vol. 22, 43–52.

HIGHLEY, D E (compiler). 1982. Fireclay. Mineral Dossier Mineral Resources Consultative Committee, No. 24.

HOLM ES, A, and HARWOOD, H F. 1929. The tholeiite dykes of the north of England. Mineralogical Magazine, Vol. 22, 1–52.

HORNUNG, G, AL-ANI A, and STEWART, R M. 1966. The composition and emplacement of the Cleveland dyke. Transactions of the Leeds Geological Association, Vol. 7, 232–249.

HUNTER, R H, and UPTON, B G J 1987. The British Isles - a Palaeozoic mantle. 107–118 in Mantle xenoliths. NIXON, P (editor). (New York: John Wiley.)

INSTITUTE OF GEOLOGICAL SCIENCES. 1967. Aeromagnetic map of Great Britain, Sheet 1. (Chessington, Surrey: Ordnance Survey for Institute of Geological Sciences.)

JARDINE, W G, and 5 others. 1988. A late Middle Devensian interstadial site at Sourlie, near Irvine, Strathclyde. Scottish Journal of Geology, Vol. 24, 288–295.

JOHNSTONE, G S. 1965. The volcanic rocks of the Misty Law-Knockside Hills district, Renfrewshire. Bulletin of the Geological Survey of Great Britain, No. 22, 53–64.

KELLER, W D. 1967. Flint clay and flint-clay facies. Clays and Clay Minerals, Vol. 16, 113–128.

KENNEDY, W Q. 1958. Tectonic evolution of the Midland Valley of Scotland. Transactions of the Geological Society of Glasgow, Vol. 23, 107–133.

LAWSON, J D, and WEEDON, D S (editors). 1992. Geological excursions around Glasgow and Girvan. (Glasgow: Geological Society of Glasgow.)

LE MAITRE, R W (editor). 1989. A classification of igneous rocks and glossary of terms. Recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks. (Oxford, London, Edinburgh: Blackwell Scientific Publications.)

LEEDER, M R. 1971. Initiation of the Northumberland Basin. Geological Magazine, Vol. 108, 511–516.

LEEDER, M R. 1974. Origin of the Northumberland Basin. Scottish Journal of Geology, Vol. 10, 283–296.

LEEDER, M R. 1976. Sedimentary facies and the origins of basin subsidence along the northern margin of the supposed Hercynian Ocean. Tectonophysics, Vol. 36, 167–179.

LEEDER, M R. 1982. Upper Palaeozoic basins of the British Isles-Caledonian inheritance versus Hercynian plate marginal processes. Journal of the Geological Society of London, Vol. 139, 479–491.

LEEDER, M R, and MCMAHON, A H. 1988. Upper Carboniferous (Silesian) basin subsidence in northern Britain. 43–52 in Sedimentation in a synorogenic basin complex: the Upper Carboniferous of northwest Europe. BESLY, B M, and KELLING, G (editors). (Glasgow and London: Blackie and Son.)

LOUGHNAN, F C. 1978. Flint clays, tonsteins and kaolinite clayrock facies. Clay Minerals, Vol. 13, 387–400.

LUMSDEN, G I, and WILSON, R B. 1979. Stratigraphical classification of the Carboniferous succession of central Scotland. Compte Rendu 8e Congres International de Stratigraphie et de Geologie du Carbonifere, Moscow 1975, Vol. 2, 27–36.

MACDONALD, R. 1975. Petrochemistry of the early Carboniferous (Dinantian) lavas of Scotland. Scottish Journal of Geology, Vol. 11, 269–314.

MACDONALD, R. 1980. Trace element evidence for mantle heterogeneity beneath the Scottish Midland Valley in the Carboniferous and Permian. Philosophical Transactions of the Royal Society of London, Vol. 280, 111–123.

MACDONALD, R, GOTTFRIED, D, FARRINGTON, M J, BROWN, F W, and SKINNER, N G. 1981. The geochemistry of a continental tholeiitic suite: late Palaeozoic quartz-dolerite dykes of Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 72, 57–74.

MACDONALD, R, THOMAS, J E, and RIZZELLO, S A. 1977. Variations in basalt chemistry with time in the Midland Valley province during the Carboniferous and Permian. Scottish Journal of Geology, Vol. 13, 11–22.

MACDONALD, R, WILSON, L, THORPE, R S, and MARTIN, A. 1988. Emplacment of the Cleveland Dyke: evidence from geochemistry, mineralogy and physical modelling. Journal of Petrology, Vol. 29, 559–583.

MACGREGOR, A G. 1928. The classification of Scottish Carboniferous olivine-basalts and mugearites. Transactions of the Geological Society of Glasgow, Vol. 18, 324–360.

MACGREGOR, A G. 1937. The Carboniferous and Permian volcanoes of Scotland. Bulletin Volcanologique Serie 2, Vol. 1, 41–58.

MACGREGOR, A G. 1944. Barytes in Central Scotland. Wartime Pamphlet of the Geological Survey of Great Britain: Scotland, No. 38.

MACGREGOR, A G. 1948. Problems of Carboniferous-Permian volcanicity in Scotland. Quarterly Journal of the Geological Society of London, Vol. 108, 133–153.

MACGREGOR, A G. 1960. Divisions of the Carboniferous on Geological Survey Scottish Maps. Bulletin of the Geological Survey of Great Britain, No. 16, 127–130.

