The Dalradian rocks of Scotland: an introduction

D. Stephenson, J. R. Mendum, D.J. Fettes and A.G. Leslie

Accepted manuscript. Published version: David Stephenson, John R. Mendum, Douglas J. Fettes, A. Graham Leslie, The Dalradian rocks of Scotland: an introduction, Proceedings of the Geologists' Association, Volume 124, Issues 1–2, 2013, Pages 3–82, ISSN 0016-7878, https://doi.org/10.1016/j.pgeola.2012.06.002.

* David Stephenson British Geological Survey, Murchison House, West Mains Road, Edinburgh EH9 3LA.

John R. Mendum British Geological Survey, Murchison House, West Mains Road, Edinburgh EH9 3LA.

Douglas J. Fettes British Geological Survey, Murchison House, West Mains Road, Edinburgh EH9 3LA.

A. Graham Leslie British Geological Survey, Murchison House, West Mains Road, Edinburgh EH9 3LA.

* Corresponding author

Keywords:

Geological Conservation Review; Dalradian Supergroup; History of Research; Metamorphism; Tectonic evolution

Abstract

The Dalradian Supergroup and its basement rocks, together with younger plutons, underpin most of the Grampian Highlands and the islands of the Inner Hebrides between the Highland Boundary and Great Glen faults. The Dalradian is a mid-Neoproterozoic to early-Ordovician sequence of largely clastic metasedimentary rocks, with some volcanic units, which were deformed and metamorphosed to varying degrees during the Early Palaeozoic Caledonian Orogeny.

Sedimentation of the lower parts of the Dalradian Supergroup, possibly commencing about 730 million years ago, took place initially in fault-bounded rift basins, within the supercontinent of Rodinia and adjacent to sectors of continental crust that were later to become the foundations of North America, Greenland and Scandinavia. Later sedimentation reflected increased instability, culminating between 600 and 570 million years ago in continental rupture, volcanicity and the development of the Iapetus Ocean. This left the crustal foundations of Scotland, together with those of North America and Greenland, on a laterally extensive passive margin to the new continent of Laurentia, where turbiditic sedimentation continued for about 85 million years. Later plate movements led to closure of the Iapetus Ocean and the multi-event Caledonian Orogeny. Most of the deformation and metamorphism of the Dalradian strata peaked at about 470 million years ago, during the mid-Ordovician Grampian Event, which has been attributed to the collision of an oceanic arc with Laurentia. The later, mid-Silurian Scandian Event, attributed to the collision of the continent of Baltica with Laurentia and the final closure of the Iapetus Ocean, apparently had little effect on the Dalradian rocks but marked the start of late-orogenic uplift and extensive magmatism in the Grampian Highlands that continued until Early Devonian times.

The Dalradian rocks thus record a wide range of sedimentary environments (alluvial, tidal, deltaic, shallow marine, turbiditic, debris flow) and a complex structural and metamorphic history. In areas of low strain, original sedimentary and volcanic structures are well preserved, even at relatively high metamorphic grades. There is convincing evidence for glacial episodes of worldwide importance and economic deposits of stratiform barium minerals are unique. The Grampian Highlands include two of the World's type-areas for metamorphic zonation, Barrovian and Buchan, with spectacular examples of the key metamorphic minerals, and various stages of migmatite development. Polyphase folding is widespread on all scales and gives rise to a range of associated cleavages and lineations. Regional dislocations, both ductile and brittle, are associated with a range of shear fabrics, breccias, clay gouges and veining.

1.1 Introduction

D. Stephenson

1.1.1 The Dalradian Supergroup

The Dalradian Supergroup is a mid-Neoproterozoic to Early Palaeozoic sequence of largely clastic sedimentary rocks, with some notable carbonate and volcanic units that were all deformed and metamorphosed to varying degrees during the mid-Ordovician Grampian Event of the Caledonian Orogeny. The Dalradian rocks, together with Caledonian intrusive igneous rocks, form the bedrock to most of the Grampian Highlands of Scotland and the islands of the Inner Hebrides between the Highland Boundary and Great Glen faults. Pre-Dalradian basement crops out in parts of the Northern Grampian Highlands and on the Isle of Islay (Figure 1.1). The Dalradian sequence, its basement and the Caledonian intrusions comprise the Grampian Terrane, one of several major crustal blocks that were juxtaposed during the Caledonian Orogeny to form the northern part of the British Isles (Figure 1.2). Dalradian rocks also occur in the Shetland Islands, east of the Walls Boundary Fault, conventionally as part of the Grampian Terrane but possibly part of a separate terrane. Siluro–Devonian and Mesozoic cover rocks crop out mainly around the margins of the Grampian Highlands.

The Grampian Terrane extends south-westwards into the northern and north-western parts of Ireland, where Dalradian rocks crop out over wide areas (Figure 1.2). There, the south-eastern terrane boundary is largely buried beneath younger rocks and is difficult to define. An extension or major splay of the Highland Boundary Fault probably does extend from Cushendun in the east to Clew Bay in the west but, unlike in Scotland, it is not defined by any strong geophysical feature. A Dalradian sequence with remarkable similarities to that of the Grampian Highlands also crops out in Connemara, well to the south of Clew Bay, and hence it seems likely that the boundary of the Grampian Terrane does not coincide with an extension of the Highland Boundary Fault in the west of Ireland and possibly extends south-eastwards as far as a line between south Antrim and Galway (Ryan et al., 1995).

On a broader scale, the terranes of the northern British Isles are inherently linked geologically to eastern North America and Greenland, which were in close proximity prior to the opening of the North Atlantic Ocean in Palaeogene times (c. 55 Ma ago) (Figure 1.3). The Dalradian Supergroup is similar in age to the Fleur de Lys Supergroup in Newfoundland (Kennedy, 1975) and its lower parts are equivalent to the Eleonore Bay Supergroup in East Greenland (Soper, 1994b; Leslie et al., 2008); the three sequences might have been deposited in adjacent basins. The Geological Conservation Review, and hence this volume, considers only the Dalradian rocks of Scotland and Shetland; for reviews of the Irish Dalradian see Alsop and Hutton (1990), Leake and Tanner (1994), Harris et al. (1994), Cooper and Johnston (in Mitchell, 2004) and chapters by J.S. Daly and D.M. Chew in Holland and Sanders (2009).

The current best estimate for the age of deposition of the oldest Dalradian rocks, adopted in this volume, is about 730 Ma. The youngest strata that can be assigned to the Dalradian has been a matter of recent debate, but it is now generally accepted that there is stratigraphical and structural continuity through from undisputed Dalradian strata into fossiliferous strata of Early Cambrian age (c. 515 Ma) and possibly continuing up into mid-Arenig age strata (Tanner and Pringle, 1999; Tanner and Sutherland, 2007). Those Early Palaeozoic strata were formerly thought to have been juxtaposed tectonically against the Dalradian sequence and consequently they have been described separately in the British Cambrian to Ordovician Stratigraphy GCR volume (Rushton et al., 1999). They were assigned originally to the Highland Border Complex, which also includes elements of an ophiolite obducted during the Caledonian Orogeny, but Tanner and Sutherland (2007) have suggested that they should be designated as a separate ‘Trossachs Group’ and included in the Dalradian Supergroup (see Introduction to Chapter 4).

The name ‘Dalradian’ is derived from that of the ancient Scots’ kingdom of Dalriada, which united the coastal areas of Argyll, Arran and Antrim between the 5th and 9th centuries AD. It was first applied to all of the metamorphic rocks that crop out between the Moine Thrust and Highland Boundary Fault by Sir Archibald Geikie in 1891. However, in his explanatory notes to the 1892 edition of Bartholomew’s 10-miles-to-one-inch Geological Map of Scotland, he did make it clear that the ‘Moine Schists’ of the Northern Highlands are different in character to the ‘Dalradian’ rocks south-east of the Great Glen. As survey work progressed, quartzofeldspathic rocks of apparent ‘Moine Schist’ facies were also identified in the northern part of the Grampian Highlands and the term ‘Dalradian’ sensu stricto became restricted to the lithologically more-diverse strata now assigned to the Appin, Argyll and Southern Highland groups of the Dalradian Supergroup.

The so-called ‘Moine’ rocks of the Grampian Highlands have been the subject of much debate and revision of assignment and terminology over the past 100 years. Originally they were referred to variously, from north-east to south-west, as ‘Granulitic Schists of the Central Highlands’ (Hinxman and Anderson, 1915), ‘Struan Flags’ (Barrow, 1904) and ‘Eilde Flags’ (Bailey, 1910). Subsequently, J.G.C. Anderson (1948) reviewed the local successions and proposed that all rocks stratigraphically below a lowermost limestone should be included in a ‘Moinian Metamorphic Assemblage’, and Johnstone (1975) introduced the term ‘Younger Moines’ to distinguish them from the Moine rocks north-west of the Great Glen Fault. Detailed mapping in the later part of the 20th century resulted in a subdivision into an older, largely migmatitic ‘basement’ referred to as the ‘Central Highland Division’ and an overlying sequence dominated by non-migmatitic quartzofeldspathic rocks termed the ‘Grampian Division’ (Piasecki and Van Breemen, 1979a,b; Piasecki, 1980). The non-migmatitic ‘cover’ rocks are now assigned formally to the Grampian Group, the lowest group in the Dalradian Supergroup, on the basis of stratigraphical and structural continuity with the overlying sequences (Harris et al., 1978). The ‘basement’ rocks, referred to for a while as the ‘Central Highland Migmatite Complex’ (Harris et al., 1994; Stephenson and Gould, 1995), and referred to in more-recent literature as the Dava and Glen Banchor successions, have been formalised as the Badenoch Group. They might yet prove to be equivalent to the Moine Supergroup of the Northern Highlands, at least in terms of their age (see Introduction to Chapter 5).

Two other units within the Grampian Terrane, whose stratigraphical affinities are uncertain but are most probably Dalradian, are represented by GCR sites in this volume. These are the Bowmore Sandstone Group of Islay and the Colonsay Group of Islay, Oronsay and Colonsay.

Much of the terminology used in this volume is necessarily complex. It is evolving continuously, hopefully towards simpler, more-logical versions that are likely to gain widespread acceptance. Hence it will differ in parts from what has been used historically and even from current usage by some authors. In addition to a glossary of terms used, the ‘Glossary and terminology’ section at the end of the volume includes brief details and explanations of radiometric dating methods and timescale adopted, lithological nomenclature (rock names), the method of numbering of structures associated with phases of deformation, the terminology of fold geometry, and the use of the stereographic projection. In each case emphasis is placed upon terms as used in this volume.

1.1.2 History of research

In 1774, the prominent ‘Dalradian’ mountain of Schiehallion was the location for a ground-breaking experiment by the Astronomer Royal, Nevil Maskelyne, who set out to measure the gravitational attraction towards the quartzite mountain of a weight suspended on a plumb-line (Smallwood, 2007). The density of the Earth was estimated to be 4.5, possibly the earliest ever geophysical calculation. However, the earliest published description of Dalradian rocks was probably in James Hutton’s Theory of the Earth (1788), in which the country rocks intruded by granite veins at the critical historical site in Glen Tilt were referred to as ‘Alpine schistus’ (see the Forest Lodge GCR site report; in the Caledonian Igneous Rocks of Great Britain GCR volume; Stephenson et al., 1999).

The first geological map of Scotland (one of the first geological maps in the world), by Louis Albert Necker de Saussure in 1808, did not differentiate the Dalradian rocks from the Moine of the Northern Highlands or the Lewisian of the Hebridean Terrane, all of which were described as ‘primitive rocks stratified as gneiss, mica slate and clay slate’. Subsequently, John MacCulloch’s 1836 map of Scotland showed several distinct metamorphic lithologies within the three, still undivided, terranes. MacCulloch had surveyed almost the whole of Scotland and Shetland himself (Bowden, 2007) and was one of the first to describe and comment upon the origin of the Highland rocks (e.g. MacCulloch, 1814, 1819). In his geological essay on Scotland, Ami Boué (1820) attempted to correlate the metamorphic rocks of Shetland with those of Scotland and was the first to suggest that those of the eastern Mainland of Shetland are equivalent to those of the Grampian Highlands. A truly remarkable geological map from this early period is that of Shetland by Samuel Hibbert (1822), which even shows foliations and lineations in the Dalradian rocks. This was the world’s first purely structural geological map and was published in conjunction with an account of tectonite rock fabric (another world first).

By the middle of the 19th century, the metamorphosed nature of the rocks was beginning to be understood but the degree of metamorphism was thought by many to be proportional to the age of the rocks, and many geologists still adhered to the Wernerian doctrine, in which crystalline rocks were thought to have been precipitated in a regular sequence from a primeval ocean (Oldroyd and Hamilton, 2002). For many years these misconceptions greatly hindered attempts to fit the main sequences of the Highlands into an overall timescale. Murchison (1851, 1859) tried to show that the rocks of the Grampian Highlands are more highly metamorphosed equivalents of those in the Southern Uplands and Geikie (1865), although recognizing them as older than those of the Southern Uplands, still assigned them to a ‘Lower Silurian’ unit. The influence of Murchison and Geikie ensured that their ideas persisted into the next century, although as late as the 1890s George Barrow was arguing that all of the Highland ‘schists’ were of the same age as the Lewisian gneisses, and merely displayed different degrees of metamorphism.

Eventually, people such as James Nicol (1844, 1852, 1863), professor at Cork and then Aberdeen University, and Robert Harkness (1861) of Cork, whilst recognizing the existence and significance of large-scale folding, started to identify local successions and trace them along strike to establish a regional lithostratigraphy. Then Archibald Geikie, who had been introduced to the Grampian Highlands by Murchison in 1860, began to rationalize the overall Scottish succession and to postulate large-scale overfolds as a means of repeating the stratigraphy (Murchison and Geikie, 1861; Geikie, 1865). Of particular note in this context is the slightly later work of Peter Macnair (1896, 1906, 1908), Curator of Natural History at the Glasgow Museums, who recognized the value of the Loch Tay Limestone as a datum line in reconstructing the broad-scale structural framework of the south-east Grampian Highlands. Remarkably, the correlation of ‘boulder beds’ on the Garvellach Isles, on Islay and at Schiehallion had been recognized early in the 19th century by MacCulloch (1819), and those on Islay were first interpreted as having a glacial origin as early as 1877 by J. Thomson. By the end of the 19th century, knowledge of large-scale nappe structures in the Alps was being applied to Scotland by the Geological Survey, initially in the North-west Highlands and subsequently to the Grampian Highlands. Systematic field surveys were underway and petrographical techniques and methods of chemical analysis were developing fast.

Although the Geological Survey had started work in Scotland in 1854, mapping to the north of the Highland Boundary Fault did not start until 1875. For the first few years this was concentrated on the sedimentary rocks around the fringes of the Grampian Highlands, but by the early 1880s, with Archibald Geikie as Director General, J. Horne, J.S. Grant Wilson, L.W. Hinxman and J. Linn were mapping large areas of the North-east Grampian Highlands and W. Gunn, C.T. Clough, J.B. Hill and H.M. Cadell were active on the Cowal peninsula. The work in Cowal led to the recognition of polyphase deformation, and the first use of minor structures to interpret the overall pattern of major folds (Clough, in Gunn et al., 1897). In 1895, work started in the Glen Coe–Appin area, with a highly experienced team that included B.N. Peach and C.T. Clough, after completion of their ground-breaking survey of the North-west Highlands, and J.S. Grant Wilson. They were joined, among others, by H.B. Maufe and E.B. Bailey who were soon to make their mark with their exposition of cauldron subsidence in the volcanic rocks of the area, as well as unravelling the complex structure and proposing a succession for the Dalradian strata. Thereafter, until well into the 20th century, survey work was concentrated in the South-west Highlands, where large areas were surveyed by J.B. Hill, H. Kynaston, R.G. Symes, S.B. Wilkinson and others. The notable exception at that time was the work of G. Barrow and E.H. Cunningham Craig in north-east Perthshire and Angus, which led to Barrow’s (1893) seminal work on metamorphic zones. Although Barrow initially attributed the zonation to contact metamorphism above concealed igneous masses, ironically the ‘Barrovian zones’ became the world standard for the progressive regional metamorphism of aluminous sedimentary rocks at medium pressure. As this concept was extended throughout the Highlands it played a major part in deciphering the geological history of the whole region.

There then followed a period dominated by the work of Edward Battersby Bailey, who was undoubtedly the most dynamic, prolific and controversial figure ever to set foot upon Dalradian rocks. Bailey was decorated for bravery in the First World War and legends abound concerning his resilience and eccentricity in the field (Figure 1.4). After distinguishing himself with the Geological Survey, and then becoming disillusioned by official restrictions upon his personal scientific investigations, he left in 1930 to take a chair at the University of Glasgow. However, he subsequently returned to the survey as Director in 1937, was knighted in 1945 and continued to publish on Highland geology well into retirement.

Bailey started work in the Highlands in 1902, and in a series of papers between 1910 and 1940 he elucidated the structure and stratigraphy of Dalradian rocks over a wide swathe of ground from Islay to Loch Awe, Loch Leven, Glen Roy, Glen Orchy, Glen Lyon, Schiehallion, Loch Tummel, Blair Atholl, Glen Shee, Glen Clunie and Braemar. Most of this was based upon his own mapping, much of it in his own time, but he also re-interpreted ground that had already been covered by others, most notably Barrow, with whom he had fundamental disagreements. This was all in addition to his ground-breaking work on the igneous rocks of Glen Coe and Mull. He could never be described as a precise detailed mapper, and was certainly not out of the same mould as the likes of Clough, Peach and Horne, but he moved over the ground at incredible speed and was a master at tracing out large-scale structures. In fact most of the major folds recognized today in the above areas are the ones that were first identified by Bailey.

It was Bailey who first made use of the ungainly verb ‘to young’ and coined the terms ‘antiform’ and ‘synform’. He also introduced the concept of ‘slides’ (or ‘fold-faults’) – very low-angle faults associated with recumbent folds, which would be described generically as 'ductile dislocations' in modern parlance. These he divided into extensional ‘lags’, along which fold limbs and significant parts of the succession can be excised, and compressional ‘thrusts’, which generally result in the repetition of parts of the succession. In 1922 he produced a comprehensive synthesis in which major slides were perceived as fundamental tectonic dislocations separating huge, Alpine-scale nappe complexes, each complex having its own stratigraphical succession and structural style. Initially three such nappe complexes were recognized and named, in ascending structural order, the Ballappel Foundation (from the type areas of Ballachulish, Appin and Loch Eilde), the Iltay Nappe (comprising most of the ground between Islay and Loch Tay), and the Loch Awe Nappe, but these were subsequently reduced to two when Bailey (in Allison, 1941) accepted the stratigraphical correlations of other investigators which removed the need to invoke a separate Loch Awe Nappe. We now know that both the structure and the stratigraphy of the Grampian Highlands are more integrated and continuous than inferred by Bailey, and hence many of the slides have lost much of their original significance as major tectonic and stratigraphical boundaries.