MACGREGOR, M, LEE, G W, and WILSON, G V. 1920. The iron ores of Scotland. Special Report on the Mineral Resources of Great Britain, Memoir of the Geological Survey, Scotland, Vol. 11.

MACGREGOR, M, and MACGREGOR, A G. 1948. British regional geology: the Midland Valley of Scotland. (2nd edition). (Edinburgh: HMSO.)

MACGREGOR, M, and MANSON, W. 1935. Variations in the thickness of the Carboniferous Limestone Series of Scotland with special reference to the Limestone Coal Group. Transactions of the Institution of Mining Engineers, Vol. 89, 115–130. MACKENZIE, J. 1818. Regarding the remains of an elephant found in Ayrshire. Memoirs of the Wernerian Natural History Society, Vol. 2, 662.

MACNAIR, P. 1917. The Hurlet sequence in the east of Scotland and the Abden fauna as an index to the position of the Hurlet Limestone. Proceedings of the Royal So ety of Edinburgh, Vol. 37, 173–209.

MCLEAN, A C. 1961. Density measurements of rocks in south-west Scotland. Proceedings of the Royal Society of Edinburgh, Vol. 68, Section B, 103–111.

MCLEAN, A C. 1966. A gravity survey in Ayrshire and its geological interpretation. Transactions of the Royal Society of Edinburgh, Vol. 66, 239–265.

MCLEAN, A C, and QURESHI, I R. 1966. Regional gravity anomalies in the western Midland Valley of Scotland. Transactions of the Royal Society of Edinburgh, Vol. 66, 267–283.

MCLEAN, A C, and WALKER,  1978. Gravity and magnetic studies of the Upper Firth of Clyde and Bute* 29–42 in The solid geology of the Clyde Sheet (55°N/06°W). MCLEAN, A C, and DEEGAN, C E (editors). Report of the Institute of Geological Sciences, No. 78/9.

MCLEAN, A C, and WREN, A E. 1978. Gravity and magnetic studies of the Lower Firth of Clyde. 7–27 in The solid geology of the Clyde Sheet (55°N/06°W). MCLEAN, A C, and DEEGAN, C E (editors). Report of the Institute of Geological Sciences, No. 78/9.

MILTON, J J W. 1972. A palaeomagnetic study of Permian and Carboniferous igneous rocks from the Midland Valley of Scotland. Unpublished MSc thesis, University of Newcastle-upon-Tyne.

MONRO, S K. 1982a. Sedimentation, stratigraphy and tectonics Vol. 12, 421–458.

MONRO, S K, LOUGHNAN, F C, and WALKER, M C. 1983. Ayrshire Bauxitic Clay - an allochthonous deposit? 47–58 in Residual deposits. WILSON, R C L (editor). Special Publication of  the Geological Society of London, No. 11.

MONTGOMERY, W. 1839. Outlines of the geology of Renfrewshire and the north of Ayrshire. Prize Essays and Transactions of the Highland and Agricultural Society Scotland, in the Dairy Basin, Ayrshire. Unpublished PhD thesis, University of Edinburgh.

MONRO, S K. 1982b. The Upper Brigantian of Central Strathclyde. Scottish Journal of Geology, Vol. 18, 323–325.

MONRO, S K. 1984. Dinantian sedimentation in the Dairy Basin, Ayrshire. European Dinantian Environments, 1st Meeting 1984. Department of Earth Sciences, Open University, 10–12.

MONRO, S K. 1985. Sedimentation controls in the Midland Valley of Scotland - the Dairy Basin case study. Compte Rendu 10e Congres International de Stratigraphie et de Giologie du Carbonifere, Madrid, 1984, Vol. 3, 343–353.

MONRO, S K. 1986. Sedimentation controls in the Namurian of the Dairy Basin, Ayrshire. Controls of Upper Carboniferous sedimentation, north-west Europe 1986. Abstract, University of Keele, 51–56.

MOORE, D J. 1979. The baryte deposits of central and southern Scotland. Unpublished PhD thesis, University of London.

MOORE, I G. 1859. On a protrusion of Silurian rock in the north of Ayrshire. Quarterly Journal of the Geological Society of London, Vol. 17, 1–4.

MORTON, D J. 1979. Palaeogeographical evolution of the Lower Old Red Sandstone basin in the western Midland Valley. Scottish Journal of Geology, Vol. 15, 97–116.

MYKURA, W. 1960. The replacement of coal by limestone and the reddening of Coal Measures in the Ayrshire Coalfield. Bulletin of the Geological Survey of Great Britain, No. 16, 69–109.

MYKURA, W. 1965. White trap in some Ayrshire coals. Scottish Journal of Geology, Vol. 1, 176–184.

MYKURA, W. 1967. The Upper Carboniferous rocks of southwest Ayrshire. Bulletin of the Geological Survey of Great Britain, No. 26, 23–98.

MYKURA, W. 1991. Old Red Sandstone. 297–346 in Geology of Scotland. CRAIG, G Y (editor). (3rd edition). (London: The Geological Society.)

NEILSON, J. 1874. On some sections of Carboniferous Limestone, near Busby. Transactions of the Geological Society of Glasgow, Vol. 4, 282–290.

NEVES, R, and 5 others. 1973. Palynological correlations within the Lower Carboniferous of Scotland and northern England. Transactions of the Royal Society of Edinburgh, Vol. 69, 23–70.

NEVES, R, GUIENN, K J, CLAYTON, G, IOANNIDES, N, and NEVILLE, R S W. 1972. A scheme of miospore zones for the British Dinantian. Compte Rendu 7e Congrès International de Stratigraphie et de Géologie du Carbonifere, Krefeld, 1971, Vol. 1, 347–353.

PALMER, A, PERRY, S P G, and TATTLING, D H. 1985. Carboniferous magnetostratigraphy. Journal of the Geological Society, London, Vol. 142, 945–955.

PATERSON, I B, and HALL, I H S. 1986. Lithostratigraphy of the late Devonian and early Carboniferous rocks in the Midland Valley of Scotland. Report of the British Geological Survey, Vol. 18, No. 3.

PATERSON, I B, HALL, I H S, and STEPHENSON, D. 1990. Geology of the Greenock district. Memoir of the British Geological Survey, Sheets 30W and part of 29E (Scotland).

PATTERSON, E M. 1945. The distribution of trace-elements in a Scottish Permo-Carboniferous teschenite and its lugaritic differentiate. Geological Magazine, Vol. 82, 230–234.

PATTERSON, E M. 1946. The teschenite-picrite sill of Saltcoats, Ayrshire. Transactions of the Geological Society of Glasgow, Vol. 21, 1–28.

PATTERSON, E M. 1949. The Old Red Sandstone rocks of the West Kilbride-Largs District, Ayrshire. Transactions of the Geological Society of Glasgow, Vol. 21, 207–236.

PEACOCK, J D, GRAHAM, D K, and WILKINSON, I P. 1978. Late-Glacial and post-Glacial marine environments at Ardyne, Scotland, and their significance in the interpretation of the history of the Clyde sea area. Report of the Institute of Geological Sciences, No. 78/17.

PEDLEY, R C. 1991. Interactive 2.5D gravity and magnetic modelling program. User Manual, Integrated Geophysical Services. (Keyworth, Nottingham: British Geological Survey.)

PENN, I E, SMITH, I F, and HOLLOWAY, S. 1984. Interpretation of a deep seismic reflection profile in the Glasgow area. Investigation of the geothermal potential of the UK. (Keyworth, Nottingham: British Geological Survey.)

PHILLIPS, W E A, STILLMAN, CJ, and MURPHY, T. 1976. A Caledonian plate tectonic model. Journal of the Geological Society, London, Vol. 132, 579–609.

PRICE, M, BIRD, M J, and GALE, I N. 1974. Hydraulic properties of the Upper Old Red Sandstone at Hunterston Peninsula, Ayrshire. British Geological Survey Technical Report, WD/ST/ 74/11.

RAMSBOTTOM, W H C. 1977. Correlation of the Scottish Upper Limestone Group (Namurian) with that of the north of England. Scottish Journal of Geology, Vol. 13, 327–330.

RAMSBOTTOM, W H C, and 6 others. 1978. A correlation of Silesian rocks in the British Isles. Geological Society of London Special Report, No. 10.

READ, W A. 1988. Controls on Silesian sedimentation in the Midland Valley of Scotland. 222–241 in Sedimentation in synorogenic basin complex: the Upper Carboniferous of Northwest Europe. BESLEY, B M, and KEELING, G (editors). (Glasgow and London: Blackie and Son.)

RICHEY, J E. 1925. The Carboniferous strata below the Top Hosie Limestone in north Ayrshire. Transactions of the Geological Society of Glasgow, Vol. 17, 204–232.

RICHEY, J E. 1928. The north Ayrshire sequence of Calciferous sandstone volcanic rocks. Transactions of the Geological Society of Glasgow, Vol. 18, 247–255.

RICHEY, J E. 1937. Areas of sedimentation of Lower Carboniferous age in the Midland Valley of Scotland. 93–110 in Summary of Progress of the Geological Survey for 1935. Part 2. (London: His Majesty's Stationery Office.)

RICHEY, J E. 1946. Pene-contemporaneous structures in the Hurlet Limestone near Kilbirnie, north Ayrshire. Transactions of the Geological Society of Glasgow, Vol. 21, 61–73.

RICHEY, J E, ANDERSON, E M, and MACGREGOR, A G. 1930. The geology of north Ayrshire. Memoir of the Geological Survey, Scotland, Sheets 22.

RICHEY, J E, WILSON, G V, and ANDERSON, E M. 1925. The economic geology of the Ayrshire coalfields, Area 1, Kilbirnie, Dairy and Kilmaurs. Memoir of the Geological Survey, Scotland.

ROBERTSON, D. 1877a. Garnock Water Post-Tertiary deposit. Transactions of the Geological Society of Glasgow, Vol. 5, 292–296.

ROBERTSON, D. 1877b. Notes on the Post-Tertiary deposit of Misk Pit near Kilwinning. Transactions of the Geological Society of Glasgow, Vol. 5, 297–309.

ROBERTSON, R H S. 1971. Ayrshire bauxitic clay, a flint clay. Clays and Clay Minerals, Vol. 19, 341.

ROBINS, N S. 1990. Hydrogeology of Scotland. (London: HMSO for the British Geological Survey.)

ROBINS, N S, and BALL, D F. 1991. Groundwater in the Quaternary deposits of Scotland: a valuable resource and a potential engineering hazard. 251–255 in Quaternary Engineering Geology. Geological Society Engineering Geology Special Publication, No. 7.