At the beginning of the 20th century, the use of sedimentary structures as way-up criteria was virtually unknown and Bailey, along with everyone else, struggled to establish the correct order of succession. Possibly the first use of graded bedding in the Scottish Highlands to determine the younging direction was by Peach et al. (1911) at a location within the Kilmory Bay GCR site, and current-bedding was first used by J.F.N. Green on Islay (Green, 1924). But the key event happened in 1924, when two graduate students from the University of Wisconsin, Sherwood Buckstaff and Olaf Rove, accompanied Thorolf Vogt of the University of Trondheim on a visit to Ballachulish and deduced that the succession established by Bailey (1910, 1922) was in fact upside down. This radical new interpretation was communicated to Bailey who was then invited to attend a Princeton University summer school in Canada in 1927, where he became convinced of the need to apply sedimentary structures to interpret folded rocks. As a result of this, in 1929, he acted as guide for a visit by a Princeton group to Scotland, where all agreed that his original (1922) order of succession needed to be reversed. Vogt and Bailey published their findings as linked articles in the Geological Magazine (Vogt, 1930; Bailey, 1930), and Bailey went on to re-appraise key sections around Loch Leven and throughout the whole area, confirming the presence of recumbent folds of many kilometres amplitude with extensive inverted limbs. The legacy of this fundamental ‘about face’ lives on to this day in the archives of the British Geological Survey (BGS), where the original 6-inch to 1 mile maps of the Loch Leven area, prepared prior to 1924, still show Bailey’s original succession.

Survey mapping was severely curtailed by the First World War (1914–18) although a detailed revision of part of the North-east Grampian Highlands was carried out by H.H. Read, after he had been discharged from active service on medical grounds. That work formed the basis of a lifetime of investigation into the structure, metamorphism and magmatism of the North-east Grampians, in a career that ranged widely through the Scottish and Irish Caledonides and took him from the Geological Survey to the University of Liverpool and the Royal School of Mines (later Imperial College), London. He was the first to recognise that the Buchan metamorphism originated at lower pressures and higher temperatures than the Barrovian sequence, became even more-widely known for his survey work in the Northern Highlands, and was also part of a large team that carried out the primary survey of the Shetland Islands in the early 1930s. Previous publications on Shetland geology had concentrated mainly upon mineral occurrences (e.g. Heddle, 1878, 1901) and the only published maps were based upon that of Hibbert in 1822. The Dalradian rocks were surveyed by D. Haldane, J. Knox, J. Phemister, H.H. Read, T. Robertson and G.V. Wilson but the only ensuing publications were those of Read, which concentrated upon the metamorphism (Read, 1933, 1934, 1936, 1937) and the Caledonian ophiolite-complex of Unst and Fetlar (see the Caledonian Igneous rocks of Great Britain GCR volume; Stephenson et al., 1999).

Geological Survey work in Scotland between the wars and during the Second World War (1939–45) was concentrated in the coalfields, the Northern Highlands and the Inner Hebrides. Some sheet memoirs based upon pre-1914 mapping of the Grampian Highlands were published and the first edition of British Regional Geology: the Grampian Highlands appeared in 1935. The latter was written by H.H. Read who, by the time of its publication, had taken the chair at Liverpool University, where his influence no doubt laid the first foundations of the Dalradian research that was to flourish there in the second half of the 20th century (see below). Other than those by E.B. Bailey, there were few ‘Dalradian’ papers by survey geologists. E.M. Anderson did small amounts of official survey mapping in the Grampian Highlands and, inspired by Bailey, he produced a paper based on his own work in the Schiehallion area that significantly advanced understanding of the overall Dalradian succession (Anderson, 1923). He retired due to ill health in 1928, but continued ‘indoor work’ for many years and is best remembered for ‘The Dynamics of Faulting and Dyke Formation’, drawing upon his experience in the Grampians and elsewhere. It was first published in 1942, revised in 1951, remained a classic textbook for many years and was last reprinted in 1972.

The most prolific university contributions to the Dalradian between the wars were undoubtedly by Glasgow, where Bailey clearly inspired both existing staff and students. His predecessor, J.W. Gregory, had published several accounts (Gregory, 1910, 1916, 1928, 1929, 1930) and the first overview of Dalradian Geology (Gregory, 1931), all of which presented an ‘alternative’ rationalization of the succession across the whole region. W.J. McCallien (1925, 1926, 1929) had already published on the Dalradian rocks of Kintyre and went on to collaborate with Bailey in Perthshire (Bailey and McCallien, 1937). A. Allison worked on the Tayvallich–Loch Awe area and it was in the discussion to his paper, in 1941, that Bailey had his final say on the gross structure of the South-west Grampian Highlands, after becoming Director of the Geological Survey. Among Bailey’s students, S.M.K. Henderson (1938) was the first to identify the Aberfoyle Anticline, but it was J.G.C. Anderson who made a major contribution to the Dalradian of the Highland Border region, in addition to his PhD studies on igneous rocks of the western Grampian Highlands. He joined the Geological Survey in Scotland in 1937, just before its activities became focussed upon war-related resource evaluation, and hence he did little ‘official’ mapping in the Highlands. However, in his own time he covered large areas in the manner of his mentor, resulting in a series of papers between 1935 and 1956, the last being published after he was appointed to the chair in Cardiff in 1949.

Work on the Dalradian at this time at Cambridge University centred around C.E. Tilley and Gertrude Elles. This included the recognition of the Portsoy Thrust in the North-east Grampian Highlands (Elles, 1931) and the refining of Barrow’s zones of regional metamorphism (Tilley, 1925). Studies of the metamorphism of basic igneous rocks in the Dalradian were also made in Cambridge (Phillips, 1930; Wiseman, 1934), complementing what was already known about the progressive metamorphism of pelitic rocks. Also of the Cambridge ‘school’ was J.F.N. Green, who was a student contemporary of Gertie Elles before embarking upon a non-geological career in the Colonial Office, pursuing his geological interests in his spare time and becoming president of both the Geologists’ Association and the Geological Society. After acclaimed work in Pembrokeshire and the Lake District, he made a major contribution to the geology of Islay, though a later paper on broader aspects of the Dalradian of the South-west Grampian Highlands was criticised by both Bailey and Elles (Green, 1924, 1931). His interest in the Dalradian probably resulted from an association with Barrow, after the latter’s move to the Geological Survey in England. Less well known is the PhD study of S.O. Agrell (1942) under the supervision of Frank Coles Phillips, an unpublished part of which involved the first petrofabric analysis of the Ben Vuirich Granite and its country rocks. They could not have known how important these fabrics would become in subsequent discussions on the timing of pre-Caledonian and Caledonian magmatic and tectonic events, but the conclusions reached accord remarkably well with current interpretations (Howarth and Leake, 2002).

By 1950, Bailey had retired from the Geological Survey but was still a considerable influence on many aspects of Highland geology. Read had moved to the chair at Imperial College, London, where he continued to publish on the North-east Grampian Highlands. He led many investigations into the Caledonian basic intrusions of the region and introduced the concept of the Banff Nappe as a largely allochthonous succession (Read, 1955). The latter view was however disputed by his colleagues John Sutton and Janet Watson, who made a brief diversion from their work on the Lewisian of the North-west Highlands to publish three papers on the crucial across-strike Dalradian section of the Banffshire coast (Sutton and Watson, 1954, 1955, 1956). Read became the authority of his day on the origin of granites, deployed many post-graduate students and researchers to all parts of the Scottish Highlands, and developed what was to become a major involvement in Dalradian and Caledonian igneous rocks of the west of Ireland. But undoubtedly one of his most momentous moves was to support Derek Flinn’s studies of the Dalradian (and other rocks) of Shetland, which were initiated as a PhD under Read (Flinn, 1953) and continued for over 50 years (e.g. Flinn, 2007). Subsequently at Imperial College, John Knill combined structural geology in the Craignish–Kilmelfort area with some of the earliest sedimentological studies of deformed and metamorphosed rocks (Knill, 1959, 1960, 1963) and Brian Amos (1960) and Brian Hackman (Stewart and Hackman, 1973) applied sedimentology to the Bowmore Sandstone and Colonsay groups of Islay respectively.

In 1948, the chair in Liverpool passed to Robert Shackleton, who had developed an interest in the Dalradian of Connemara and Donegal whilst at Imperial College. In Liverpool, he also turned his attention to the Grampian Highlands, in particular the Highland Border region, where he worked with research students (Stone, 1957; Stringer, 1957). This led to the conclusion that the Aberfoyle Anticline, and hence the closure of the Tay Nappe, is downward facing in the Highland Border region, and stimulated a debate on the geometry and mechanism of emplacement of the nappe that continues to the present day. The resulting seminal publication is frequently cited as the first explanation of the concept of structural facing in polyphase terrains to facilitate understanding of the structural evolution of an area (Shackleton, 1958). Shackleton’s tenure also saw the move to Liverpool of Derek Flinn, the PhD study of the Colonsay Group on Colonsay and Oronsay by A.D. (Sandy) Stewart (1960) and the start of John Roberts’ investigations in the South-west Grampian Highlands (Roberts, 1963).

Meanwhile, in Glasgow, research on the Dalradian was continuing under the direction of Basil King, and later Don Bowes, notable students being Nick Rast around Schiehallion (PhD, 1956), Donald Ramsay in Glen Lyon (PhD, 1959), Ken Jones in the Ben More–Stob Binnein area (PhD Swansea, 1959) and Harry Convery in the Ben Ledi–Balqhidder area. Rast became a key foundation in a ‘dynasty’ that dominated university Dalradian research for the next forty years. He first took up a post in Aberystwyth, where he extended his own work eastwards across the Loch Tay Fault by supervising the research of Brian Sturt, south of Loch Tummel (1959), and Tony Harris, between Loch Tummel and Blair Atholl (1960). When Shackleton moved to Leeds in 1962, Rast took over most of the Dalradian research in Liverpool, supervising Jack Treagus south of Loch Rannoch (1964a) and Peter Thomas to the north in what was later to become known as the Geal-charn–Ossian Steep Belt (1965). Those studies were two of the earliest to concentrate on rocks below and around the Boundary Slide. He also supervised Martin Litherland around Glen Creran (1970), A.N. Basahel on Islay (1971), Duncan France around the Boundary Slide at Bridge of Orchy (1971) and Graham Borradaile in the northern Loch Awe Syncline (1972a). It was during this era that Rast and Litherland (1970) made the first attempt to correlate the Dalradian successions of the South-west Grampians and Lochaber with those of the Central Grampians, which ultimately fed into the first correlations across the whole Grampian Terrane by Harris and Pitcher (1975). Wallace Pitcher, Shackleton’s successor at Liverpool, whilst mostly pursuing his Caledonian interests in Donegal, was promoting an appraisal of Dalradian tillites that culminated in the comprehensive Geological Society memoir by Tony Spencer (1971). Numerous metamorphic aspects of the Dalradian of Scotland were also investigated by Mike Atherton, Mike Brotherton and John Mather (PhD, 1968), leading to a major review article (Atherton, 1977), and the metamorphic history of north-east Shetland was studied by Roger Key (PhD, 1972).

Rast left Liverpool in 1971 and was replaced by Tony Harris, who by then had worked on Dalradian rocks for the Geological Survey in the Elgin area and in the Highland Border, around Dunkeld. Harris took up several lines of Dalradian research, involving many research students, notably Peter Gower on the Loch Tay Limestone (1973), Harry Bradbury (1978) and Richard Smith (1980) on the Pitlochry–Blair Atholl district, Lindsay Parson around Fort Augustus (1982), Andrew Highton (1983) and Nick Lindsay (1988) on the problems surrounding the structural history and stratigraphical correlation of the Grampian Group and its migmatitic ‘basement’, and Phillip Rose on the emplacement and evolution of the Tay Nappe (1989). Harris, Bradbury and Smith were all involved in a contract with the Geological Survey to map Sheet 55E (Pitlochry, 1981). At the same time, Jack Treagus was developing his own research in Manchester, mainly in the Central Grampian Highlands and including collaborative work with John Roberts in Newcastle on the Loch Leven area and on the Banffshire coast. He directed research students in the Braemar area (Paul Upton, 1983), Glen Lyon (Phillip Nell, 1984) and the South-west Grampians (Charlie Bendall, 1995), and collaborated with Peter Thomas, by this time at the University of Strathclyde, on a new edition of Sheet 55W (Schiehallion, 2000), with accompanying memoir, for the Geological Survey.

In Cambridge, Tilley’s influence upon metamorphic studies continued into the 1950s, introducing Henry Pantin to the basic meta-igneous rocks of Ben Vrackie (1952) and encouraging Graham Chinner to re-examine the pelitic gneisses of Glen Clova (1957). Chinner then went on to direct numerous studies, in particular on high-grade migmatitic rocks (e.g. John Ashworth, 1972; Eileen McLellan, 1983). He was also responsible for introducing Ben Harte to the south-east Grampians (1966), and Keith Watkins to the Balquhidder–Crianlarich area (1982). Ashworth and Harte both went on to expand their studies elsewhere.

Studies of contemporaneous igneous rocks were also taking place at Bristol University under Bernard Leake, whose main interest lay in Connemara but who supervised Peter van de Kamp on the Green Beds (1968) and collaborated with J.R. Wilson on the ‘epidiorites’ of Tayvallich (Wilson and Leake, 1972). After moving to take up the chair in Glasgow, Leake wrote a review of volcanism in the Dalradian (Leake, 1982). Meanwhile, in Edinburgh (see below), Colin Graham had started a major study of volcanic and subvolcanic basic meta-igneous rocks in the South-west Grampians (Graham, 1976) and had also contributed to a review (Graham and Bradbury, 1981). Several more-recent studies of Dalradian volcanic and volcaniclastic rocks and of magmatism in general have been linked in some way to BGS mapping projects (Goodman and Winchester, 1993; Macdonald et al., 2005; Pickett et al., 2006; Macdonald and Fettes, 2007; Fettes et al., 2011) and David Chew of Trinity College, Dublin has incorporated the geochemistry of metavolcanic rocks into wider studies of Dalradian tectonics (e.g. Chew et al., 2009)..

In Edinburgh, Mike Johnson re-examined the structure and metamorphism of the Banffshire Dalradian sections, with the help of PhD studies by Vic Loudon (1963) and Doug Fettes (1968). He also wrote reviews of the Dalradian for the first three editions of Geology of Scotland (1965, 1983, 1991). Ben Harte expanded his interests, notably through his own work in the ‘Tarffside Nappe’ and other areas close to the Highland Border and through the PhD work of John Booth (1984) and Tim Dempster (1983). The latter’s work on metamorphism of the Dalradian and particularly on its uplift history has been continued at Glasgow. Harte also supervised the study by Neil Hudson on Buchan-type metamorphism of pelites (1976), which led on to studies at Derbyshire College, later University of Derby, of calcsilicate metamorphism by Stuart Kearns (1989), pelitic migmatites by Tim Johnson (1999) and the Portsoy Shear-zone by Jim Carty (2001). Colin Graham started his study of basic meta-igneous rocks with a PhD in 1973 and subsequently expanded into studies of low- to intermediate-grade metamorphism and fluid movement in the South-west Grampians that have included PhD students Ken Greig (1987), Peter Dymoke (1989), Alasdair Skelton (1993) and Chris Thomas (1999).

Following the earlier studies at Liverpool and Imperial College, sedimentological understanding of the lower grade Dalradian rocks of the South-west Grampian Highlands was significantly advanced by the work of Roger Anderton, first as a PhD in Reading (1974), then at the University of Strathclyde, and culminating in a seminal review paper (Anderton, 1985). Subsequent notable sedimentological PhD studies have been those of Brian Glover (1989) and Chris Banks (2005), both on the Grampian Group of the Northern Grampians at the University of Keele, and Elaine Burt (2002) on the Southern Highland Group at Kingston University. At Keele, John Winchester’s interest in the Dalradian had originally been through whole-rock geochemistry but broader structural and sedimentological aspects of the Grampian Group in particular have been pursued through a number of research students in the Northern Grampian area, Keith Whittles (1981), Paula Haselock (1982) and C.T. Okonkwo (1985). Subsequently, whilst at the University of Greenwich, Haselock collaborated with colleagues and the BGS to produce Sheet 73E (Foyers, 1996).

Much of the Dalradian research at Aberdeen University in the latter part of the 20th century centred around the Caledonian basic intrusions of the North-east Grampians and more-general mapping contracts with the BGS. The latter were initially set up by Iain Munro and Bill Ashcroft, with other staff such as Graham Leslie, Alan Crane, Sally Goodman and Maarten Krabbendam becoming involved subsequently, together with PhD students such as Ben Kneller (1988). The first map and memoir to be produced in this way was Sheet 77 (Aberdeen, 1982), and this was followed by 87W (Ellon, 1985), 65E (Ballater, 1995) and 56W (Glen Shee, 1999).

At Birmingham, Alan Wright worked particularly on the stratigraphy and whole-rock geochemistry of the Appin Group (Wright, 1988) and supervised A.H. Hickman, who produced a radical re-appraisal of the stratigraphy and structure of the crucial area between Glen Roy and Lismore (Hickman, 1975, 1978). Also at Birmingham, Ian Fairchild, following on from his PhD studies at Nottingham, made detailed mineralogical and geochemical studies of the dolomitic beds at the base of the Argyll Group on Islay and both he and Wright contributed to the compilation of Sheet 27 (North Islay, 1994) for the BGS. Meanwhile Bill Fitches, Alex Maltman and Roddy Muir in Aberystwyth were concentrating upon the Colonsay Group in south-west Islay and its relationship with the basement of Rhinns Complex, whilst M.R. Bentley (PhD, 1986) re-appraised the outcrops on Colonsay.

Later work at Glasgow University has been largely by Geoff Tanner and Tim Dempster, with significant discussion and counter argument on broader tectonic themes by Brian Bluck. Tanner turned to addressing several key issues in the Scottish Dalradian after a long involvement with the Dalradian of Connemara that started with Shackleton at Leeds. In a series of seminal papers since 1994, he has established the emplacement age of the Ben Vuirich Granite relative to the Grampian deformation and metamorphism (Tanner and Leslie, 1994; Tanner, 1995; Tanner et al., 2006) and has confirmed the structural and stratigraphical continuity between the Southern Highland Group and the older parts of the Highland Border Complex (Tanner, 1995; Tanner and Pringle, 1999; Tanner and Sutherland, 2007). A collaboration with Peter Thomas has led to a classic detailed description of recumbent folds between Tyndrum and Glen Orchy (Tanner and Thomas, 2010) and structural investigations in the Highland Border area around Loch Lomond, Cowal and Bute are contributing to understanding the geometry and development of the Tay Nappe.