ROLLIN, K E. 1995. Focused deconvolution and geophysical models of igneous activity in the Midland Valley of Scotland. (Abstract D030), 57th European Association of Geoscientists and Engineers Conference and Technical Exhibition, Glasgow, Scotland. EAGE 1995.

RUSSELL, M J, and SMYTHE, D K. 1983. Origin of the Oslo Graben in relation to the Hercynian-Alleghenian Orogeny and lithosphere rifting in the North Atlantic. 457–472 in Developments in geotectonics, 19: Processes of continental rifting. MORGAN, P, and BARKER, B H (editors). Tectonophysics, Vol. 94.

SCOTT, B. 1967. Barytes mineralisation at Gasswater mine, Ayrshire, Scotland. Transactions of the Institution of Mining and Metallurgy (Section B: Applied earth science), Vol. 76, B40–51.

SCOTT, W B. 1986. Nodular carbonates in the Lower Carboniferous, Cementstone Group of the Tweed Embayment, Berwickshire: evidence for a former sulphate evaporite facies. Scottish Journal of Geology, Vol. 22, 325–345.

SHEARMAN, D J, MOSSOP, G, DUNSMORE, H, and MARTIN, M. 1972. Origin of gypsum veins by hydraulic fracture. Transactions of the Institution of Mining and Metallurgy (Section B: Applied earth science), Vol. 81, B149–155.

SHIELLS, K A G. 1966. A new productid brachiopod from the Upper Visean of Scotland. Palaeontology, Vol. 9, 426–447.

SHIELLS, K A G, and PENN, I E. 1971. Notes on the geology of Trearne Quarry (Upper Visean), Ayrshire and on the palaeoecology of its productid brachipods. Scottish Journal of Geology, Vol. 7, 29–49.

SHOTTON, W F, BLUNDELL, D J, and WILLIAMS, R E G. 1970. Birmingham University Radiocarbon Dates IV. Radiocarbon, Vol. 12, 385–389.

SIMPSON, J B, and MACGREGOR, A G. 1932. The economic geology of the Ayrshire coalfields, Area IV, Dailly, Patna, Rankinston, Dalmellington and New Cumnock. Memoir of the Geological Survey, Scotland.

SISSONS, J B. 1967. Glacial stages and radiocarbon dates in Scotland. Scottish Journal of Geology, Vol. 3, 375–381.

SISSONS, J B, and BROOKS, C L. 1971. Dating of early postglacial land and sea level changes in the western Forth valley. Nature, London, Physical Sciences, Vol. 234, 124–127.

SMEDLEY, P L. 1986a. Petrochemistry of Dinantian volcanism in northern Britain. Unpublished PhD thesis, University of Edinburgh.

SMEDLEY, P L. 1986b. The relationship between talc-alkaline volcanism and within-plate continental rift volcanism: evidence from Scottish Palaeozoic lavas. Earth and Planetary Science Letters, Vol. 77, 113–128.

SMEDLEY, P L. 1988. Trace element and isotope variations in Scottish and Irish Dinantian volcanism: evidence for an OIB-like mantle source. Journal of Petrology, Vol. 29, 413–443.

SMELLIE, W R. 1916. The igneous rocks of Bute. Transactions of the Geological Society of Glasgow, Vol. 15, 363–364.

SMITH, J. 1882. The geology of Dalry. Transactions of the Geological Society of Glasgow, Vol. 6, 277–282.

SMITH, J. 1891. The Great Ice Age in the Greenock Valley. Transactions of the Geological Society of Glasgow, Vol. 9, 151–191.

SMITH, J. 1893. The Ardrossan Shell Mound, with an account of its excavation. Transactions of the Geological Society of Glasgow, Vol. 9, 355–366.

SMITH, J. 1895a. The geological position of the Irvine Whale Bed. Transactions of the Geological Society of Glasgow, Vol. 10, 29–50.

SMITH, J. 1895b. On a bed of ironstone occurring in trap tuff in the parishes of Stevenson, Dalry and Kilwinning. Transactions of the Geological Society of Glasgow, Vol. 10, 133–136.

SMITH, J. 1898. The drift or glacial deposits of Ayrshire. Transactions of the Geological Society of Glasgow, Vol. 11 (Supplement), 1–134.

SPEIGHT, I M, SKELHORN, R R, SLOAN, T, and KNAPP, R J. 1982. The dyke swarms of Scotland. 449–460 in Igneous rocks of the British Isles. SUTHERLAND, D S (editor). (Chichester: Wiley.)

STEDMAN, C. 1988. Namurian E1 tectonics and sedimentation in the Midland Valley of Scotland: rifting versus strike-slip influence. 242–254 in Sedimentation in a synorogenic basin complex: the Upper Carboniferous of northwest Europe. BESLY, B M, and KELLING, G (editors). (Glasgow and London: Blackie and Son.)

STEPHENSON, D, and COATS, S. 1983. Baryte and copper mineralisation in the Renfrewshire Hills, central Scotland. Mineral Reconnaissance Report, Institute of Geological Sciences, No. 67.

STONE, P. 1980. The pre-rifting geology of the southern part of Scotland. British Association, Section C, No. 65.