During the mid 20th century, Geological Survey work in the Highlands was almost entirely north-west of the Great Glen. The only Dalradian mapped was on the fringes of the Grampian Highlands on sheets 95 (Elgin), 39 (Stirling) and 48W (Perth). However, considerable data on the Dalradian of the Central Grampian Highlands was obtained as a result of logging tunnels and other excavations for new hydro-electric schemes, mostly by Scot Johnstone and Donald Smith. Sheet 38W (Ben Lomond) and parts of surrounding areas were revised in the early 1980s by Doug Fettes, John Mendum, and Bill Henderson, and then in 1982 the East Grampians Project set out to complete the revision mapping of the North-east Grampian Highlands. Key players were Doug Fettes, John Mendum, David Stephenson, David Gould, Chris Thomas, Graham Smith and Steve Robertson, aided by specialists in geophysical, geochemical and mineral surveys and by several university contract teams (see above). Some nineteen 1:50 000 sheets were produced over a twenty-year period. This was followed by the Monadhliath Project in 1986, staffed mainly by Frank May, Roger Key, Colin Clark, Martin Smith, Steve Robertson, Andrew Highton, Donald Smith and Richard Smith. This project aimed to complete the mapping of the Grampian Group and its ‘basement’ (i.e. the Dava and Glen Banchor subgroups = Badenoch Group) in the Northern Grampian Highlands, a significant part of which was still classed as ‘primary survey’. The East Grampian Project completed the lithostratigraphical rationalisation and correlation of the Appin and Argyll groups between the established sequences in the Central Grampian Highlands and on the north coast and recognized several important shear-zones and long-lived crustal lineaments (Fettes et al., 1986, 1991). The whole-rock geochemistry of metacarbonate rocks proved to be a useful indicator of depositional environment and a correlation tool (Thomas, 1989) and pelitic rocks revealed a fascinating story of high-pressure metamorphic overprinting west of the Portsoy Lineament (Beddoe-Stephens, 1990). The most significant results to emerge from the Monadhliath Project have been the establishment of a coherent lithostratigraphy for the Grampian Group, linked to early Dalradian basin architecture, and the confirmation of basement–cover relationships in the Northern Grampians (e.g. Robertson and Smith, 1999; Smith et al., 1999).

Other work in the Grampian Highlands by the BGS that has provided invaluable background information to so many studies includes the regional gravity and aeromagnetic surveys (published in 1977 and 1978 respectively) and the regional stream-sediment surveys that formed the basis for geochemical atlases (Great Glen, 1987; Argyll, 1990; East Grampians area, 1992; Southern Scotland, 1993). Work for the Mineral Reconnaissance Programme, led mainly by Mike Gallagher, Graham Smith and Stan Coats was spectacularly successful in its identification of the stratabound barium materialization near Aberfeldy (e.g. Coats et al., 1984), and that provided the impetus for investigations in other areas such as Tyndrum, Duntanlich, Coire Kander and The Lecht. Apart from the baryte at Aberfeldy and Duntanlich and a gold prospect at Cononish, economic prospects have been disappointing but the investigations have contributed to the mapping programme and inspired a wealth of scientific spin-off studies.

More-recent mapping by the BGS has been on Sheet 38E (Aberfoyle, 2005) and work is ongoing on sheets 46W and 46E in the Bridge of Orchy–Killin area (mainly by Graham Leslie, Maarten Krabbendam, Richard Smith and Chris Thomas). Other recent work includes the dating of detrital zircons and Sm-Nd signatures of whole-rock as indicators of sediment provenance (e.g. Cawood et al., 2003), the search for orogenic unconformities, especially within the Argyll Group, that might correlate with possible examples in the west of Ireland (e.g. Dempster et al., 2002; Hutton and Alsop, 2004), and the ongoing modelling of the Tay Nappe and regional deformation mechanisms (Krabbendam et al., 1997). The identification of various glacial deposits within the sequence and their correlation with known global glaciations has led to much speculation (e.g. McCay et al., 2006) and this has been linked to global C-isotope profiles, largely by Tony Prave at St Andrews (Prave et al., 2009a, 2009b).

Radiometric dating of events that have affected Dalradian rocks has been carried out at the Scottish Universities Environmental Research Centre (SUERC, formerly SURRC) and at the Natural Environment Research Council’s Isotope Geology Laboratories (NIGL). At SUERC, Otto van Breemen, in collaboration with Mark Piasecki at Hull, produced the first crucial dates on pegmatites emplaced in slides that cut what they regarded as pre-Grampian Group basement (Piasecki and van Breemen, 1979a,b, 1983). Subsequent fieldwork has confirmed that relationship and more-modern high-precision zircon and monazite dates have been obtained by Steve Noble at NIGL, in collaboration with Euan Hyslop and Andrew Highton at the BGS (Noble et al., 1996; Highton et al., 1999). Other key dates, notably for the emplacement of the the pre-metamorphic Ben Vuirich Granite, and the North-east Grampian Basic Suite that was almost coeval with the peak of Grampian deformation and metamorphism, were produced initially under Robert Pankhurst at the forerunner to NIGL (Pankhurst, 1970; Pankhurst and Pidgeon, 1976). These have been repeated, together with dates of the Tayvallich volcanic rocks and mineral ages that plot late-tectonic cooling and uplift, using high-precision methods at SUERC and elsewhere. The key workers have been Alex Halliday, Graeme Rogers and Tim Dempster, in collaboration with projects at Glasgow and Edinburgh, who are responsible for virtually the whole temporal framework currently in use (Halliday et al., 1989; Rogers et al., 1989; Dempster et al., 1995, 2002). The possibility of more-direct dating of Dalradian sedimentation has been raised by encouraging results of Re–`Os dating at Durham University (Rooney et al., 2011).

Field excursion guides to the Dalradian have been published by Read (1960) and Treagus (2009). Individual excursions are described in several regional guides i.e. Arran (MacDonald and Herriot, 1983), Aberdeen area (Trewin et al., 1987), Glasgow and Girvan (Lawson and Weedon, 1992), Fife and Angus (MacGregor, 1996). Excursions in the Grampian Group and its inferred basement, all formerly regarded as Moine, are included in the original excursion guide to Moine geology (Allison et al., 1988) and a series of excursions to the Dalradian of the South-west Grampian Highlands, described by various authors, comprise a whole part issue of the Scottish Journal of Geology (volume 13, part 2, 1977).

1.2 GCR site selection

D. Stephenson

Metamorphic rocks in the more-sparsely populated areas of Great Britain are on the whole less prone to damage than sequences in the more-developed areas. They are none the less vulnerable to large-scale activities, some long established and obvious, such as quarrying and landfill, and others related to more-recent exploitation of the rural landscape such as coastal defences, hydro-electric schemes, wind farms and power transmission lines.

The greatest threat is undoubtedly the possibility of large areas of rock being obscured by man-made constructions or large volumes being removed by excavations. Some of the harder and more-resistant lithologies are an important source of construction materials and are hence particularly vulnerable to large-scale commercial extraction. Uses are many and varied but as demand changes with time, new uses are constantly emerging, so that no rock can be considered absolutely safe from future exploitation. However, with careful management both disused and active quarries, road cuttings and other excavations can provide highly instructive exposures, especially in areas of poor natural exposure. On a smaller scale, fossils, minerals and fine detail of delicate features can be lost easily through injudicious hammering, for casual or commercial collecting or even for bona fide research purposes. Much of the value of the GCR sites is derived from their research potential, but sampling does need to be controlled carefully and there is a clear need for better dissemination of information about protected sites.

The GCR sites in this volume vary greatly in size and character. There are long coastal sections (e.g. Kilchiaran to Ardnave Point, Rubha a’Mhail, West Tayvallich Peninsula, Cullen to Troup Head, Fraserburgh to Rosehearty) as well as numerous smaller coastal exposures such as those on the Isles of Islay and Jura and on the mainland of the South-west Grampian Highlands. River sections dominate the inland areas (e.g. River Leven Section, River Orchy, River E, Glen Ey Gorge, Bridge of Brown, Bridge of Avon, Kymah Burn) but road cuttings provide valuable additional sites (A9 and River Garry, The Slochd). At many of these sites exposure approaches 100 percent. Some of the larger sites occupy upland areas and mountain summits (Stob Ban, Ben Lawers, Aonach Beag and Geal-charn, Ben Alder, Ben Vuirich, Cairn Leuchan), where large-scale structures can be demonstrated, commonly in three dimensions, or where key exposures protrude from otherwise poorly exposed terrain.

The Geological Conservation Review (GCR) aims to identify the most important sites in order that the scientific case for their protection and conservation is fully documented as a public record, with the ultimate aim of formal notification as a Site of Special Scientific Interest (SSSI). The notification of SSSIs under the National Parks and Access to the Countryside Act 1949 and subsequently under the Wildlife and Countryside Act 1981, is the main mechanism of legal protection in Great Britain, and in Scotland this is the responsibility of Scottish Natural Heritage. At the time of writing most, but not all, of the sites described in this volume have been notified. The origins, aims and operation of the review, together with comments on the law and practical considerations of earth-science conservation, are explained fully in Volume 1 of the GCR series, An Introduction to the Geological Conservation Review (Ellis et al., 1996). The GCR has identified three fundamental site-selection criteria; these are international importance, presence of exceptional features andrepresentativeness. Each site must satisfy at least one of these criteria, many of them satisfy two and some fall into all three categories (Table 1.1), such as the Garvellach Isles site that displays a tillite of international importance for its detailed features as well as for its stratigraphical and chronological significance. In addition to the GCR sites, significant, well-exposed local successions and structural features may be designated as ‘Regionally Important Geological/Geomorphological Sites’ (RIGS) so that, even though such status carries no legal protection, their importance is at least recognized and recorded.

Features, events and processes that are fundamental to the understanding of the geological history, composition and structure of Britain are arranged for GCR purposes into subject ‘blocks’. The Dalradian rocks of Scotland comprise a single block. Within each block, sites fall into natural groupings, termed ‘networks’, which in this volume are based upon geographical areas (Figure 1.5). The boundaries between the areas follow significant stratigraphical or structural boundaries wherever possible and the Highland Border Region and North-east Grampian Highlands are well defined in this way. Boundaries between the South-west, Central and Northern Grampian Highlands are more arbitrary and to some extent reflect the areas of interest of particular groups of researchers or periods of research. The six networks, each represented by a single chapter, contain 85 sites, which are listed in Table 1.1 together with their principal reasons for selection.

Site selection is inevitably subjective and some readers may feel that vital features or occurrences have been omitted or that others are over-represented. But the declared aim of the GCR is to identify the minimum number and area of sites needed to demonstrate the current understanding of the diversity and range of features within each block or network. To identify too many sites would not only make the whole exercise unwieldy and devalue the importance of the exceptional sites, but it would also make justification and defence of the legal protection afforded to those sites more difficult to maintain.

Although this volume is titled Dalradian rocks, the GCR sites that it contains illustrate not only the lithostratigraphy and sedimentology of the Dalradian rocks and their basement but also the structures and metamorphism that affect them as a result of the Caledonian Orogeny. Some sites have been selected specifically to illustrate just one of those aspects, but many illustrate two or more. A few have historical significance.

The overall international importance of the Dalradian Supergroup has already been discussed in a historical context, highlighting significant contributions to the understanding of the processes of deformation and metamorphism. These contributions are acknowledged in several GCR sites such as those at Glen Esk and along the north coast (e.g. Fraserburgh to Rosehearty), where metamorphic zones at different confining pressures were first established. The sites that illustrate Neoproterozoic glacial espisodes, most notably at the Garvellach Isles and Caol Isla continue to attract attention for the worldwide significance of their deposits, the ages of which are still controversial. The granite at the Ben Vuirich site has yielded valuable information on the timing of deformation and metamorphism during the Caledonian Orogeny and unique economic deposits of stratabound barium minerals represented by the Craig an Chanaich to Frenich Burn site are of undoubted international value.

Exceptional features, invaluable for research and/or teaching purposes, are exhibited at many, if not most, of the GCR sites in this volume. For example, a range of sedimentary structures are particularly well displayed at the A9 and River Garry and River E sites, and many sites have first-rate examples of specific features such as dewatering structures (Caol Isla), slump structures (Lussa Bay, Kinuachdrach, Port Selma), sandstone dykes (Surnaig Farm), debris flows (Black Mill Bay) and Bouma sequences (Rubha na Magach). World-class examples of tillites occur at the Garvellach Isles site, algal stromatolites at Rubha a’Mhail, and pseudomorphs after gypsum at Craignish Point. Many of the sites show minor fold structures but these are particular features at Black Mill Bay, Fearnach Bay and Port Cill Maluaig. Fold interference structures are seen at the Creag nan Caisean–Meall Reamhar site and classic variations in style of cleavage occur in the related sites at Little Glen Shee, Craig a’Barns and Rotmell. Various types of migmatite are splendidly exposed at the Balnacraig, Dinnet and Cairnbulg to St Combs sites. Pillow lavas are present at several of the sites that include metavolcanic rocks (West Tayvallich Peninsula, Muckle Fergie Burn, Black Water) but the most exceptional meta-igneous features are probably the enigmatic spinifex-like textures at the Cunningsburgh site.

The criterion of representativeness aims to ensure that all key stratigraphical units and the most significant structural features are represented. It is difficult to do this whilst keeping the number of sites within reason. However, all of the main stratigraphical features of the Dalradian succession (e.g. those shown on (Figure 1.6)) are represented by sites, as well as some near-contemporaneous igneous intrusions. Selected structures include major and minor folds from each of the main phases of deformation, principal dislocations (thrusts, slides and faults), shear-zones and high-strain zones. Barrovian and Buchan metamorphism are represented as well as the highest grade rocks in the Dalradian and examples of polyphase metamorphism.

Clearly it would be impossible to represent all along-strike regional variations of the stratigraphy, structure and metamorphism by GCR sites. However, an attempt has been made to include broad descriptions of such variations in appropriate chapter introductions, together with references to key publications. Hence, this volume does constitute a complete review of the Dalradian Supergroup and the deformational and metamorphic effects of the Caledonian Orogeny throughout the whole Grampian Terrane.

Metavolcanic rocks within the stratigraphical column provide time markers and hence are important targets for radiometric dating. Precise U-Pb zircon dates have been obtained from the Tayvallich Volcanic Formation at the top of the Argyll Group, which give much added value to the West Tayvallich Peninsula GCR site. Unfortunately, as yet there have been no suitable targets that would date the tillites at the base of the Argyll Group (e.g. Garvellach Isles GCR site) that constitute the other significant chronostratigraphical marker in the Dalradian succession. Intrusions are less-precise chronostratigraphical markers than lavas but the deformed granite represented by the Ben Vuirich GCR site has yielded a number of increasingly precise radiometric dates that, together with dates from the Portsoy Granite (Cullen to Troup Head GCR site), have set time limits to Argyll Group deposition and early phases of the Caledonian Orogeny. The Tayvallich and Ben Vuirich dates are at present two of the mainstays of late-Neoproterozoic and Caledonian chronology in the whole North Atlantic region. Key radiometric evidence for the age of the Badenoch Group basement rocks has come from metasedimentary rocks at The Slochd GCR site.

Some sites are important in more than just the context of Dalradian rocks, their deformation and metamorphism. In particular, the Ardsheal Peninsula is also a GCR site in the Silurian and Devonian Plutonic Rocks block, as the type-area for the Appinite Suite of small ultramafic to felsic intrusions (Stephenson et al., 1999). The Keltie Water GCR site has profound implications for relationships between the Dalradian and the Highland Border Complex and needs to be considered along with two sites (Leny Quarry and Lime Craig Quarry), which are described in Volume 18 of the GCR series, British Cambrian to Ordovician Stratigraphy (Rushton et al., 1999).

The broader aspects of Dalradian regional geology are outlined in the various sections of this chapter, with variations applicable to each network described in the following chapter introductions. In some cases, sections of general discussion apply to two or more related sites and this has necessitated a slight change of format in the Little Glen Shee, Craig a’Barns and Rotmell site reports. The voluminous and widespread Caledonian igneous rocks that are emplaced within and upon the Dalradian rocks are described in the Caledonian Igneous Rocks of Great Britain GCR volume(Stephenson et al., 1999). For details of post-Dalradian rocks in the Grampian Highlands the reader is referred to The Geology of Scotland (Trewin, 2002) and the volume in the British Geological Survey’s British Regional Geology series (Stephenson and Gould, 1995).

Table 1.1 Dalradian rocks of Scotland: GCR networks and site selection criteria.

1.3 Basement to the Dalradian basins

D. Stephenson

Our knowledge of the immediate basement to the Dalradian rocks of the Grampian Highlands is limited to outcrops on the islands of Islay and Colonsay in the south-west (the Rhinns Complex) and in parts of the Northern Grampian Highlands (the Badenoch Group, formerly known informally as the Dava and Glen Banchor successions) (Figure 1.1). The existence of an unmodified contact with the basement is difficult to demonstrate in either area, but a stratigraphical and orogenic unconformity can be inferred from omission and overstep of strata on a regional scale and from structural and metamorphic hiatuses. Where contacts are exposed they are commonly seen to coincide with zones of high strain and shearing.

The Rhinns Complex was once thought to be part of the Lewisian Gneiss Complex of the Hebridean Terrane. It is now regarded as part of an extensive tract of Palaeoproterozoic juvenile crust that includes the Ketilidian belt of southern Greenland and the Svecofennian belt of Scandinavia and may form a link between those two segments (Marcantonio et al., 1988; Muir et al., 1989, 1992; Park, 1994). It extends south-westwards at least to the island of Inishtrahull off the northern coast of Ireland and Bentley et al. (1988) proposed that the three outcrops define a small allochthonous Colonsay–western Islay Terrane, bounded by the Great Glen and Loch Gruinart faults. Subsequently, it has been suggested that it might also underlie much of the Grampian Highlands, where the isotopic signatures of Caledonian granites are consistent with derivation, at least in part, from comparable juvenile crust forming a block measuring at least 600 x 100km to the south-east of the Great Glen Fault (Dickin and Bowes, 1991).

Regional gravity and magnetic evidence indicate the presence of a distinctively different, lower density, basement beneath the south-eastern part of the Grampian Highlands (i.e. beneath the Tay Nappe), which seems to be a continuation of the basement that underlies the Midland Valley of Scotland (Rollin, 1994; Trewin and Rollin, 2002). The geophysical evidence also indicates the existence of a different, high-density, basement beneath the Buchan Block of the North-east Grampian Highlands.

The Dava and Glen Banchor subgroups of the Badenoch Group in the Northern Grampian Highlands are somewhat younger sequences of mainly gneissose and locally migmatitic metasedimentary rocks, comparable in lithology to parts of the Moine Supergroup to the north-west of the Great Glen Fault, and showing evidence of having experienced at least some elements of a Neoproterozoic, Knoydartian orogenic event. Gneissose metasedimentary units in the Buchan Block have also been interpreted as part of a Proterozoic ‘basement’ to the Dalradian by some authors (Sturt et al., 1977; Ramsay and Sturt, 1979), although this is not currently accepted (see the Introduction to Chapter 6).

1.3.1 Rhinns Complex

The Rhinns Complex crops out over an area of about 20km² on the Rhinns of western Islay and as a very small inlier at the north end of Colonsay (Muir, 1990; Muir et al., 1992, 1994a, 1994b) (see (Figure 2.1)). On Islay, granitic and syenitic gneisses were all affected by deformation and amphibolite-facies metamorphism prior to the intrusion of gabbro sheets and further intense multiple deformation (Wilkinson, 1907). The inlier on Colonsay covers only c. 0.3 km² and is largely obscured by blown sand (Cunningham Craig et al., 1911). The exposures there are of quartzofeldspathic gneiss, much of it pegmatitic, with dark knots, streaks and layers of amphibolite. The metasyenites and metagabbros represent an alkaline igneous association and are characterized by major- and trace-element patterns similar to subduction-related igneous rocks generated in continental margins or island-arcs (Muir et al., 1992, 1994a). Isotope studies have shown that they were emplaced as magmas consisting dominantly of juvenile material derived from a depleted mantle source.

In places the gneisses have suffered considerable crushing, mylonitization and metamorphic downgrading and have been intersliced with the overlying low-grade metasedimentary rocks of the Colonsay Group (Muir et al., 1995). The intensity of the cataclastic and mylonitic effects increases as the overlying sequence is approached, but the actual contact is rarely seen. Where it is exposed, it is marked by a high-strain zone of phyllonitization and mylonitization. On Islay this is termed the Kilchiaran Shear-zone or Bruichladdich Slide.

Palaeoproterozoic U-Pb zircon ages obtained from metasyenites on Islay (1782 ± 5 Ma; Marcantonio et al., 1988) and Inishtrahull (1779 ± 3 Ma; Daly et al., 1991) have been interpreted as crystallization ages of the protolith. That is about the time that the Lewisian Gneiss Complex of the Hebridean Terrane was undergoing tectonothermal reworking during the Laxfordian Event (Mendum et al., 2009). However, the stable isotope studies have shown that the Rhinns rocks are not reworked Archaean crust but are derived dominantly from juvenile mantle material, effectively ruling out any direct correlation with the Lewisian Gneiss Complex, in which there is no evidence for major addition of mantle material at c. 1800 Ma.

1.3.2 Badenoch Group

Large areas of middle amphibolite-facies, gneissose and locally migmatitic, psammite and semipelite in the Northern Grampian Highlands were originally regarded as stratigraphically equivalent to the surrounding psammitic rocks, and all were equated with the Moine succession north-west of the Great Glen Fault, being referred to generally as the ‘Younger Moines’ (e.g. Johnstone, 1975; Harris and Pitcher, 1975). Detailed mapping and radiometric age determinations in the 1970s led to a radical re-interpretation of the stratigraphical and structural significance of the migmatitic rocks. These became regarded as a distinct tectonostratigraphical unit, termed the ‘Central Highland Division’, and were viewed as an older basement to the adjacent ‘Grampian Division’ psammites (Piasecki and van Breemen, 1979a, 1979b; Piasecki, 1980; Piasecki and Temperley, 1988a). The traditional view that the Central Highland Division comprises migmatized versions of the adjacent rocks (though not by this time correlated with the Moine) was restated by Lindsay et al. (1989) and the matter remained unresolved for some time, with the generally gneissose parts of the succession regarded for convenience as a tectonothermal lithodemic unit termed the ‘Central Highland Migmatite Complex’ (Harris et al., 1994; Stephenson and Gould, 1995; Highton, 1999). Recent mapping by the British Geological Survey (BGS) has shown that the migmatization is controlled generally by the lithology of the protolith, but has confirmed the existence of a basement–cover relationship and has mapped out the contact. ‘Basement’ successions have been documented in two geographically separate areas (see (Figure 5.1)) and these have been referred to informally as the Dava and Glen Banchor successions in recent publications (Robertson and Smith, 1999; Smith et al., 1999; Strachan et al., 2002). They have now been formalised as subgroups of the Badenoch Group. Additional small inliers occur in the Aviemore area at Kincraig and Ord Ban and enigmatic gneissose rocks in Gleann Liath, near Foyers, described by Mould (1946), could be a fault slice of a similar basement succession.

A series of ductile shear-zones, known collectively as the Grampian Shear-zone (formerly the Grampian Slide or Slide-zone), can be traced discontinuously throughout the outcrop of the Glen Banchor Subgroup and largely delimit the Dava Subgroup. The shear-zones have been variously interpreted as a deformed unconformity between basement and cover (Piasecki, 1980), a zone of tectonic interleaving of the Grampian Group with a migmatite complex (Highton, 1992; Hyslop and Piasecki, 1999), and a zone of distributed shear located entirely within the Badenoch Group (Smith et al., 1999).

Many of the shear-zones incorporate a distinctive suite of syntectonic, foliated veins of pegmatitic granite and quartz, and it is believed that these segregations formed during early ductile shearing along the slides (Hyslop, 1992; Hyslop and Piasecki, 1999). Several dates were obtained by Rb-Sr methods on combined muscovite and whole rock, which cluster quite tightly around 750 Ma (Piasecki and van Breemen, 1979a, 1983). More-precise U-Pb analyses of monazites from such pegmatites have provided ages of 808 +11/–9 Ma and 806 ± 3 Ma, and a concordant age of 804 +13/–12 Ma has been obtained from the host mylonite matrix (Noble et al., 1996). U-Pb dating of single zircon grains within kyanite-bearing migmatites has yielded an age of 840 ± 11 Ma (Highton et al., 1999). These U-Pb dates provide a minimum age for the Badenoch Group and confirm that it was affected by a tectonothermal event, comparable in age to the Knoydartian that affected the Moine Supergroup of the Northern Highlands (Fettes et al., 1986). No such dates have been obtained from rocks assigned to, or cutting, the Grampian Group, which record only later, Caledonian events, supporting the notion of an orogenic unconformity at or near to its base. Thus the rocks of the Badenoch Group are now regarded as forming the basement to Grampian Group depositional basins, and may possibly be comparable in age to the Moine Supergroup.

A full discussion of the basement–cover relationships and the evidence for a stratigraphical, structural and metamorphic break is given in the Introduction to Chapter 5 of the present volume.

1.4 Dalradian lithostratigraphy

D. Stephenson

The Dalradian Supergroup is dominated by well-differentiated sequences of variably metamorphosed marine clastic sedimentary rocks and metacarbonate rocks. Some fluvial interludes are recognized in the earliest parts and localized metavolcanic rocks occur in the later parts. The aggregate total thickness of the succession adds up to at least 25 km, although the complete thickness was never deposited at one place. It is more likely that depocentres migrated south-eastwards with time and individual basins are unlikely to have accumulated more than 15 km of sediment.

The formal hierarchical lithostratigraphy of the supergroup is shown in (Figure 1.6). This builds on the original division of the Dalradian into three groups by Harris and Pitcher (1975), with subsequent incorporation of the Grampian Group by Harris et al. (1978), and further rationalization and modification by Winchester and Glover (1988), Harris et al. (1994), Stephenson and Gould (1995) and Smith et al. (1999). The Grampian, Appin and Argyll groups are divided into subgroups. A suggestion by Banks and Winchester (2004) that the lowest subgroup of the Grampian Group should be elevated to group status as the Glenshirra Group has not been accepted here.

The two lower groups (Grampian and Appin groups) fall within the Cryogenian System/Period and the two higher groups (Argyll and Southern Highland groups) fall mainly within the Ediacaran System/Period but extend into the Middle Cambrian Series. Outcrops of the four groups are shown on (Figure 1.1), which illustrates the overall younging from north-west to south-east. Subgroup outcrops are shown on the figures to acccompany the regional chapter introductions (Figure 2.1), 3.0, (Figure 4.1), (Figure 5.1), (Figure 6.1).

Similar sedimentary associations extend throughout the Grampian Highlands along a strike length of some 320 km. In fact in some parts of the succession, a remarkable stratigraphical consistency is preserved over the whole Dalradian outcrop from Connemara in the west of Ireland to the Banffshire coast, a distance of 700km. A simple ‘layer-cake’ stratigraphy, in some cases allowing detailed correlation down to at least member level, is most apparent in the middle part of the Appin Group. That sequence, from the calcareous top part of the Lochaber Subgroup, through the whole of the Ballachulish Subgroup, to the basal metamudstones and metalimestones of the Blair Atholl Subgroup, is recognizable almost everywhere. In other parts of the Dalradian Supergroup, correlation between local successions is complicated by lateral facies changes, diachronous boundaries, local unconformities, non-sequences, tectonic discontinuities and changes in metamorphic grade. Many formations have only local extent but laterally persistent facies associations do allow broad correlations at subgroup level over most of the outcrop.

Certain key formations of distinctive lithology have been traced throughout the Grampian Highlands and, in some cases, through north-western Ireland. Most of these are highlighted on (Figure 1.6). They provide lithostratigraphical markers and some have widespread chronostratigraphical significance. Foremost among the latter are the tillites (e.g. the Port Askaig Tillite Formation) that mark the base of the Argyll Group throughout most of its outcrop and record a major glacial event. Correlation with worldwide glaciations is currently a matter of debate but is taken here to be with the 635 Ma Marinoan global event (see p. 47 [in Dating the Dalradian sedimentation]). The top of the Argyll Group is marked in the South-west Grampian Highlands and parts of Northern Ireland by large volumes of basic volcanic and subvolcanic rocks (e.g. the Tayvallich Volcanic Formation). These represent a major rift-related magmatic event during the early stages of development of the Iapetus Ocean that has been dated radiometrically at 600–595 Ma (see p. 47 [in Dating the Dalradian sedimentation]). Other events, although readily recognizable in the lithostratigraphy, are probably more diachronous. These include transgressive flooding surfaces, such as those at the bases of the Ballachulish and Easdale subgroups, and basin shallowing events marked by calcareous lithologies at the tops of those two subgroups.

On the Shetland Islands, lithological associations are typical of the Dalradian as a whole. In particular, metalimestones and metamudstones in the middle of the succession suggest a broad correlation with the Appin and Argyll groups, and metavolcanic formations have been assumed to be broadly coeval with those at the top of the Argyll Group and in the Southern Highland Group. But, any detailed correlations with the Dalradian of mainland Scotland that have been suggested are extremely tenuous.

Correlations in parts of the succession are complicated by syn-depositional, possibly listric faulting (e.g. Anderton, 1985, 1988) and detailed field observations are now revealing previously undetected intrabasinal unconformities and periods of non-deposition. The scale and significance of such discontinuities is currently a matter of much debate, in some cases leading to speculation that major stratigraphical and orogenic breaks occur within the Dalradian successions of both Scotland and Ireland (Prave, 1999; Alsop et al., 2000; Hutton and Alsop, 2004), though this has been challenged by Tanner (2005).

The original depositional thicknesses of sedimentary units are difficult to estimate due to tectonic thickening or thinning during polyphase folding and ductile shearing. And regional metamorphism not only obliterates original sedimentary features but can also change the lithological and mineralogical characteristics to such an extent that some lithostratigraphically equivalent units have been given different formation names in areas of differing metamorphic grade. However, the intensity of metamorphism and deformation varies considerably on a regional basis. In the South-west Grampian Highlands and in most of the Highland Border region, the middle to upper parts of the Dalradian succession have been affected only by low- to medium-grade metamorphism and are not strongly deformed. Hence, their basin architecture, sedimentological environment and depositional processes have been deduced with some confidence (e.g. Roberts, 1966a; Hickman, 1975; Anderton, 1976, 1979; Litherland, 1980; Burt, 2002). As a consequence of this detailed knowledge, most early reviews of Dalradian evolution were heavily biased towards interpretations from those areas (e.g. Knill, 1963; Harris et al., 1978; Johnson, 1965, 1983, 1991; Anderton, 1982, 1985, 1988). In areas of higher grade metamorphism, interpretations of the original nature of the rocks are generally more difficult, although sedimentary structures are preserved to remarkably high grade in some psammitic rocks. These have enabled several sedimentological studies to be made, particularly in the Grampian Group (Winchester and Glover, 1988; Glover and Winchester, 1989; Glover, 1993; Banks, 2007) and sequence-stratigraphy has been applied to some extent in the lowermost parts (Glover and McKie, 1996; Banks, 2005).

For many years, the regional correlation of local formations within the Dalradian succession could not be attempted with any confidence and the first Grampian-wide lithostratigraphical scheme was that of Harris and Pitcher (1975). Since then most gaps in the mapping have been filled, correlations have been widened to at least subgroup level and most units have been formalized by inclusion in the BGS Lexicon of Named Rock Units (https://www.bgs.ac.uk/technologies/the-bgs-lexicon-of-named-rock-units/). However, despite attempts to rationalize the plethora of local names, historical uncertainties are still reflected by the fact that even key regionally recognized formations have different names in different areas. The name changes are usually confined to places where there is a break in continuous outcrop. For example, the major quartzite formation that dominates the Islay Subgroup is known variously as the Jura Quartzite, the Schiehallion Quartzite, the Creag Leacach Quartzite, the Kymah Quartzite and the Durn Hill Quartzite. A similar number of names are used for the same quartzite in Ireland.

Other relics of historical nomenclature persist in formal names that do not accord with accepted stratigraphical practice but have been adopted because the name is well established. Thus, for example, we have the Ballachulish Slate Formation immediately succeeding the Ballachulish Limestone Formation and both a Lower and an Upper Erins quartzite formation. In addition a few informal names remain, usually because a formal rank and correlation have yet to be established. These are placed in parentheses and not capitalized (e.g. the Stuartfield ‘division’ of the North-east Grampian Highlands). References to previous uses of the terms ‘division’ and ‘group’ are also parenthesized.

1.4.1 Grampian Group

The Grampian Group crops out over an area of approximately 4250 km² in a broad NE-trending area extending from Glen Orchy to near Elgin, with an isolated outcrop on the north coast around Cullen. It forms most of the Northern Grampian Highlands and large parts of the Central Grampian Highlands as defined in this volume (chapters 3 and 5). The group consists mainly of psammites and semipelites, with some quartzites, all at amphibolite-facies metamorphic grade. The coarse-grained siliciclastic Scatsta Group at the base of the Shetland Dalradian succession could well be a correlative, as could much of the Colonsay Group on Islay and Colonsay. It is also possible that the Bowmore Sandstone Group on Islay may be assigned to the Grampian Group. These problematical units are discussed separately below.

Regional mapping in the Northern Grampian Highlands has shown that deposition of the group occurred within a series of linked NE-trending basins bounded by major syndepositional faults and crustal lineaments (see (Figure 5.2)). Stratigraphical successions have been established in each of the main basins, although only the Corrieyairack Basin is known in detail, whilst the Cromdale and Strathtummel basins are still the subject of ongoing research. Correlation between the basins is tentative in parts. However, following the establishment of informal local stratigraphies (e.g. Thomas, 1980), an initial division into subgroups proposed by Winchester and Glover (1988) has now been expanded and adopted across the region (Glover and Winchester, 1989; Smith et al., 1999; Banks, 2005). The three subgroups are characterized by different lithofacies associations and are interpreted to represent distinct phases of early and syn-rift extension followed by a protracted period of post-rift thermal subsidence (Glover et al., 1995).

The nature of the contact between the gneissose and migmatitic basement of the Badenoch Group and the essentially non-gneissose lithologies of the Grampian Group is considered in some detail in the section on Basement to the Dalradian Basins and in the Introduction to Chapter 5.

For much of its length in the North-east and Central Grampian Highlands, the contact of the Grampian Group with the overlying Appin Group is defined tectonically by a regional high-strain zone of enigmatic origin and nature, known as the Boundary Slide-zone. However, to the south-west of the Schiehallion district the slide cuts up the succession so that a conformable Grampian–Appin group succession can be observed in places such as Glen Orchy and Kinlochleven and is inferred elsewhere. In those areas, the contact has been placed traditionally at the base of the Eilde Quartzite, now the base of the Loch Treig Schist and Quartzite Formation, although it is possible that unconformities exist locally (Winchester and Glover, 1988; Glover, 1993). A sedimentary transition is also recognized in the far north-east, where the boundary is rather arbitrarily defined to be above any major quartzite units.

The Grampian Group is geochemically distinctive and shares more characteristics with the Moine Supergroup of the Northern Highlands Terrane than it does with younger Dalradian groups. This is particularly noticeable in the regional geochemical maps, compiled from stream sediment data by the British Geological Survey (BGS) (Plant et al., 1999; Stone et al., 1999). The Moine Supergroup and Grampian Group are depleted in base elements such as Mg, Ni and Cu but are enriched in incompatible elements, Ba, K, La, Rb, Sr, U, and Zr, relative to later Dalradian sequences. In broad terms, this has been attributed to the change from dominantly psammitic sequences to more-mixed shelf and basin lithologies, which is reflected in a reduction in detrital minerals such as K-feldspar and an increase in marine clay minerals. Whole-rock analyses of the psammitic and semipelitic rocks are also chemically distinguishable from those in the overlying Appin Group and there are slight chemical differences between semipelites at different stratigraphical levels (Lambert et al., 1982; Haselock, 1984; Winchester and Glover, 1988).

Throughout the sedimentation of the Grampian Group there was a constant supply of detritus from a hinterland of exposed quartzofeldspathic gneiss and granitic rock (Hickman, 1975; Glover and Winchester, 1989). This is particularly noticeable in pebbly beds of the Glenshirra Subgroup. Analysis of detrital zircon populations indicates source areas of Proterozoic rocks, with a marked absence of any Archean detritus (Cawood et al., 2003). That study highlighted differences in provenance with time. In the Glenshirra Subgroup, populations are dominated by 1.8 Ga detritus with subsidiary 1.2 Ga detritus, suggesting that ‘Rhinnian-type’ basement was an important component of the source area at that time. Peaks at 2.0 Ga and 1.4 Ga characterize the Corrieyairack Subgroup, with progressive dilution by 1.1–0.9 Ga Grenvillian detritus in the more-mature sediments of the Glen Spean Subgroup. However, a study of detrital zircons and whole-rock Nd isotopes by Banks et al. (2007) revealed marked differences in provenance signatures between Grampian Group basins, which were attributed to sediment transport directions. Thus the Corrieyairack Basin contains voluminous 1.8 Ga detritus, inferred to have been derived from Ketilidian/Rhinian igneous basement in the Labrador region of Laurentia to the west, whereas the more easterly Strathtummel Basin was dominated by 1.0 – 0.9 Ga Grenvillian/Sveconorwegian sources in Baltica to the east. Further-travelled basin-axial deposition supplied a greater variety of late-Paleoproterozoic (1.8 – 1.6 Ga) and early Mesoproterozoic (1.6 – 1.4Ga) detritus to both basins.