THOMPSON, R N. 1982. Magmatism of the British Tertiary volcanic province. Scottish Journal of Geology Vol. 18, 49–107.

TOMKEIEFF, S I. 1937. Petrochemistry of the Scottish Carboniferous-Permian igneous rocks. Bulletin Volcanologique Serie 2, Vol. 1, 59–87.

TYRRELL, G W. 1909. The classification of the post-Carboniferous intrusive igneous rocks of the west of Scotland. Transactions of the Geological Society of Glasgow, Vol. 13, 298–317.

TYRRELL, G W. 1912. The late-Palaeozoic alkaline igneous rocks of the west of Scotland. Geological Magazine, Vol. 9, 125–127.

TYRRELL, G W. 1917a. The igneous geol6gy of the Cumbrae Islands, Firth of Clyde. Transactions of the Geological Society of Glasgow, Vol. 16, 244–274.

TYRRELL, G W. 1917b. Some Tertiary dykes of the Clyde area. Geological Magazine, Vol. 4, 305–315 and 350–356.

TYRRELL, G W. 1923. Classification and age of the analcitebearing igneous rocks of Scotland. Geological Magazine, Vol. 60, 249–260.

TYRRELL, G W. 1928a. A further contribution to the petrography of the late-Palaeozoic igneous suite of the west of Scotland. Transactions of the Geological Society of Glasgow, Vol. 18, 259–294.

TYRRELL, G W. 1928b. On some dolerite sills containing analcite-syenite in central Ayrshire. Quarterly Journal of the Geological Society of London, Vol. 84, 540–569.

UPTON, B G J, ASPEN, P, and HUNTER, R H. 1984. Xenoliths and their implications for the deep geology of the Midland Valley of Scotland and adjacent regions. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 75, 65–70.

UPTON, B J G, ASPEN, P, and CHAPMAN, N A. 1983. The upper mantle and deep crust beneath the British Isles: evidence from inclusions in volcanic rocks. Journal of the Geological Society of London, Vol. 140, 105–121.

WAGNER, R H. 1983. A Lower Rotliegend flora from Ayrshire. Scottish Journal of Geology, Vol. 19, 135–155.

WALKER, F. 1934. The term "crinanite". Geological Magazine, Vol. 71, 122–128.

WALLIS, S M. 1989. Petrology and geochemistry of Upper Carboniferous-Lower Permian volcanic rocks in Scotland. Unpublished PhD thesis, University of Edinburgh.

WHYTE, F, and MACDONALD, J G. 1974. Lower Carboniferous vulcanicity in the northern part of the Clyde plateau. Scottish Journal of Geology, Vol. 10, 187–198.

WHYTE, M A. 1981. The Upper Brigantian (Lower Carboniferous) of central Strathclyde. Scottish Journal of Geology, Vol. 17, 227–246.

WILSON, G V. 1916. Preliminary notes on volcanic rocks in north-west Ayrshire. Transactions of the Geological Society of Glasgow, Vol. 16, 86–99.

WILSON, G V. 1922. The Ayrshire bauxitic clay. Memoir of the Geological Survey, Scotland.

WILSON, R B. 1979. The base of the Lower Limestone Group (Visean) in north Ayrshire. Scottish Journal of Geology, Vol. 15, 313–319.

WILSON, R B. 1989. A study of the Dinantian marine macrofossils of central Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 80, 91–126.

YOUNG, J. 1895. On the occurrence of pyroxene or augite in large crystals in two localities in western Scotland. Transactions of the Geological Society of Glasgow, Vol. 10, 13–15.

YOUNG, J, and CRAIG, R. 1869. Notes on the occurrence of seeds of freshwater plants and arctic shells along with the remains of the mammoth arid reindeer in beds under the boulder clay of Kilmaurs. Transactions of the Geological Society of Glasgow, Vol. 3, 310–321.

Author citations for fossil species

To satisfy the rules and recommendations of the international codes of botanical and zoological nomenclature, authors of species cited in Chapters 10 and 15 are listed below.