1.4.1.1 Glenshirra Subgroup

The base of the Grampian Group is not seen in undisturbed contact with the underlying Badenoch Group rocks. The oldest exposed unit is the Glenshirra Subgroup, which has a type area in the core of the Glenshirra Dome in upper Speyside and in several fault- and shear-bounded inliers closer to the Great Glen Fault between Loch Lochy and Strath Errick (Haselock et al., 1982; Okonkwo, 1988; May and Highton, 1997).

The subgroup comprises sequences, up to 2 km in thickness, of geochemically distinct, immature feldspathic psammite and beds of pebbly psammite and metaconglomerate. The metaconglomerates thicken and increase in abundance both up section and towards the Great Glen Fault. Abundant sedimentary structures are of a type indicative of deposition by traction currents in shallow marine environments with periodic influxes of fluviatile deposits (Banks and Winchester, 2004). The progressive westward thickening and coarsening of the strata imply the presence of a basin margin to the west or north-west, approximately coincident with the present trace of the Great Glen Fault.

Banks and Winchester (2004) argued that the largely sub-aerial environments of the Glenshirra Subgroup contrast so markedly with the deep-water marine environments of the succeeding Grampian Group sequences, that the Glenshirra should be afforded the status of a separate group.

1.4.1.2 Corrieyairack Subgroup

Around the Glenshirra Dome and in the upper reaches of the River Nairn, the Glenshirra Subgroup rocks are conformably overlain by a distinctive and regionally widespread succession of semipelite and striped semipelite and psammite. This marks the base of the overlying Corrieyairack Subgroup, which is the major component of the Grampian Group in the southern Monadhliath mountains (Haselock et al., 1982; Okonkwo, 1988). The abrupt change to a semipelitic succession records a basin-wide flooding event heralding the start of a major period of widespread subsidence and rift-related extension (Banks, 2005).

There the basal semipelites are overlain by 4–5 km of siliciclastic rocks deposited by prograding turbidite complexes (Banks, 2005). Variations in sediment supply and source area are indicated by changes in the proportions of plagioclase and K-feldspar, whereas variations in bed thickness and form reflect depositional processes (Glover and Winchester, 1989; Key et al., 1997). Overlying semipelites record a reduction in sand supply and the development of shelf conditions along the basin margins as recorded by the lateral thickness and facies changes. The youngest parts of the subgroup record a renewed influx of sand-dominated turbidites, deposited by extensive fan-lobe systems derived from the north-west. These pass southwards and eastwards in the Glen Spean and Drumochter areas into shelf environments with several units of distinctive quartzite (Glover and Winchester, 1989; Glover et al., 1995; Robertson and Smith, 1999; Banks, 2005).

Throughout the subgroup, rapid local facies and thickness variations indicate contemporary tectonism and, together with the progressive overstep onto an interbasin high in the Glen Banchor area, they permit the tracing of outlines of former basin margins (Glover and Winchester, 1989; Robertson and Smith, 1999). Locally, as at Ord Ban and Kincraig near Aviemore, basement rocks of the Glen Banchor Subgroup are overlain directly by a condensed shallow-marine shelf sequence consisting of semipelite interbedded with calcsilicate rocks, thin quartzites, metacarbonate rocks and concordant sheets of amphibolite. These strata, previously assigned to the now defunct ‘Ord Ban Subgroup’ of Winchester and Glover (1988), are now redesignated as the Kincraig Formation. Comparable lithologies that constitute the Grantown Formation near Grantown-on-Spey (McIntyre, 1951; Highton, 1999) were also previously assigned tentatively to the ‘Ord Ban Subgroup’ but this formation includes metacarbonate rocks with a distinctly different lithogeochemical signature, more like those of the lower Appin Group (BGS unpublished data).

1.4.1.3 Glen Spean Subgroup

In all three major basins, the deep-water turbiditic rocks of the uppermost Corrieyairick Subgroup are conformably overlain by shallow-marine shelf sedimentary rocks of the Glen Spean Subgroup, consisting of a mature mixed sequence of semipelite and psammite with quartzites in the upper parts. Reduced subsidence and relative tectonic stability at this time are interpreted to represent post-rift thermal subsidence (Glover et al., 1995).

In the south-west of the Corrieyairack Basin, rocks of the subgroup were formerly known as the ‘Eilde Flags’. Around Spean Bridge, the succession is approximately 4000m thick (Winchester and Glover, 1991; Glover, 1993) but it thins southwards to about 100m in the Kinlochleven area and on the Black Mount. A thin development also occurs in the south-east of the basin, on the flank of the Glen Banchor ‘high’. Similar lithologies form a thick succession in the Strathtummel Basin, where a near-continuous section is exposed in the A9 road cuttings (see Chapter 3). There, the whole succession was formerly referred to as the ‘Struan Flags’, but Thomas (1980, 1988) described a ‘Drumochter succession’ of flaggy psammites and semipelites, overlain by a predominantly psammitic ‘Strathtummel succession’. Those informal successions were designated as new ‘Atholl’ and ‘Strathtummel’ subgroups by Treagus (2000) and on the BGS 1:50 000 Sheet 55W (Schiehallion, 2000). However, they are almost certainly local equivalents of the Glen Spean Subgroup and will be referred to as such in the future. Shallow-water sedimentary structures are well preserved in many of the psammitic beds and these have been described in detail and interpreted by Banks (2007).

In the Cromdale Basin, the upper 2–3 km of Grampian Group strata are dominated by micaceous psammites and quartzites, which are assigned to the Glen Spean Subgroup. Shallow-water quartzites become dominant towards the top of the subgroup, in the Cromdale Hills, around Rothes and along the north coast at Cullen, where they pass conformably upwards into Appin Group successions.

1.4.2 Appin Group

The Appin Group was derived from a varied sequence of shelf sediments and comprises pelites, semipelites, quartzites, calcsilicate rocks, metalimestones and metadolostones, usually with rapid alternations of facies (Anderton, 1985; Wright, 1988). Local successions are easily established and the group has been divided into three subgroups. Lateral facies changes, thickness variations and local unconformities are well documented in several areas, but most formations and certain key beds can be traced over large distances and there is an overall general consistency of facies from Connemara in western Ireland to the Moray Firth coast. Correlations between local successions have thus been made with reasonable confidence throughout the Grampian Highlands (Harris et al., 1994, fig. 14; Stephenson and Gould, 1995, fig. 10), aided in some areas by detailed studies of the whole-rock geochemistry of a variety of lithologies. Such studies have been more successful in the Appin Group than in other parts of the Dalradian succession (Lambert et al., 1981; 1982; Hickman and Wright, 1983; Rock et al., 1986). Of particular use are the geochemical studies of metacarbonate units, some of which retain distinctive geochemical characteristics over considerable distances (Rock, 1986; Thomas, 1989, 1993).

Rocks of the Appin Group crop out over some 2100 km² in a relatively narrow outcrop extending throughout the Grampian Highlands over a strike length of 325 km (Figure 1.1). In the south-west a complete sequence, which continues up into the overlying Argyll Group, is recognized in the core of the Islay Anticline. Thick developments occur around Appin and in Lochaber, which are type areas for the two lowest subgroups, the Lochaber and Ballachulish subgroups. South-eastwards from Appin, rapid facies changes and considerable attenuation occur; higher parts of the group were either not deposited, or are cut out by unconformities, or have been excised by possible tectonic dislocation along the Boundary Slide-zone. As a result, only a condensed and possibly incomplete sequence of Lochaber Subgroup rocks is present from Glen Orchy to Glen Lyon. A more-complete although still condensed sequence, which passes up conformably into the Argyll Group, reappears to the north of Schiehallion and expands rapidly eastwards to Blair Atholl, the type area of the highest, Blair Atholl Subgroup. The complete sequence is then traceable north-eastwards to Braemar. To the north of the Cairngorm and Glengairn granite plutons a similar succession has been traced northwards to link with the succession on the Moray Firth coast. Appin Group strata also occur as outliers in narrow fold hinges in the Northern Grampian Highlands, both in the Glen Roy area, where they represent an extension of the Lochaber stratigraphy, and farther to the east in the Geal-charn–Ossian Steep Belt. On Shetland, a varied succession of siliciclastic and carbonate rocks, typical of the Appin and Argyll groups, occurs in the middle of the ‘East Mainland Succession’ and it is this Whiteness Group that most strongly suggests a correlation with the Dalradian of mainland Scotland.

The shallow-marine environment that dominated the Glen Spean Subgroup continued up into Appin Group times, when sedimentation occurred generally on a shallow tidal shelf, overlying crust that was gently subsiding due to lithospheric stretching (Anderton, 1985; Wright, 1988). Conditions fluctuated between an open marine oxidizing environment and stagnant euxinic lagoons to produce the alternating sandstone and black pyritic mudstone sequences (Figure 1.6). The overall lateral persistence of sedimentary facies suggests that similar processes of deposition were widespread and that supply of sediment remained constant. As in the preceding Grampian Group, this was from a north-western landmass, although palaeocurrent indicators show that sediment was distributed along the shelf by tidal longshore currents. In detail, a series of NE-trending basins might have developed, bounded by synsedimentary growth faults, such as those that have been identified in the Glen Creran–Loch Leven district (Hickman, 1975; Litherland, 1980).

1.4.2.1 Lochaber Subgroup

In many parts of the Grampian Highlands a sequence of semipelites and pelites with lenticular interbedded quartzites succeeds the underlying more-persistently psammitic rocks of the Grampian Group. This conformable relationship persists throughout much of the type area, between Port Appin and Glen Roy and in the mountainous ground around Kinlochleven. Elsewhere, significant stratigraphical omissions suggest the presence of local unconformities, such as around the Geal-charn–Ossian Steep Belt, where over 1200 m of strata are absent locally, possibly as a result of footwall uplift at a basin margin (Glover et al., 1995; Robertson and Smith, 1999). The subgroup is absent between Onich and Glen Spean, where the Ballachulish Subgroup is in direct contact with the Grampian Group along what was formerly interpreted as the Fort William Slide (Bailey, 1934, 1960) and subsequently re-interpreted as a localized unconformity (Glover, 1993). Between Glen Orchy and Braemar the Lochaber Subgroup is represented only by a highly condensed sequence within and above the Boundary Slide-zone but farther north a thick sequence, broadly comparable with that of the type area, rests conformably upon the Grampian Group.

In its type area, the Lochaber Subgroup has a maximum aggregate thickness of 4200m (Hinxman et al., 1923; Bailey, 1934). Hickman (1975) defined type sections for some of the units within a continuous section along the River Leven and Loch Leven. Distinctive quartzite units are traceable over strike lengths of tens of kilometres, although they do taper out laterally and might be diachronous. Around Glen Coe and Kinlochleven there are three major quartzites (the Eilde, Binnein and Glen Coe quartzites), all of which thin north-eastwards towards Loch Treig, where the succession becomes dominantly semipelitic, and they taper out north of Glen Spean (Key et al., 1997). The quartzites become finer grained and less feldspathic towards the north-east, a mineralogical change that is reflected in their whole-rock chemistry (Hickman and Wright, 1983). Contacts between the quartzites and interbedded semipelites are commonly transitional over several metres with fine-scale interleaving of the two lithologies and consequently the alternating units are all classed as members within a single Loch Treig Schist and Quartzite Formation.

The upper part of the subgroup in its type area comprises a homogeneous sequence, up to 2200 m thick, of distinctive striped, greenish grey phyllitic or schistose semipelite and pelite with minor thin quartzites, termed the Leven Schist Formation. The pelites and semipelites are less feldspathic than those in the Grampian Group, and this is matched by distinct changes in the regional whole-rock chemistry and provenance (Lambert et al., 1982; Winchester and Glover, 1988; Stone et al., 1999). The uppermost strata become increasingly calcareous and contain thin beds of metalimestone in the Glen Spean area.

To the south-west, the upper part of the Lochaber Subgroup is exposed in the core of the Islay Anticline, where it consists of a lower quartzite unit with pebble beds containing extrabasinal granite clasts, and an overlying semipelitic unit which becomes more calcareous upwards, comparable to the Leven Schists (Rast and Litherland, 1970; Wright, 1988). However, south-east of the type area a tripartite division is recognized in which basal dark grey pelites and semipelites are overlain by striped semipelites with thin metacarbonate beds, which in turn are overlain by pale greyish green semipelites (Hickman, 1975; Litherland, 1980).

On the north-western edge of the Geal-charn–Ossian Steep Belt, between Ben Alder and Kingussie, semipelites and quartzites similar to those of the Loch Treig Schist and Quartzite Formation have been assigned to the lower part of the Lochaber Subgroup (Robertson and Smith, 1999). At the top of this succession is the Kinlochlaggan Boulder Bed, containing clasts of a variety of intra- and extrabasinal rock types including granite (Treagus, 1969, 1981; Evans and Tanner, 1996). These beds were formerly interpreted as tillites, deposited directly from ice-sheets, but the clasts are now thought to have originated as ice-rafted dropstones. Either way this is the earliest evidence for glacial activity in the Dalradian succession.

Between Glen Orchy and Braemar, thin developments of the Lochaber Subgroup have been recognized in condensed sequences, rarely exceeding a few hundred metres in thickness (Treagus and King, 1978; Roberts and Treagus, 1979; Upton, 1986; Treagus, 1987, 2000). The junction with the underlying Grampian Group lies within a zone of high strain, the Boundary Slide-zone, in which several formations are strongly attenuated or even excised locally along tectonic breaks. However, in some areas, there seems to be a continuous overall stratigraphical transition from the Grampian Group into the Lochaber Subgroup, as in the Gilbert’s Bridge GCR site in Glen Tilt. The reduced thickness is not solely attributed to deformation in the slide-zone; there would seem to be considerable sedimentological thinning in this area, as there is in many of the succeeding Appin Group formations. Recognition of component formations is hampered by the highly tectonized state of lithological units such as the Beoil Schist Formation in the Schiehallion area and the Tom Anthon Mica Schist Formation south-west of Braemar. Calcareous Leven Schist-type lithologies occur in the Meall Dubh Striped Pelite Formation of the Schiehallion area and in the Glen Banvie Formation of Glen Tilt.

North of the granitic plutons that mark the Deeside Lineament, Grampian Group psammites are overlain conformably by micaceous psammites with thin quartzites and semipelites, passing up locally into slightly calcareous lithologies. The subgroup thickens markedly northwards, where the Findlater Flag Formation, passes transitionally upwards into calcareous psammites and semipelites locally termed the Pitlurg Calcareous Flag Formation and Cairnfield Calcareous Flag Formation.

Throughout its outcrop, the lower part of the subgroup records the continuation of relatively shallow marine conditions from Grampian Group times and the basal psammites, quartzites and pelites show close affinities with the Grampian Group. The major quartzites exhibit well-preserved shallow-water sedimentary structures, such as cross-bedding, grading, ripple-marks, slump and dewatering structures and have been interpreted generally as nearshore tidal sand bodies (Hickman, 1975; Wright, 1988). They probably represented periodic basin shoaling events in a delta that extended north-eastwards over a distance of about 40km, with facies varying from proximal at Appin to distal at Glen Roy (Glover et al., 1995; Glover and McKie, 1996). Farther east, where the quartzites and associated psammites pass into semipelite, the complex facies variations reflect generally deeper water environments (Key et al., 1997). Gradually the coarse siliciclastic basin-fills gave way to widespread deep-water muddy sedimentation during the first of several marine transgressions that typify subsequent sedimentary cycles of the Appin and Argyll groups. In almost all areas, the highest beds of the subgroup are variably calcareous with tremolitic amphibole, and rare units of impure metacarbonate rock, heralding the more-persistent and more-widespread carbonate sedimentation of the Ballachulish Subgroup. They have been interpreted as reflecting the establishment of local lagoonal environments with seawater-precipitated magnesium-rich carbonate rocks and possible seasonal desiccation (Thomas, 1989; Stephenson, 1993). This broadly calcareous part of the succession is a useful stratigraphical marker.

1.4.2.2 Ballachulish Subgroup

This subgroup, more than any other in the Dalradian, exhibits a remarkable lateral continuity of lithological type; key elements of the sequence can be traced, almost on a bed-for-bed basis, from Connemara to the Moray Firth coast, attesting to widespread stability and relatively uniform subsidence at this time.

In most parts of the Grampian Highlands, the contact of the Lochaber Subgroup with the Ballachulish Subgroup is conformable; there is a transition from calcareous semipelites and tremolitic calcsilicate rocks to a background lithology of graphitic pelites with more-persistent discrete beds of metacarbonate rock that are commonly dolomitic. Notable exceptions occur between Onich and Glen Spean, where the basal Ballachulish Subgroup rests with local unconformity on the Grampian Group (Glover, 1993), and on the north-western edge of the Geal-charn–Ossian Steep Belt where, as a result of non-deposition and erosion on a longstanding footwall ‘high’, the upper part of the Ballachulish Subgroup oversteps the lower part and the entire Grampian Group to rest directly on the Badenoch Group (Robertson and Smith, 1999). In the Boundary Slide-zone, despite the highly attenuated successions, a complete stratigraphical transition can be recognized in most areas.

Five formations are recognized in the Lochaber–Appin area (Bailey, 1960; Hickman, 1975; Litherland, 1980). The Ballachulish Limestone Formation and Ballachulish Slate Formation together account for over 500 m of succession in the type area, around Ballachulish and Onich, but thin northwards around fold closures in the area of Glen Roy, where they are partly excised by slides (Key et al., 1997). Phyllitic grey-green calcsilicate rocks, cream and grey metadolostones and dark bluish grey metalimestones define the lower formation within a background lithology of slaty and phyllitic graphitic pelites that become dominant upwards. Intercalations of graded psammite and quartzite on scales from a few millimetres to about a metre become numerous in the upper part of the pelites where they form the distinctively striped Appin Transition Formation. Thesucceeding Appin Quartzite Formation is about 300m thick in its type area but it thins considerably to the north-east like the quartzites of the Lochaber Subgroup. This massive, clean, locally feldspathic quartzite is distinctive from other Dalradian quartzites throughout its strike length, and is characterized by sedimentary structures such as cross-bedding, ripple marks and graded bedding. The overlying Appin Phyllite and Limestone Formation consists of an alternating sequence of phyllitic and flaggy semipelites and psammites, metacarbonate rocks and thin quartzites. It attains a total thickness of up to 400m in the type area. The metacarbonate rocks, which are more prevalent in the lower part, include several very distinctive lithologies, such as the pure white Onich Limestone and the aptly named ‘tiger-rock’ of Bailey (1960), which consists of regularly-spaced 5to 10cm layers of orange-weathering dolomitic carbonate and dark grey semipelite.

South-west of the type area, representatives of the Appin Quartzite and the Appin Phyllite and Limestone formations crop out on small islands in the Firth of Lorn, and still farther to the south-west, after a gap of 75km, the subgroup crops out in the core of the Islay Anticline (Rast and Litherland, 1970).

To the south-east of Appin, the type succession thins rapidly across strike and marked facies changes occur (Litherland, 1980, figure 6). In Glen Creran the lower part of the sequence might be cut out by an unconformity and local non-sequences within the Appin Phyllite and Limestone Formation suggest some syndepositional tectonic control (Litherland, 1980). Farther to the south-east, beyond the Etive Pluton, the subgroup is absent. Small outliers of schistose calcareous rocks below the Boundary Slide-zone in the Bridge of Orchy area were formerly assumed to represent the Ballachulish Limestone (Bailey and Macgregor, 1912; Thomas and Treagus, 1968; Roberts and Treagus, 1979) but recent mapping by the Geological Survey has re-assigned them all to the Lochaber Subgroup. The remainder of the Appin Group succession is absent from this area, probably as a result of sedimentological factors with further tectonic attenuation in the Boundary Slide-zone.

The attenuated Ballachulish Subgroup re-appears north-eastwards as a condensed sequence totalling only 100m at Errochty Water, but increases to 700m in the Schiehallion area(Treagus and King, 1978; Treagus, 2000). The subgroup continues to the Loch Tay Fault at Foss where it is displaced north-eastwards to the Blair Atholl area (Pantin, 1961; Smith and Harris, 1976). It can then be traced in continuous outcrops north-eastwards to Braemar (Upton, 1986; Goodman et al., 1997; Crane et al., 2002). Throughout this continuous strike length of some 65km the succession can be matched almost bed for bed with that in the type areas of Lochaber and Appin.

The subgroup is well developed to the north of the granitic plutons of the Deeside Lineament and the distinctive formations can be traced as far as the Keith area (see Chapter 6). However, the units become ill defined north of Keith, where marked facies changes probably reflect the original basin architecture. There, apart from a condensed sequence of metalimestone, graphitic pelite and quartzite around Deskford, and a thick sequence of graphitic pelite with thin basal metacarbonate rocks at Sandend Bay, much of the subgroup appears to be absent.

The base of the Ballachulish Subgroup marks a significant break from the siliciclastic rock-dominated successions of the Grampian Group and Lochaber Subgroup to the limestone–mudstone–sandstone facies associations of the higher parts of the Appin Group and much of the Argyll Group. It is interesting to note that this change coincides with the appearance of Archaean detrital zircon grains in the sediment load (Cawood et al., 2003). The limestones and graphitic mudstones of the lowest formations indicate a major marine transgression and widespread subsidence, with the progressive development of shallow, tidally influenced shelf sedimentation and anoxic lagoonal environments (Anderton, 1985; Wright, 1988). The more-persistent mudstones of the Ballachulish Slate Formation have been interpreted as prodelta clay deposits, and encroachment of upward-coarsening, fine quartz sands from the delta into deeper water produced the overlying sandstones such as the Appin Quartzite. A return to interbedded limestone, calcareous mudstone and siltstone deposition (the Appin Phyllite and Limestone Formation) indicates renewed transgression. Considerable along-strike continuity of facies is seen in all of the formations but the shelf must also have been relatively narrow to explain the rapid down-dip facies changes seen, for example, between Appin and Glen Creran (Litherland, 1980). The sporadic distribution of all of the formations, particularly in the NE-trending block between the Etive and Rannoch Moor plutons and the Bridge of Balgie Fault probably reflects non-deposition and/or erosion due to local uplift rather than tectonic excision.

1.4.2.3 Blair Atholl Subgroup

The Blair Atholl Subgroup maintains a generally constant lithology of dark pelites and metalimestones from Connemara to the Moray Firth, although local successions differ in detail, making bed-for-bed matching difficult. In some areas, notably Islay, the Blair Atholl district and parts of the North-east Grampian Highlands, the upper part of the subgroup is less pelitic with banded semipelites, micaceous psammites, metalimestones and metadolostones comprising a distinctive ‘pale group’.

In the Ardsheal Peninsula of the Appin area, the Ballachulish Subgroup is seen to be overlain by some 300m of dominantly slaty pyritiferous pelites and semipelites with minor dark-grey metalimestones and some more-psammitic beds, all comprising the Cuil Bay Slate Formation (Hickman, 1975). The pelites and semipelites pass upwards into the Lismore Limestone Formation, a 1km-thick graphitic sequence of flaggy, banded, blue-grey metalimestones with thin pelite members, which forms the whole island of Lismore. It is also seen as an inlier in the core of the Loch Don Anticline in south-eastern Mull. Hickman (1975) has divided the formation into fifteen members and recognises several limestone–mudstone cycles. Slump folds and syndepositional breccias indicate periods of sediment instability along basin margins. Individual limestone formations are laterally persistent and contain ooids, stromatolites and lenticles of chert.

Owing to the south-westerly plunge of the major folds, any higher beds of the subgroup that might have been deposited in the Appin–Lismore area lie beneath the Firth of Lorn, but an extended sequence has been recognized in the Islay Anticline (Rast and Litherland, 1970). There, graphitic pelites and a metalimestone, equated with the Cuil Bay Slate and Lismore Limestone, are overlain by more pelites, semipelites and the distinctive, partly ooidal and stromatolitic, Lossit Limestone (Spencer, 1971) (see Chapter 2, Introduction).

To the south-east of the Appin area, in Glen Creran, the subgroup thins and facies changes similar to those in the underlying Ballachulish Subgroup result in a more-semipelitic succession (Litherland, 1980). Like its predecessor, the Blair Atholl Subgroup is absent eastwards from Glen Creran due to a presumed combination of sedimentary thinning and possible movement within the Boundary Slide-zone.

Strata of the Blair Atholl Subgroup reappear east of Loch Errochty in stratigraphical continuity with the Ballachulish Subgroup and a complete sequence, 250to 350m thick, is present between there and the Loch Tay Fault at Foss (Treagus and King,1978; Treagus, 2000). A lower sequence of dark graphitic pelites and metalimestones is equated with the Cuil Bay Slate and Lismore Limestone and an overlying non-graphitic ‘pale group’ is generally composed of ribs of psammite and quartzite in a graded pelitic or semipelitic matrix. At the top of the subgroup is a pale cream-weathering dolomitic metalimestone, equated with the Lossit Limestone.

Across the Loch Tay Fault, around Blair Atholl, a similar continuous succession from the Ballachulish into the Blair Atholl subgroup has been demonstrated (Smith and Harris, 1976). This constituted the original type succession for the Blair Atholl ‘series’ (Bailey, 1925; Pantin, 1961). Thick beds of dark graphitic metalimestone in the lower part of the Blair Atholl Subgroup are a distinctive feature in this area and the main formations can be followed north-eastwards through the Glen Shee area almost to Braemar (Upton, 1986; Goodman et al., 1997; Crane et al., 2002).

North of the Deeside Lineament the Blair Atholl Subgroup consists mainly of semipelites, which are more-pelitic, graphitic and calcareous locally. A prominent thick metalimestone, the Inchrory Limestone Formation, occurs in its central part and minor metalimestones occur locally. The metalimestones thicken considerably around Fordyce and dominate the subgroup in the north coast section.

In many areas, the junction between the Ballachulish and Blair Atholl subgroups is taken at a fairly abrupt change from a background lithology dominated by semipelite and micaceous psammite to one dominated by graphitic pelite, marking a change to deeper de-oxygenated marine conditions. In contrast with the lateral stratigraphical continuity exhibited by the underlying strata, the greater variation seen between local successions might indicate deposition in a series of smaller basins. Basin margins are possibly marked by thinning of units and lateral changes of facies into coarser grained siliciclastic deposits (semipelites and micaceous psammites) and more-argillaceous limestones. Incomplete stratigraphical sequences in some areas might indicate that sediment supply periodically exceeded subsidence, so that emergence led to local non-deposition.

1.4.3 Argyll Group

The base of the Argyll Group as originally defined was marked by a tillite or a sequence of tillites, almost certainly deposited beneath ice sheets during a glacial event of restricted duration. The number of tillites and the character of the formations that contain them vary, but they are present throughout most of the Dalradian outcrop and hence constitute an important chronostratigraphical marker, not only in the British Isles but throughout the Caledonides. Unfortunately there are no obvious associated targets for direct radiometric dating and consequently there has been much debate as to with which of several dated global glacial events they might be correlated (see p. 47 for a full discussion).

The group has been divided into four subgroups. The tillite-bearing sequence is followed by local carbonate successions, interpreted by some as 'cap carbonates', and by thick shallow-water shelf and deltaic quartzites of regional extent. Together, these constitute the oldest, Islay Subgroup, which marks the end of the stable shelf conditions that characterized Appin Group deposition. The succeeding Easdale, Crinan and Tayvallich subgroups are generally characterized by basin deposits, turbidites and unstable basin margin slump deposits, with only one widespread major shallow-water interlude, in the upper Easdale Subgroup. It is rarely possible to trace individual beds in the Easdale and Crinan subgroups for any great distance and correlations are made on the grounds of general similarity of facies. In contrast, for much of its outcrop the Tayvallich Subgroup is dominated by a calcsilicate- and carbonate-rock unit, which forms another of the principal marker bands of the Dalradian succession. The Argyll Group is also characterized by penecontemporaneous volcanic activity, evidence of which is found throughout the group but which reaches a maximum development in the Tayvallich Subgroup. This volcanism has provided the most-precise radiometric dates in the whole Dalradian succession (see p. 47). Further evidence for tectonic instability comes from widespread stratabound exhalative mineralization in the Easdale Subgroup, which includes economically significant sulphide and bedded baryte deposits near Aberfeldy.

Rocks of the Argyll Group crop out over an area of some 5700km² (Figure 1.1). Extensive outcrops occur on Islay and Jura, and the type successions for all four subgroups are found in the South-west Grampian Highlands. Between the Etive Pluton and the Bridge of Balgie Fault, the Islay Subgroup, like much of the preceding Appin Group, is absent, although most of the younger units are continuous. Farther to the north-east a full succession is present through the Tummel Steep Belt to the Glen Shee area, with the higher units continuing to Deeside. In the North-east Grampian Highlands, west of the Portsoy Shear-zone, the lowest units have been traced intermittently to the north coast. East of the Portsoy Shear-zone, the higher parts of the group are believed to be present in an undivided gneissose sequence forming a horseshoe outcrop pattern around the Turriff Syncline. In Shetland, the absence of a tillite formation hampers identification of an Appin/Argyll group boundary, but the metalimestone-bearing upper part of the Whiteness Group is thought to be broadly equivalent to the Argyll Group of the mainland Dalradian.

In the Islay Subgroup, conditions of deposition on or close to an extensive continental shelf were similar to those of the preceding Appin Group. The Easdale Subgroup then records the onset of renewed instability with cycles of rapid basin deepening and variable lithofacies within a series of NE-trending basins. Locally thick clastic units indicate that sediment input kept pace with extension in some of the basins, and crustal rupturing allowed sub-marine volcanism to reach a climax in late Argyll Group time (Fettes et al., 2011).

1.4.3.1 Islay Subgroup

This largely psammitic subgroup is dominated in most areas by a thick quartzite formation. The quartzite and the underlying basal tillite-bearing formation are persistent and distinctive, enabling a good correlation of beds from Connemara to the Moray Firth.

The basal Port Askaig Tillite Formation is one of the most obvious and readily recognized lithostratigraphical units within the Dalradian succession. In the type area on Islay and in the Garvellach Isles, the combined sequence could be up to 870m thick, with up to 47 separate tillites recognized (Spencer, 1971, 1981). It consists of a sequence of metasandstones, metasiltstones, metaconglomerates, metadolostones and metadiamictites that have commonly been referred to informally as ‘boulder beds’. These beds range from 0.5m to 65m in thickness and contain boulders up to 2m in diameter. The lower beds contain clasts of dolostone, probably derived locally from within the formation or from the underlying Blair Atholl Subgroup. However, the higher beds contain clasts of granite and gneissose granite of extrabasinal origin (Fitches et al., 1990). This division on the basis of clast content is also recognizable in metadiamictite sequences outwith the type area. The metadiamictites are generally regarded as tillites, deposited from grounded ice sheets, but isolated clasts in associated varved siltstones have been interpreted as dropstones from floating ice.

The succeeding Bonahaven Dolomite Formation includes metasandstones, metamudstones and impure dolomitic rocks with stromatolites and has been the subject of intense sedimentological, petrographical and geochemical studies (Hackman and Knill, 1962; Klein, 1970; Spencer and Spencer, 1972; Fairchild, 1980a, 1980b, 1985, 1991; Kessler and Gollop, 1988). Outcrops are restricted to north-east Islay, where the succession has been divided into four members, with a total thickness of up to 350 m.

The Jura Quartzite Formation (formerly also referred to as the Islay Quartzite) marks an abrupt change to a thick succession of cross-bedded and pebbly quartzites characteristic of a slightly deeper water tidal shelf environment (Anderton, 1976). The quartzites form almost all of the islands of Jura and Scarba and crop out on both limbs of the Islay Anticline on Islay. The thickest development of over 5000m is on Jura from where the formation thins markedly along strike, to both the south-west and north-east (see Anderton, 1979, figure 1). Sedimentary structures imply dominant palaeocurrent flow directions towards the north-north-east throughout the formation, with a general change towards finer grained facies observed in the same direction.

Farther to the north-east in the Ardmucknish area, the Islay Subgroup is reduced in thickness to between 300and 800m (Litherland, 1980). The sequence there is similar to that of the type area in that it includes two dolomitic metadiamictites at the base, followed by dolomitic metacarbonate rocks and pebbly, cross-bedded quartzites. From Ardmucknish the subgroup increases in thickness across strike to between 4500 and 7000m around Glen Creran, some 10km to the east, where a distinctly different succession, consisting predominantly of flaggy and pebbly quartzites and semipelites with no metadiamictites, was termed the Creran succession by Litherland (1980). Graded turbidites and ‘green beds’ of possible volcaniclastic origin in the lower part of this succession are features which become more common in the subgroup in the North-east Grampian Highlands.

Eastwards from the Etive Pluton to the Bridge of Balgie Fault, the Islay Subgroup is not present. Rocks of the overlying Easdale and Crinan subgroups rest directly upon the Lochaber Subgroup with an intervening zone of highly strained rocks that have been interpreted as representing the Boundary Slide-zone (Roberts and Treagus, 1979). If rocks of the Islay Subgroup were not deposited in this area, it is necessary to invoke a fault-bounded structural ‘high’, separating rapidly subsiding basins on either side, to account for the great thicknesses observed in the adjoining areas. It is thought that non-deposition, contemporaneous erosion and resultant unconformities above such ‘highs’ might account for many gaps in local Islay Subgroup successions (Pantin, 1961; Harris and Pitcher, 1975; Harris et al., 1994).

In the Schiehallion and Pitlochry areas, the Islay Subgroup is well developed, with a succession comparable with that of the type area (Bailey, 1925; Bailey and McCallien, 1937; Pantin, 1961; Harris, 1963; Treagus, 2000). The Schiehallion Boulder Bed Formation can be traced throughout most of the area, but is particularly well developed on the northern slopes of Schiehallion itself and is well exposed at the Tempar Burn GCR site. The lower part has a matrix of calcareous semipelite with only carbonate-rock clasts, whereas the upper part is more siliceous with large clasts of granite, syenite and quartzite. The overlying Schiehallion Quartzite Formation is typically massive and rarely cross-bedded. In the lower part, conglomeratic beds contain boulders of granite identical to those of the metadiamictites; dolomitic beds, consisting of tremolitic metalimestone and calcareous pelites, are well developed locally. Farther north-east, towards Braemar, thin developments of pebbly, granitic metadiamictite occur locally at the base of the massive Creag Leacach Quartzite Formation (Upton, 1986; Goodman et al., 1997; Crane et al., 2002).

To the north of the Deeside Lineament the lower part of the Islay Subgroup consists of two interdigitating and diachronous formations, the psammitic Ladder Hills Formation and the Nochty Semipelite and Limestone Formation. Thin beds of metadiamictite, typically underlain by a thin metadolostone, occur locally towards the top of the Ladder Hills Formation and within the lower units of the overlying Kymah Quartzite. Minor basic tuffs and pillow lavas occur locally. The Kymah Quartzite varies considerably in thickness between 10m and 500m, probably reflecting the influence of the underlying basin architecture. However, the whole subgroup is probably cut out structurally by the Portsoy Shear-zone to the west of Huntly. It re-appears close to the north coast as the Durn Hill Quartzite Formation, which overlies a psammitic formation characterized by the presence of metadiamictite (Spencer and Pitcher, 1968).

A dramatic climatic change that was possibly worldwide took place at the beginning of Argyll Group times. On page 47 it is argued that this is most likely to be equated with the Marinoan glacial period at c. 635 Ma. Tillites were probably deposited on a shallow-marine shelf by up to seventeen successive pulses of grounded ice, possibly advancing from the south-east. Pre-existing intrabasinal sedimentary rocks were eroded by the glaciers and then covered by marine tills in which extrabasinal granitoid debris becomes increasingly common upwards. Isolated dropstones in varved siltstones infer local flotation of the ice sheet but large-scale deposition from ice rafts and reworking by downslope mass-flow, as suggested by Eyles and Eyles (1983), had been rejected by Spencer (1971). Periods of emergence and periglacial weathering between the glacial cycles are suggested by polygonal sandstone wedges, interpreted as ice wedges (Eyles and Clark, 1985). Beach conglomerates then heralded the start of a marine transgression followed by the next glacial advance. Dolomitic limestones with stromatolites, suggestive of warm water, were deposited during some of the interglacial periods, and on Islay, as in Donegal (McCay et al., 2006), a sequence that includes dolomitic metacarbonate rocks has been interpreted as a ‘cap carbonate’ such as commonly occurs above tillite sequences of various ages worldwide. The paradox of having an ice sheet at sea level, followed immediately by warm-water carbonate precipitation in apparently low latitudes has so far eluded a completely satisfactory explanation.

After the final retreat of the ice, the cold climate conditions apparently ameliorated and shelf sedimentation resumed, resulting in the deposition of the widespread quartzite formations. Locally thick accumulations of psammite and quartzite indicate that sediment input kept pace with extension in a series of NE-trending basins. The closest shoreline remained north-west of Islay and Jura with at least 100km of open sea to the south-east (Anderton, 1985). The change to a tidal, shallow-water environment was probably caused by a combination of source-area uplift and a tectonically induced marine transgression. This change in sedimentation is also marked by increasing volumes of Archaean detrital grains above the level of the tillite formations (Cawood et al., 2003). In some areas, thin but significant accumulations of volcaniclastic detritus and local pillow lavas could be early signs of basin instability that was soon to become widespread.