Chapter 10: Carboniferous

• Actinocyathus floriformis floriformis (Martin, 1809)
• Actinopteria persulcata (McCoy, 1851)
• Actinopteria regularis (Etheridge jun., 1873)
• Alwynia vesicularis (de Koninck, 1851)
• Angiospirifer trigonalis (Martin, 1809)
• Angyomphalus radians (de Koninck, 1843)
• Annuliconcha concentrica (Hind, 1904)
• Anthraconaia cymbula (Wright, 1929)
• Anthraconaia irvinensis Brand, 1989
• Anthraconaia modiolaris (J de C Sowerby, 1840)
• Anthraconaia pulchella Broadhurst, 1959
• Anthraconaia salteri (Leitch, 1940)
• Anthraconaia spathulata Weir, 1967
• Anthraconeilo luciniformis (Phillips, 1836)
• Anthraconeilo laevirostrum (Portlock, 1843)
• Anthraconeilo mansoni Wilson, 1967
• Anthraconeilo pentonensis (Hind, 1899)
• Anthracosia acutella (Wright, 1929)
• Anthracosia atra (Trueman, 1929)
• Anthracosia beaniana King, 1856
• Anthracosia disjuncta Trueman & Weir, 1951
• Anthracosia fulva (Davies & Trueman, 1927)
• Anthracosia ovum Trueman & Weir, 1951
• Anthracosia phrygiana (Wright, 1929)
• Anthracosia regularis (Trueman, 1929)
• Anthracosphaerium affine (Davies & Trueman, 1927)
• Antiquatonia costata (J de C Sowerby, 1827)
• Antiquatonia hindi (Muir-Wood, 1928)
• Antiquatonia insculpta (Muir-Wood, 1928)
• Antiquatonia muricata (Phillips, 1836)
• Antiquatonia sulcata (J. Sowerby, 1822)
• Apatognath us geminus (Hinde, 1900)
• Aulina rotiformis Smith, 1917
• Aviculopecten inaequalis Hind, 1903
• Aviculopecten knockonniensis (McCoy, 1844)
• Aviculopecten plicatus (Sowerby, 1823)
• Aviculopecten tabulatus (McCoy, 1844)
• Amnia youngiana (Davidson, 1860)
• Baylea parka (E G Thomas, 1940)
• Bellerophon anthracophilus Frech, 1906
• Bellerophon sowerbii (d'Orbigny, 1840)
• Borestus wrighti (E G Thomas, 1940)
• Calymmatotheca stangeri (Stur, 1877)
• Calymmatotheca stangeri schlehani (Stur, 1877)
• Carbonicola oslancis Wright, 1929
• Carbonicola pseudorobusta Trueman, 1929
• Carbonicola subconstricta (J Sowerby, 1813)
• Cardiomorpha hindi Wilson, 1960
• Cardiomorpha parka Hind, 1898
• Claviphyllum eruca (McCoy, 1851)
• Composita ambigua (J Sowerby, 1822)
• Crurithylis urii (Fleming, 1828)
• Cypricardella rectangularis (McCoy, 1844)
• Cyrtoceras .rugosum (Fleming, 1828)
• Dibunophyllum linnense Hill, 1939
• Diphyphyllumfasciculatum (Fleming, 1828)
• Echinoconchus defensus (Thomas, 1914)
• Echinoconchus elegans (McCoy, 1855)
• Echinoconchus punctatus (J Sowerby, 1822)
• Echinoconchus subelegans (Thomas, 1914)
• Edmondia josepha de Koninck, 1842
• Edmondia maccoyi Hind, 1899
• Edmondia punctatella (Jones, 1865)
• Edmondia senelis (Phillips, 1836)
• Edmondia sulcata (Fleming, 1828)
• Edmondia unioniformis (Phillips, 1836)
• Eomarginfera lobata Sowerby, 1821)
• Eomor inijera longispina (J Sowerby, 1814)
• Eomarginifera setosa (Phillips, 1836)
• Euphemites ardenensis (Weir, 1931)
• Euphemites urii (Fleming, 1828)
• Geisina arcuata (Bean, 1836)
• Globosochonetes parseptus Brunton, 1968
• Hesperiella thomsoni (de Koninck, 1883)
• Hyalostelia parallela (McCoy, 1844)
• Isogramma pachti (Dittmar, 1872)
• Kochiproductus coronus Shiells, 1968
• Krotovia spinulosa (J. Sowerby, 1814)
• Lachiymula latior (McCoy, 1852)
• Latischisma globosa E G Thomas, 1940
• Leiopteria lunulata (Phillips, 1836)
• Leiopteria thompsoni (Portlock, 1843)
• Lepidodendron rhodeanum Sternberg, 1825
• Lepidophloios scoticus Kidston, 1885
• Leptagonia caledonica Brand, 1972
• Leptagonia smithi Brand, 1972
• Limipecten dissirnilis (Fleming, 1828)
• Lingula rnytilloides (J Sowerby, 1812)
• Lingula squamiformis Phillips, 1836
• Lingula straeleni Demanet, 1934
• Lithophaga lingualis (Phillips, 1836)
• Liralingua wilsoni Graham, 1970
• Lonsdaleia caledonica Smith, 1916
• Macrochilina obtusa de Koninck, 1881
• Naiadites alatus Weir, 1956
• Naiadites crassus (Fleming, 1828)
• Naiadites obliques Dix & Trueman, 1932
• Naiadites quadrates (J de C Sowerby, 1840)
• Neuropteris gigantea (Sternberg, 1821)
• Nuculopsis gibbosa (Fleming, 1828)
• Orbiculoidea bulla (McCoy, 1852)
• Orbiculoidea cincta (Portlock, 1843)
• Orbiculoidea craigii (Davidson, 1877)
• Orthonema pygmaeum Donald, 1889
• Overtonia fimbriata (J de C Sowerby, 1824)
• Paladin eichwaldi (Fischer de Waldheim, 1825)
• Paleyoldia macgregori Wilson, 1967
• Paracarbonicola peruetusta (Bennison, 1954)
• Pernopecten sowerbii (McCoy, 1844)
• Phestia attenuata (Fleming, 1828)
• Planolites ophthalmoides Jessen, 1949
• Posidonia comigata (Ethelitidge jun., 1873)
• Productus carbonarius de Koninck, 1842
• Productus concinnus (J Sowerby, 1821
• Promarginifera trearnensis Shiells, 1966
• Pterinopectinella dumontiana (de Koninck, 1843)
• Pterinopectinella granola (J de C Sowerby, 827)
• Pugilis kilbridensis (Muir-Wood, 1928)
• Pugilispugilis (Philliik, 1836)
• Pugilisscotica Sowerby, 1814)
• Pustula pustulosa (Phillips, 1836)
• Retispira decussata (Fleming, 1828)
• Retispira exilis (de Koninck, 1883)
• Retispira striata (Fleming, 1828)
• Rhodea goepperti (Ettingshausen, 1865)
• Rugosochonetes celticus Muir-Wood, 1962
• Sanguinolites abdenensis Etheridge jun., 1877
• Sanguinolites costellatus (McCoy, 1851)
• Sanguinolites plicatus (Portlock, 1843)
• Sanguinolites striatolamellosus de Koninck, 1842
• Schizodus axiniformis (Phillips, 1836)
• Schizodus portlocki (Brown, 1849)
• Schizophoria resupinata (Martin, 1809)
• Schuchertella radialis (Phillips, 1836)
• Sedgwickia gigantea McCoy, 1844
• Sedgwickia lirnosa (Fleming, 1828)
• Sedgwickia ovata Hind, 1899
• Sedgwickia scotica Hind, 1899
• Sedgwickia subothicularis Hind, 1899
• Serpuloides carbonarius (McCoy, 1844)
• Sinuatella sinuata de Koninck, 1851
• Siphonodendron junceum (Fleming, 1828)
• Siphonodendron pauciradiale (McCoy, 1844)
• Solemya primaeva Phillips, 1836
• Solenomorpha minor (McCoy, 1844)
• Sphenophyllum tenerrimum Stur, 1868
• Spirifer duplicicosta (Phillips, 1836)
• Spirifer striates (Martin, 1809)
• Stigmaria ficoides (Sternberg, 1820)
• Straparolus (Euomphalus) carbonarius (J. de C Sowerby, 1814)
• Streblochondria elliptica (Phillips, 1836)
• Streblopteria ornata (Etheridge jun., 1873)
• Strobeus obtusus (de Koninck, 1881)
• Strobeus ventricosus (de Koninck, 1881)
• Telangium affine (Lindley & Hutton, 1832)
• Thysanophyllum minus Thompson, 1880
• Tornquistia scotica Brand, 1970
• Tornquistia youngi Wilson, 1966
• Trigonoglossa scotica (Davidson, 1860)
• Tylonautilus nodiferus (Armstrong, 1866)
• Wilkingia elliptica (Phillips, 1836)
• Wilkingia maxima (Portlock, 1843)
• Zaphrentites constricts (Carruthers, 1910)
• Zaphrentites disjuncta (Carruthers, 1910)