1.4.3.2 Easdale Subgroup

The base of this subgroup is marked in most places by a sharp change to finer grained rocks showing features typical of deep-water sedimentation and turbidity currents, with local incursions of coarse-grained mass-flow deposits. Higher parts of the subgroup are dominated by calcareous clastic rocks with local metacarbonate rocks, representing shallower water sedimentation. These general characteristics are preserved throughout the outcrop from Islay almost to the north coast and, although the detailed stratigraphy is less continuous than in preceding subgroups, individual units and some very distinctive sequences can be traced for up to 100km.

Along the south-eastern coasts of Islay and Jura, the Jura Quartzite Formation is overlain by the Jura Slate, which constitutes the basal member of the Scarba Conglomerate Formation (Anderton, 1979). This formation is about 450m thick, consisting of metamudstones that pass upwards into quartzite and metaconglomerate. Sedimentary structures such as graded beds with erosional bases suggest deposition from turbidity currents on a subsiding off-shore platform shelf. Farther north, on Scarba and adjoining islands, coarse debris flows are considered to have slumped downslope northwards into a fault-bounded basin. To the north of Scarba the debris flows and associated quartzites become finer-grained and thinner as they become interbedded with and pass upwards into the deeper water Easdale Slate Formation (Anderton, 1979, 1985). The Easdale Slate consists predominantly of graphitic, pyritic metamudstones, best developed on the islands of Seil and Luing where they have been quarried extensively for roofing slates. Thin beds of poorly graded metasandstones represent incursions of distal turbidites. Still farther north, in the Ardmucknish area, the Selma Black Slates and overlying Selma Breccia areinterpreted as equivalent to the Jura Slate and Scarba Conglomerate. The overlying Culcharan Black Slates (200–500 m) and the graded, pebbly Culcharan Quartzite (200 m) were correlated with the Easdale Slate by Litherland (1980).

To the east of Ardmucknish, the Creran succession of Litherland (1980) is difficult to correlate with established Argyll Group successions elsewhere (see above). Anderton (1985) regarded the whole succession as typical of the Easdale Subgroup but Litherland (1987), whilst accepting the similarity of parts of the topmost Beinn Donn Quartzite Formation to the Killiecrankie Schist and Cairn Mairg Quartzite of Perthshire (see below), re-iterated his 1980 view that the whole succession is equivalent to the Islay Subgroup.

Throughout its strike length, the upper part of the Easdale Subgroup is dominated by a variably calcareous shallow-marine facies. In the South-west Grampian Highlands this facies is represented by three major formations, separated geographically and cropping out on different limbs of major folds but all broadly equivalent stratigraphically (Roberts, 1966a). On the north-west limb of the Loch Awe Syncline, the Easdale Slate passes upwards, locally via the 20m-thick Degnish Limestone Formation, into the Craignish Phyllite Formation. The latter consists of abundant alternations of laminated phyllitic metamudstones, quartzites, metalimestones and pebbly metasandstones, with a total thickness of up to 4500m. Many of the sediments were deposited on tidal flats and in subtidal environments with gypsum pseudomorphs, well preserved in the Craignish Point GCR site, that indicate periodical emergence. Above the Craignish Phyllite, cross-bedded, dark grey metalimestones, intercalated with metamudstones comprise the Shira Limestone and Slate Formation, which is a persistent local marker unit, up to 300m thick. On Islay and Jura, on the south-eastern limb of the Islay Anticline, the Port Ellen Formation and its basal Kilbride Limestone Member consist of similar lithologies and occupy a similar stratigraphical position. There they are overlain by the Laphroaig Quartzite Formation.

On the south-eastern limb of the Loch Awe Syncline, and passing down into the core of the Ardrishaig Anticline, the upper part of the subgroup is represented by the Ardrishaig Phyllite Formation, overlain locally by the Shira Limestone. The Ardrishaig Phyllite is similar lithologically to the Craignish and Port Ellen phyllite formations but sedimentary structures are not so well preserved owing to a higher metamorphic grade. The amount of deformation makes estimates of thickness difficult but over 4000m are recorded in places. On the south-eastern limb of the Ardrishaig Anticline, a diachronous, predominantly quartzitic unit with an apparent thickness of up to 5000m, the Lower Erins Quartzite, replaces the upper part of the Ardrishaig Phyllite in Knapdale. There, the quartzite is succeeded by the dominantly pelitic Stornoway Phyllite Formation (with a more-graphitic and calcareousStronachullin Phyllite Member locally). Towards the north-east, around upper Loch Fyne, the Ardrishaig Phyllite is overlain directly by the St Catherine’s Graphitic Schist, consisting of up to 200 m of graphitic metamudstones with thinly bedded metalimestones that are probably equivalent to the Shira Limestone. Minor ‘green beds’ of detrital volcanic material occur in the south-eastern parts of the Ardrishaig Phyllite and Lower Erins Quartzite outcrop, and numerous basic sills could be near-contemporaneous with sedimentation.

The Easdale is the earliestsubgroup in the Dalradian that can be traced continuously from the South-west Grampian Highlands through the Central Grampian Highlands to the Pitlochry area and beyond. Black slaty pelites overlain by Ardrishaig Phyllite-type lithologies in the area around Dalmally can be equated with both the Easdale Slate to the west and the Ben Eagach Schist to the east, hence providing a crucial link (Roberts and Treagus, 1975). Throughout this area, a common succession may be recognized above the Boundary Slide-zone which, with only local absences, consists of the Killiecrankie Schist–Carn Mairg Quartzite–Ben Eagach Schist–Ben Lawers Schist formations.

The Killiecrankie Schist Formation is a deep-water facies consisting dominantly of semipelite and pelite but with abundant intercalated psammite and pebbly quartzite and many concordant bands of amphibolitic metabasalt. Where there is no quartzite to represent the Islay Subgroup, as in most of the ground between Blair Atholl and Glen Shee, it is difficult to distinguish between the Killiecrankie Schist and the underlying Blair Atholl Subgroup, especially where the sequence is attenuated by tectonic slides, and the facies is probably continuous across the subgroup boundary (Goodman et al., 1997; Crane et al., 2002). The distinctive Carn Mairg Quartzite ranges from feldspathic, pebbly quartzite, to a psammitic metagreywacke and typically shows graded bedding. Locally it is absent and it is probably best regarded as the thickest and most extensive of many sandy turbidites that swept periodically into the fine-grained basin sediments now represented by the Killiecrankie and Ben Eagach schist formations.

The Ben Eagach Schist Formation can be traced almost continuously from north of Tyndrum eastwards to the Glen Shee area. It consists predominantly of distinctive graphitic pelites, with impersistent thin metalimestones and amphibolites. Most of the stratabound mineralization of the Argyll Group occurs in this formation, extending intermittently over a strike length of at least 90km, from Loch Lyon to Loch Kander (Coats et al., 1984). Most notable is the 50m-thick bed of baryte and celsian with sulphides around Ben Eagach and Farragon Hill, north of Aberfeldy, which has been mined commercially in recent times.

The Ben Lawers Schist Formation forms a wide continuous outcrop extending from Tyndrum to the Braemar area. North of Loch Tay it also occupies the core of the Ben Lawers Synform. The dominant lithology is a calcareous pelite or semipelite with some thin beds of quartzite and metalimestone. Hornblende-schists of metasedimentary origin are common, but chloritic ‘green beds’ of volcaniclastic origin and pods of basic and ultrabasic meta-igneous material are also recorded. A persistent zone of stratabound sulphide occurs near the top of the formation.

Above the Ben Lawers Schist, the top of the subgroup is marked by a variety of lithologies. In the Glen Lyon area, the Auchlyne Formation (formerly the Sron Bheag Schist) consists mainly of psammite, semipelite and possible volcaniclastic rocks with various calcsilicate rocks and metalimestones. East of the Loch Tay Fault the Farragon Volcanic Formation, a complex sequence of quartzite, psammite, metabasalt and volcaniclastic ‘green beds’ represents the earliest major volcanic episode in the Dalradian succession.

Only the upper formations of the subgroup can be traced with any confidence north-east of the Pitlochry area. In the Glen Shee–Braemar area, the Creag Leacach Quartzite of the Islay Subgroup passes up through a transition member of interlaminated quartzite and pelite into the graphitic Glas Maol Schist, which is correlated with the Ben Eagach Schist (Upton, 1986). Farther north-east, in the Ballater area, the subgroup consists mainly of psammite and semipelite, the Craig nam Ban Psammite Formation, in the lower part, with semipelite, calcsilicate rock, amphibolite and thin impure metalimestones, the Glen Girnock Calcareous Formation, equivalent to the Ben Lawers Schist, in the upper part (Smith et al., 2002). The Meall Dubh Metabasite Formation, a unit of metabasalt and volcaniclastic rocks, occurs at the top of the subgroup in an equivalent position to the Farragon Volcanic Formation. Around the Ladder Hills, a more-typical Easdale Subgroup sequence consists of semipelite and psammite of the Culchavie Striped Formation, the Glenbuchat Graphitic Schist Formation and a sequence of calcareous semipelite and minor psammite with metalimestone and calcsilicate beds, termed the Badenyon Schist and Limestone Formation. Basic pillow lavas and tuffaceous beds of the Delnadamph Volcanic Member occur locally within the graphitic pelite and semipelite.

Still farther north, most of the succession above the Islay Subgroup is cut out by the Portsoy Shear-zone. Within and immediately to the east of the shear-zone, various local successions include Easdale-type facies but firm correlations with established successions are impossible. On the north coast, immediately west of the shear-zone, the subgroup is represented by a sheared and attenuated succession consisting of the Castle Point Pelite and Portsoy Limestone formations.

An initial shelf-deepening event at the start of the Easdale Subgroup is indicated by a rapid change to finer grained sediments showing features of deep-water sedimentation and turbidity currents, with a general tendency to become finer towards the east and north-east. Fault-controlled, steep shelf-slope sedimentation resulted in local incursions of very coarse-grained, mass-flow deposits, as seen in the Scarba Conglomerate Formation (Anderton, 1979, 1985). Marked along-strike facies changes indicate that, in contrast to the preceding Appin Group, deposition occurred within more-discrete fault-bounded marginal basins having a general north-east trend (Anderton, 1985, 1988). The interbasinal highs are marked by local thinning, facies changes, erosion or non-deposition; they commonly coincide with long-lasting major lineaments trending north-east or north-west (Fettes et al., 1986; Graham, 1986). Syngenetic barium and base-metal mineralization has been attributed to ponding of exhalative brines in basins adjacent to active faults, which in turn controlled sea-water infiltration into buried sediments (Coats et al., 1984).

Widespread shallow-water shelf and tidal-flat sedimentation returned during late Easdale Subgroup times due to infilling of the local basins with fine-grained sediment to produce, for example, the Craignish and Ardrishaig phyllites and the Ben Lawers Schist. Although thin beds of volcaniclastic rock and local pillow lavas do occur in the Islay Subgroup, the first major volcanic episode in the Dalradian succession occurred in late Easdale Subgroup time and is represented by the Farragon Volcanic Formation, the Meall Dubh Metabasite Formation and the Delnadamph Volcanic Member. Taken together with the tendency towards unstable sedimentation and the slightly earlier syngenetic mineralization, these occurrences point to an increased extension in this part of Rodinia, possibly heralding the break up that led eventually to the separation of Laurentia and formation of the Iapetus Ocean (Fettes et al., 2011).

1.4.3.3 Crinan Subgroup

Throughout its outcrop, this subgroup exhibits a relatively simple stratigraphy in which many local successions are dominated by a single thick formation. Sedimentation was generally of deep-water, turbiditic type with marked variations in thickness and diachronous facies changes. The Crinan Grit Formation of the South-west Grampian Highlands passes north-eastwards into the generally thinner-bedded and finer-grained Ben Lui Schist Formation of the Central Grampian Highlands. In most areas of the North-east Grampian Highlands, the Crinan and Tayvallich subgroups are difficult to separate; they crop out in a broad horseshoe of migmatitic and gneissose psammite and semipelite around the Turriff Syncline.

The Crinan Grit Formation, and its local equivalent the Ardmore Formation on the south-east coast of Islay, constitute the whole subgroup on both limbs of the Loch Awe Syncline (Knill, 1959; Roberts, 1966a; Borradaile, 1973, 1979). The base is probably diachronous and is interbedded locally with lithologies indistinguishable from the underlying Craignish Phyllite. In some areas a local unconformity is implied by a basal conglomerate. On the north-western limb of the syncline and around the head of Loch Awe, the subgroup consists of 100–550m of white quartzite with thin beds of gritty psammite and interbedded green or grey phyllitic semipelite. Locally, thin metalimestones and slaty metamudstone units are present. A lens of pale green metatuff occurs near Craignish, and the overlying psammites have a more chloritic matrix, reflecting a volcaniclastic component. On the south-eastern limb of the syncline the thickness increases to over 3000m with the incoming of many thick-bedded psammites containing angular fragments of detrital feldspar, mica and carbonate. Pebbly and conglomeratic quartzites increase towards the top of the formation and in places become the dominant lithology.

On the south-eastern limb of the Ardrishaig Anticline, the subgroup crops out in a continuous strip along the west coast of Kintyre and through Loch Fyne to the Central Grampian Highlands. In south Knapdale the lower part of the subgroup is represented by the Upper Erins Quartzite Formation, which lenses out to the north-east. Grey-green, phyllitic pelites and semipelites occur locally and pebbly quartzites, usually graded, become prevalent towards the top of the formation (Roberts, 1966a). Above the Upper Erins Quartzite, the Stonefield Schist Formation consists of schistose semipelite and pelite, with subordinate psammite and quartzite and numerous lenticular quartzose metalimestones. When traced north-eastwards, across Loch Fyne, this formation becomes dominantly pelitic and passes north-eastwards into the Ben Lui Schist Formation of the Central Grampian Highlands.

Around Tyndrum, the Ben Challum Quartzite Formation marks the base of the subgroup locally. It consists of up to 500m of quartzite and micaceous semipelite with minor amphibolite of possible volcaniclastic origin and two bands of low-grade stratabound pyrite-chalcopyrite-sphalerite mineralization. There is a possible correlation with a zone of similar sulphide mineralization in the Upper Erins Quartzite in the Meall Mor area. The Ben Lui Schist Formation crops out continuously from Loch Fyne to Ben Vuirich, notably over wide areas on both limbs of the Ben Lawers Synform and including structural outliers on the lower limb of the Tay Nappe around Lochearnhead. The formation consists mainly of garnetiferous semipelite and graded psammite, although sparse hornblende-schist and impersistent beds of metalimestone have been recorded in the Killin area (Johnstone and Smith, 1965) and near Pitlochry (Sturt, 1961). At the base of the formation, between Tyndrum and Glen Lyon, is a zone up to 50m thick containing lenticular bands of talcose or chloritic soft green schist containing lenses and pods of dolomite and magnesite with minor amounts of chromium, copper and nickel minerals. They probably represent sediments derived from the erosion of local ultramafic rocks, possibly of ophiolitic origin (Fortey and Smith, 1986; Power and Pirrie, 2000) but more-likely derived from subcontinental lithospheric mantle exposed during the development of extensional basins (Chew, 2001).

To the north-east of Pitlochry, the metamorphic grade of the Ben Lui Schist increases and the unit grades into migmatites that constitute the Duchray Hill Gneiss Member of Glen Isla and the equivalent Queen’s Hill Gneiss Formation of Deeside (Read, 1927, 1928; see Chapter 6, Introduction). There, the formation also includes numerous bands of gneissose amphibolite that probably represent basic intrusions.

Generally gneissose and migmatitic semipelitic rocks that form a broad horseshoe outcrop around the Turriff Syncline, from mid-Donside to Fraserburgh have been assigned to the higher parts of the Argyll Group (Read, 1955; Harris and Pitcher, 1975; Harris et al., 1994; Stephenson and Gould, 1995). Most are assumed to represent the Crinan Subgroup, although the Easdale and Tayvallich subgroups occur in some areas. Schistose and gneissose psammites and pelites of the Craigievar Formation in mid-Donside are considered to be equivalent to the psammites and semipelites of the Aberdeen Formation to the east and north-east (Munro, 1986). In the lower Ythan valley, gneisses of the Ellon Formation are distinguished from those of the Aberdeen Formation by their lack of regular lithological banding, their poor fissility and a foliated, streaky appearance (Read, 1952; Munro, 1986). To the north and east they grade into the structurally overlying Stuartfield ‘division’ of semipelite, pelite, psammite and metagreywacke. The upper part of this ‘division’ has a more coherent stratigraphy involving massive channel quartzites up to 500m thick (e.g. the Mormond Hill Quartzite Member) and calcareous beds, and is termed the Strichen Formation. The calcareous beds have been taken to indicate that the formation spans the boundary between the Crinan and Tayvallich subgroups (Kneller, 1988).

Still farther north is the Inzie Head Gneiss Formation, which exhibits a wide range of relict metasedimentary lithologies with a general migmatitic appearance(Read and Farquhar, 1956), whereas on the west side of the Turriff Syncline, the Cowhythe Psammite Formation generally preserves original compositional banding with migmatization concentrated mainly in the semipelitic units (Read, 1923). Both of these formations are well represented by GCR sites described in Chapter 6.

The Crinan Subgroup marks a rapid shelf-deepening and basin-forming event. Well-developed grading, channelling and large-scale slump folding in the Crinan Grits are indicative of proximal turbidite deposits, which Anderton (1985) considered were deposited in submarine-fan channels flowing axially in a major NE-trending basin along the line of what is now the Loch Awe Syncline. As the grain size in the subgroup as a whole fines from south-west to north-east, a major input from the south-west seems likely. But the grain size also becomes finer from north-west to south-east, so that the Upper Erins Quartzite and the Stonefield Schist/Ben Lui Schist are all finer grained than the Crinan Grits. Many authors have regarded the finer grained beds as more distal turbidites, deposited in the same basin as the Crinan Grits (e.g. Harris et al., 1978), but Anderton (1985) suggested that they were deposited in a separate parallel basin. Volcanic activity is not so evident as in the preceding Easdale Subgroup but reworked volcaniclastic deposits occur in the lower part of the subgroup and ultramafic rocks indicate the possible exposure of sub-continental lithospheric mantle somewhere in the basin (Chew, 2001). Slope instability with contemporaneous earthquakes as a result of intrabasinal faulting is suggested by the soft-sediment structures preserved in the Crinan Grits.