Chapter 15: Quaternary

• Acteon tornatilis (Linné, 1758)
• Arctica islandica (Linné, 1767)
• Astarte borealis (Schumacher, 1817)
• Astarte elliptica (Brown, 1827)
• Astarte montagui (Dillwyn, 1817)
• Bos primigenius Bojanus, 1826
• Buccinum cyaneum Bruguière, 1792
• Caecum glabrum (Montagu, 1803)
• Chlamys islandica (Muller, 1776)
• Coelodonta antiquitatis Blumenbach, 1803
• Corbula gibba (Olivi, 1792)
• Hiatella arctica (Linne, 1767)
• Littorina littorea (Linne, 1758)
• Lunatia pallida (Broderip & Sowerby, 1829)
• Macoma calcarea (Gmelin, 1790)
• Mammuthus (Elephas) primigenius Blumenbach, 1799
• Melobesia polymorpha Linné, 1767
• Monodonta lineata (da Costa, 1778)
• Mya truncates Linné,  1758
• Natica groenlandica Möller, 1842
• Nuculana pernula (Muller, 1776)
• Onoba semicostata (Montagu, 1803)
• Rangifer tarandus Untie, 1758
• Thyasira gouldi (Philippi, 1845)
• Turritella communis Risso, 1826
• Velutina undata Brown, 1839
• Xylophaga dorsalis (Turton, 1819)
• Yoldiella fraterna (Verrill & Bush, 1898)
• Yoldiella lenticula (Möller, 1842)

Figures, plates and tables

Figures

(Figure 1a) Outline geology of the Irvine district: sedimentary and extrusive igneous rocks.

(Figure 1b) Principal igneous intrusions of the Irvine district.

(Figure 2) Location and topography of the Irvine district.

(Figure 3) Reconstruction of the Stratheden Group palaeogeography in central Scotland.

(Figure 4) Grain-size analysis of sandstones from Fairlie Glen and the Vale of Eden.

(Figure 5) Stratigraphical relationships of the Inverclyde Group formations in the northern part of the Irvine district.

(Figure 6) Distribution of members within the Clyde Plateau Volcanic Formation in the Kilbirnie Hills.

(Figure 7) Distribution of members within the Clyde Plateau Volcanic Formation in the Beith–Barrhead Hills.

(Figure 8) Range of composition of Visean igneous rocks in the Irvine district.

(Figure 9) Thickness variations of the Kirkwood Formation.

(Figure 10) Correlation of Visean limestones between Irvine and Paisley. Adapted from Wilson (1979).

(Figure 11) Borehole sections in the Kirkwood and Lower Limestone formations in the northern part of the Irvine district.

(Figure 12) Geographical extent of some of the marine limestones in the northern part of the Irvine district

(Figure 13) Dockra Limestone: variation in thickness and lithofacies, and distribution of the White Post.

(Figure 14) Dockra Limestone to Top Hosie Limestone: thickness variation of the interval, and distribution of sandstones within it.

(Figure 15) Generalised sections of Namurian and Westphalian rocks in the Irvine district.

(Figure 16) Limestone Coal Formation: generalised section north of the Dusk Water Fault.

(Figure 17) Isopach map for the Limestone Coal Formation.

(Figure 18) Lithology ratio map for the Limestone Coal Formation.

(Figure 19) Borehole sections through the Upper Limestone Formation and the lower part of the Passage Formation in the northern part of the Irvine district.

(Figure 20) Thickness variations of the Upper Limestone Formation in the northern part of the Irvine district.

(Figure 21) Thickness variations of the Troon Volcanic Member.

(Figure 22) Composition of the Troon and Mauchline lavas, and of late Carboniferous to Permian sills in the Irvine district.

(Figure 23) Sections in the Ayrshire Bauxitic Clay Member.

(Figure 24) Generalised section of Coal Measures between the Dusk Water and Inchgotrick faults

(Figure 25) Bouguer gravity anomalies in the Irvine district and surrounding area.

(Figure 26) Aeromagnetic anomalies in the Irvine district and surrounding area.

(Figure 27) Section A: geophysical profiles and inferred crustal structure, part of a 2.5D crustal model across the western Midland Valley.

(Figure 28) Section B: geophysical profiles and inferred crustal structure, part of a 2.5D crustal model across the western Midland Valley.

(Figure 29) Main structural features of the Irvine district.

(Figure 30) Glacial features in the Irvine district.

Plates

(Plate 1) Lenses of vein-quartz conglomerate within sandstones of the Seamill Sandstone Formation, Largs shore (D1483).

(Plate 2) Trough cross-bedding in sandstones of the Seamill Sandstone Formation, Seamill shore (D1480).

(Plate 3) Cornstone bed in the Kinnesswood Formation, Meikle Busbie (D 2449).

(Plate 4) Nodular cornstone in the Foul Port Mudstone Member of the Kinnesswood Formation, Great Cumbrae (D3677).

(Plate 5) Section in the argillaceous Lugton facies of the Dockra Limestone, north end of Trearne Quarry (D2435).

(Plate 6) Bioclastic limestone of the Trearne facies of the Dockra Limestone with colonial coral Siphonodendron junceum, south end of Trearne Quarry (D2433).

(Plate 7) Structures at the base of the White Post, Paduff Burn (D2442).

(Plate 8) Old Mill Quarry, showing the relationship between the White Post and the rest of the Dockra Limestone (D2761).

(Plate 9) Photomicrograph of the top of the Dockra Limestone, Old Mill Quarry, showing ostracods and algal stromatolites (MNS3212).

(Front cover) Cover photograph: Main rock platform south of Seamill with prominent cliff feature at 15 m OD

(Rear cover)

Tables

(Table 1) Dewatering rates for former mines and pits.

(Table 2) Major ion concentrations (mg l⁻¹) in water from the Coal Measures.

(Table 3) Correlation of Lower Devonian rocks at Hunterston with those at Glen Rosa, Arran. Based on Downie and Lister (1969).

(Table 4) Formations in the Stratheden Group.

(Table 5) Grain-size analyses of sandstones from Fairlie Glen and the Vale of Eden. Information from an Open University undergraduate study by M J C Ainsworth (1989), supervised by S K Monro.

(Table 6) Classification of the Carboniferous strata.

(Table 7) Lithostratigraphy of the Inverclyde Group in the Irvine district.

(Table 8) Whole rock analyses of Visean igneous rocks from the Irvine district.

(Table 9) Nomenclature of basic igneous rocks of Carboniferous and Permian age in the Midland Valley of Scotland.

(Table 10) Classification of Namurian strata in the Irvine district and comparison with that for the central part of the Midland Valley of Scotland.

(Table 11) K-Ar age determinations for the Troon Volcanic Member. From Wallis (1989).

(Table 12) Whole rock analyses of lavas of the Troon Volcanic Member from the Irvine district.

(Table 13) Taxa in the principal Carboniferous marine horizons of the Irvine district.

(Table 14) Whole rock analyses from the Irvine district: lavas of the Mauchline Volcanic Formation and associated intrusions and blocks in vents.

(Table 15) Recorded xenolith types from vents in the Irvine district.

(Table 16) Whole rock analyses of sills from the Irvine district.

(Table 17) Whole rock analyses of Palaeogene tholeiitic dykes from the Irvine district and adjacent areas.

(Table 18) Summary of the late Devensian and Flandrian events recorded in the Irvine district.

Tables

(Table 1) Dewatering rates for former mines and pits

(Table 2) Major ion concentrations (mg 1⁻¹) in water from the Coal Measures

(Table 3) Correlation of Lower Devonian rocks at Hunterston with those at Glen Rosa, Arran. Based on Downie and Lister (1969)

(Table 5) Grain-size analyses of sandstones from Fairlie Glen and the Vale of Eden

Information from an Open University undergraduate study by M J C Ainsworth (1989), supervised by S K Monro.

(Table 8) Whole rock analyses of Viséan igneous rocks from the Irvine district

(Table 9) Nomenclature of basic igneous rocks of Carboniferous and Permian age in the Midland Valley of Scotland

(Table 11) K-Ar age determinations for the Troon Volcanic Member

(Table 12) Whole rock analyses of lavas of the Troon Volcanic Member from the Irvine district