1.4.3.4 Tayvallich Subgroup

This subgroup is characterized by carbonate sequences, accompanied in the South-west Grampian Highlands by extensive basic volcanic rocks and subvolcanic sills. The Tayvallich Limestone and its lateral equivalents, the Loch Tay Limestone and Deeside Limestone, together constitute one of the most persistent marker bands of the Dalradian Supergroup, stretching from Donegal to the Banchory area (Gower, 1973). The youngest limestone of the Shetland Dalradian succession, the Laxfirth Limestone, is a possible equivalent. The basic volcanicity resulted in the thickest developments of volcanic and subvolcanic rocks seen in the Dalradian succession. However, this volcanicity is less extensive laterally than that of earlier, Easdale Subgroup times and Fettes et al. (2011) have speculated that it might have been associated with localized pull-apart basins on the rifted Laurentian margin, with the main full-scale rifting of the continental crust occurring outboard of the preserved Dalradian sequences.

In the Loch Awe Syncline and the subsidiary Tayvallich Syncline to the south-west, the Tayvallich Slate and Limestone Formation exhibits marked facies changes from north-west to south-east comparable to those in the underlying Crinan Subgroup. The overall thickness varies considerably, mainly due to variations in the amount of sedimentary and volcanic material. The formation reaches a maximum of 1200m, but the total thickness of metalimestone is relatively constant, at about 100m. On the north-western limb of the Loch Awe Syncline ooidal and graded, gritty metalimestones are interbedded with graphitic metamudstones. These lithologies probably represent shelf sedimentation. Coarse, slumped limestone-breccias, metaconglomerates and gritty metasandstones, all suggestive of an unstable shelf margin, attain maximum development along the axial zone of the syncline. On the south-eastern limb thinner lenses of limestone-breccia and metaconglomerate are interbedded with turbiditic metasandstone (Knill, 1963). Around the north-eastern closure of the Loch Awe Syncline, the upper part of the Tayvallich Slate and Limestone Formation includes the Kilchrenan Grit and the Kilchrenan Conglomerate members(Borradaile, 1973, 1977). The latter is a matrix-supported ‘boulder bed’, up to 30m thick, consisting of well-rounded quartzite boulders in a matrix of gritty black metamudstone. It has been interpreted as a slump conglomerate (Kilburn et al., 1965). In the south and west, the upper beds of metalimestone include several layers of volcaniclastic debris, including fragments of pillows, set in a carbonate matrix and the transition into the overlying Tayvallich Volcanic Formation is a complex interdigitation of metalimestone, clastic metasedimentary rocks and metavolcanic rocks.

Metavolcanic rocks occupy much of the core of the Loch Awe Syncline, with a smaller outcrop in the core of the Tayvallich Syncline. The Tayvallich Volcanic Formation consists of up to 2000m of commonly pillowed basic lava, hyaloclastite and a variety of volcaniclastic volcanic rocks (Borradaile, 1973; Graham, 1976). Extrusion of the main volcanic pile was clearly sub-marine but away from the main centre of volcanicity, which corresponds to the axis of the Loch Awe Syncline, ash-fall tuffs suggest that volcanism evolved into a subaerial environment. Felsic tuffs have yielded U-Pb zircon ages of 601 ± 4 Ma (Dempster et al., 2002).

The volcaniclastic rocks include breccias and waterlain pebbly deposits such as the Loch na Cille ‘Boulder Bed’ of the Tayvallich Peninsula (Gower, 1977). The possibility of a major stratigraphical break at this horizon has been proposed by Prave (1999) and Alsop et al. (2000) who pointed to the presence of cleaved clasts of metsedimentary rock with up to two deformational fabrics. Alternatively the clasts could merely represent the erosion of a significantly older metasedimentary source. This boulder bed has also been correlated tentatively with a younger tillite that has been recognized in the Dalradian of Donegal, which could represent the Varanger glaciations at c. 620–590 Ma (Condon and Prave, 2000).

Basic sills intrude the whole succession from the Craignish and Ardrishaig phyllites upwards; their total thickness attains 3000m in places. These were originally thought to be contemporaneous with the lavas (Borradaile, 1973; Graham, 1976). However, in places they can be seen to have broken through limestone and developed pillowed margins within the unconsolidated sediments and overlying water, so that they might well be near-contemporaneous with their host sediments (Graham and Borradaile, 1984; Graham, 1986). A suite of NW-trending metabasalt dykes on Jura could represent feeders to the lavas and sills (Graham and Borradaile, 1984).

To the south-east of the Ardrishaig Anticline the subgroup is represented by the Loch Tay Limestone which maintains a thickness of around 100m along strike from Campbeltown to Glen Doll (Bailey, 1925; Elles, 1926; Johnstone and Smith, 1965; McCallien, 1929; Roberts, 1966a; Crane et al., 2002). Thinner developments are also present to the south-east of the main outcrop, at the base of structural outliers of Ben Lui Schist in the inverted limb of the Tay Nappe around Lochearnhead (Johnstone and Smith, 1965; Mendum and Fettes, 1985; Watkins, 1984). Throughout its strike length the formation consists of thick beds of crystalline metalimestone, locally with various schistose semipelites, calcsilicate rocks and psammites. Thin, graphitic metamudstones and metagreywackes are present locally and grading in all lithologies suggests a distal turbidite origin. Intrusive basic meta-igneous rocks are present throughout the outcrop, although there is no continuous volcanic sequence comparable with the Tayvallich Volcanic Formation.

The Loch Tay Limestone can be traced almost continuously north-eastwards as far as Glen Doll, beyond which the Water of Tanar Limestone Formation and Deeside Limestone Formation are probably lateral equivalents. The latter formations are composed largely of calcareous semipelite and psammite with calcsilicate rocks, amphbolite and only thin beds of impure metalimestone. Around the head of Glen Esk and in Middle Deeside the calcareous rocks are overlain by the dominantly psammitic Tarfside Psammite Formation, which is also assigned to the Tayvallich Subgroup on account of locally abundant calcsilicate and amphibolite beds (Harte, 1979).

North of Deeside, the gneissose units that are assigned generally to the Crinan Subgroup probably include some Tayvallich Subgroup rocks, as is indicated by the presence of calcsilicate and metalimestone beds. Notable examples are the calcareous upper parts of the Strichen Formationand its lateral equivalent on the north coast, the Kinnairds Head Formation. In the coast section to the west of the Turriff Syncline, the Tayvallich Subgroup is well defined by a 1200m-thick sequence, termed the Boyne Limestone Formation, which includes the 200m-thick Boyne Castle Limestone Member (Read, 1923; Sutton and Watson, 1955). Inland, the metalimestones can only be traced for some 2.5km, making it difficult to define the Argyll–Southern Highland group boundary farther south.

However, in the Cabrach area, basaltic metavolcanic rocks occur within a turbiditic sequence of pelite, semipelite and pebbly psammite termed the Blackwater Formation (MacGregor and Roberts, 1963; Macdonald et al., 2005). Since the formation appears to pass upwards into Southern Highland Group lithologies, the volcanic rocks have been correlated tentatively with the Tayvallich Volcanic Formation, with which they share some tholeiitic geochemical characteristics.

The change from the coarse siliceous Crinan Grits to massive carbonate beds with dispersed clastic carbonate containing siliciclastic material implies a major change and reduction of sediment source. Rapid lateral variations in facies and thickness and intercalations of graphitic metamudstone, associated with unconformities, overstep relations and pebbly beds indicate deposition of the metalimestones by low-density turbidity currents. They were therefore probably deposited in pre-existing Crinan Subgroup basins following a reduction in the supply of clastic sediment to the fringing shelves (Anderton, 1985).

1.4.4 Southern Highland Group

The Southern Highland Group consists of a c. 4 km-thick turbiditic sequence made up mainly of coarse-grained, poorly sorted metagreywacke with subordinate fine-grained metamudstone. Minor intercalations of metalimestone occur locally. Volcaniclastic ‘green beds’ are widespread at several levels, notably in the basal part of the group, there is one thick local development of basic lavas, and widespread thin ‘brown beds’ within metamudstones of the Highland Border have been interpreted as more-evolved ash-fall tuffs. The metasedimentary rocks are markedly more chloritic than those of the underlying Argyll Group, partly due to the generally lower metamorphic grade, but probably also reflecting the volcaniclastic input. They are also more feldspathic, with high-grade metamorphic and granitic rock fragments, suggesting a less-mature source area.

The group can be traced from County Mayo in western Ireland to the North-east Grampian Highlands. Much of its outcrop occurs in areas of gentle regional dip such as the Flat Belt of the Tay Nappe, so that its outcrop covers a wide area of some 4900 km². The main outcrop is a belt up to 34 km wide, extending along the whole south-eastern edge of the Grampian Highlands from the Mull of Kintyre to Stonehaven and Aberdeen (Chapter 4). A small outlier occurs in the core of the Loch Awe Syncline, and in the North-east Grampian Highlands the group occupies the broad core of the Turriff Syncline and a small outlier on the east coast around Collieston.

Although several local successions have been established, detailed correlations are seldom possible over any distance, as the general lateral and vertical persistence of turbidite facies means that there are few reliable stratigraphical marker beds. Green beds are useful locally, but metamudstone beds are probably highly diachronous. Correlations are further complicated by across-strike changes in metamorphic grade, which significantly alter the appearance of the rocks and have led to different names for units that are probably equivalent stratigraphically. Consequently no subgroups are recognized and in terms of thickness and uniformity of facies the whole group is comparable to a subgroup in other parts of the Dalradian succession.

In the core of the Loch Awe Syncline, the Tayvallich Volcanic Formation is overlain by up to 1100m of chloritic gritty metasandstone and metamudstone, calcareous in parts, which together comprise the Loch Avich Grit Formation. The succeeding Loch Avich Lavas Formationconsists of 300 to 500m of basaltic pillow lavas with no significant sedimentary intercalations (Borradaile, 1973).

In the main outcrop of the group south-west of Loch Lomond, the Loch Tay Limestone is succeeded by a typical, predominantly metagreywacke sequence (McCallien, 1929; Roberts, 1966a). The basal part of this sequence is the Glen Sluan Schist Formation,consisting of up to 500m of pelitic turbidites, lithologically similar to the main part of the succession but separated from it by the Green Beds Formation. Here, the latter form a well-defined stratigraphical unit up to 1000m thick, consisting of metagreywacke and fine-grained quartzite, interbedded with the well-foliated, green schists containing abundant chlorite and epidote. These lithologies become hornblende-schist at higher metamorphic grades. Lenses of obvious detrital material, pebbles of quartz and graded bedding indicate a sedimentary origin. The beds probably represent an influx of detrital basic volcanic material, a feature reflected in their chemistry and mineralogy (Phillips, 1930; van de Kamp, 1970). Mafic sheets within the formation represent contemporaneous shallow intrusions. The volcaniclastic components are interpreted as recording in part, the erosion of the nearby Loch Avich lavas, but could also have resulted from contemporaneous volcanism in the hinterland (Pickett et al., 2006). This interpretation is not however strongly supported by the detrital zircon data, which are dominated by Archaean detritus (Cawood et al. 2003). Above the Green Beds in this area is the Beinn Bheula Schist Formation, which consists predominantly of fine-grained metagreywacke with lenticular developments of pebbly metasandstone and green metamudstone.

In the South-west Grampian Highlands, most of the Southern Highland Group outcrop lies on the inverted limb of the Tay Nappe, which is folded into the late broad Cowal Antiform. Consequently, the lower parts of the group are also exposed on the south-eastern limb of that fold. There the Beinn Bheula Schist passes stratigraphically downwards into the Dunoon Phyllite Formation (equivalent to the Glen Sluan Schist), with only local developments of green beds that are too sparse to warrant a formal formation. The Dunoon Phyllites lie within the complex hinge-zone of the Tay Nappe, which takes the form of a downward-facing (i.e. synformal) anticline known as the Aberfoyle Anticline. Consequently the rocks that crop out farther to the south-east, between the Dunoon Phyllites and the Highland Boundary Fault-zone, are regarded as younger than the phyllites and broadly equivalent to the Beinn Bheula Schist (Anderson, 1947a; Roberts, 1966a; Paterson et al., 1990). In Cowal and Bute, these pebbly metasandstones with metaconglomerate and local metalimestone and metamudstone were fotrmerly known as the Bullrock Greywacke and Innellan 'group' and have now been united as the St Ninian Formation. In north Arran a similar synformal anticline structure exists, with the Loch Ranza Slate Formation occupying the core and the North Sannox Grits Formation on each limb.

To the north-east of Loch Lomond, boundaries between metagreywacke and metamudstone units are demonstrably diachronous. Hence, although the metamudstones generally occur in the cores of downward-facing anticlines, neither the local lithostratigraphical units nor the fold axes are likely to be continuous along the whole length of the Highland Border. Metamudstone in the core of the obvious main anticline in the Aberfoyle area is assigned to the Aberfoyle Slate Formation, whereas other metamudstone units, there and in the Callander area, are now recognized to be facies variations within younger metagreywacke. To the north-west of the Aberfoyle Anticline, throughout this area, the Southern Highland Group consists predominantly of rocks of turbiditic greywacke facies, most of which are assigned to the Ben Ledi Grit Formation (e.g. Mendum and Fettes, 1985). Green beds and basic intrusions are widespread at various horizons in the lower part of this formation, and in the area of the Trossachs these define a separate Loch Katrine Volcaniclastic Formation, underlain and overlain respectively by the Ardnandave and Creag Innich sandstone formations. Detrital sodic feldspar is common in many of the metasandstones between Loch Fyne and Loch Tay and a redistribution of sodium during metamorphism has given rise to many porphyroblastic albite-schists in the metamudstone units (Bowes and Convery, 1966; Watkins, 1983).

Correlation of lithostratigraphical units in the Highland Border Steep Belt to the south-east of the Aberfoyle Anticline, is more complex than previously supposed (Harris, 1962, 1972; Harris and Fettes, 1972; Tanner, 1995, 1997; Bluck and Ingham, 1997; Tanner and Sutherland, 2007). On that limb, the whole of the metagreywacke sequence was formerly known as the ‘Leny Grits’. Most of it is now assigned to the Ben Ledi Grit Formation, but in the Callander area, the upper part is the Keltie Water Grit Formation, which includes the Leny Limestone with its fossils of undoubted latest Early Cambrian age. This formation is now accepted to be in stratigraphical and structural continuity with the Ben Ledi Grits and thus provides the only reliable biostratigraphical age for the top of Dalradian Supergroup. A full discussion of the long-term debate concerning the affinities and significance of this unit and its relationship to adjoining fault slices of the Highland Border Complex can be found in Chapter 4.

To the north-east of Callander the south-eastern limb and hinge of the Aberfoyle Anticline are cut-out by the Highland Boundary Fault. Consequently in the Dunkeld area, the whole sequence in the Highland Border Steep Belt youngs to the north-west. Continuity of the Birnam Slate and Grit Formation (oldest) and the Dunkeld Grit Formation (youngest) with the Ben Ledi Grits is difficult to confirm and they could represent a higher stratigraphical level. In a similar structural position to the north-east are the Cairn Gibbs Psammite Formation of the Glen Shee area and the Rottal Schist Formation of Glen Clova but there too, stratigraphical continuity is uncertain.

To the east of the Loch Tay Fault, the lower parts of the Southern Highland Group, consisting dominantly of more-pelitic metagreywacke sequences, are exposed only along the north-western edge of its outcrop. Between Loch Tay and Glen Shee, these constitute the Pitlochry Schist Formation (Treagus, 2000), from Glen Shee eastwards the equivalent unit is the Mount Blair Psammmite and Semipelite Formation (Crane et al., 2002), and in the Glen Clova area this basal unit has been termed the Longshank Gneiss Formation. Green beds are abundant at several stratigraphical levels but are notably absent north-east from Glen Clova. Soft, fine-grained, amorphous ‘brown beds’, up to 10 cm thick, have been identified at several localities over a distance of 96 km along strike, from the east side of Loch Lomond to near Fettercairn (Batchelor, 2004a, 2004b). Their chemistry and mineralogy suggests that they originated as mildly alkaline rhyolitic or trachytic tuff from a relatively distant source.

Farther to the north-east, stratigraphical continuity is interrupted by an extensive right-way-up sequence, which was assigned by Harte (1979) to the Tarfside Nappe, a major recumbent structure below the Tay Nappe. In that sequence a lower more-pelitic unit, the Glen Effock Schist Formation, passes upwards into the higher Glen Lethnot Grit Formation, characterised by beds of pebbly psammite, and hence a broad correlation is suggested with the Tay Nappe successions. On the east coast, north of Stonehaven, the sequence reverts to being upside down and has been assigned to the Glen Lethnot Grit on lithological criteria.

A transition from lagoonal calcareous silts and muds of the Argyll Group into turbidites that define the Southern Highland Group is well seen on both limbs of the Turriff Syncline. The turbiditic psammites with subordinate semipelites and pelites are referred to as the Whitehills Grit Formation on the western limb (Read, 1923; Sutton and Watson, 1955) and the Rosehearty and Methlick formationson the eastern limb (Read and Farquhar, 1956). In the core of the syncline the Macduff Formation is a finer grained, more-distal turbidite facies (Sutton and Watson, 1955). A more-persistent semipelitic facies to the south-west has been termed the Clashindarroch Formation. Around the closure of the syncline, in the Correen Hills, the Southern Highland Group is represented by the Suie Hill Formation, which consists dominantly of semipelite and gritty psammite with prominent pelite units. An eastward-younging sequence of turbiditic rocks on the east coast, north of Aberdeen, is termed theCollieston Formation and is assigned to the Southern Highland Group (Read and Farquhar, 1956; Munro, 1986). This predominantly psammitic graded sequence includes characteristic ‘knotted‘ pelite containing andalusite and cordierite.

Some exotic boulders and pebbles in the higher exposed part of the Macduff Formation have been attributed to ice-rafting or debris flows linked to marine tills (Sutton and Watson, 1954; Hambrey and Waddams, 1981; Stoker et al., 1999). Correlations with various glacial periods, some as young as Ordovician, have been suggested, though a correlation with the glacigenic deposit in the Southern Highland Group of Donegal (Condon and Prave, 2000) and the Varanger tillites (accepting their revised age of 620–590 Ma) seems to be the most likely on present evidence (see below).

In general character the clastic sediments of the Southern Highland Group are similar to those of the preceding Argyll Group and sedimentary structures are commonly preserved in immature turbidite-dominated sedimentary facies associations (Harris et al., 1978; Anderton, 1980, 1985; Burt, 2002; Pickett et al., 2006). Metagreywacke and metasandstone are typically coarse-grained, poorly sorted, graded, and are commonly composite. Bouma sequences have been identified in thin-bedded turbidites of the Macduff Formation. Some of the channels are asymmetrical with steep banks showing evidence of slumping and the bases of the coarser metasandstones commonly display large saucer-shaped loadcasts. Minor silty beds show small-scale cross-bedding.

These sediments mark a return to rapid basin deepening, which persistently stayed ahead of the sedimentary and volcanic fill. The sedimentary structures imply that the coarse-grained sediments were probably deposited in slope-apron or ramp settings with channels on the lower slopes and inner zones of major deep-water sub-marine fans, with the finer sediments being laid down as overbank deposits or as outer fan facies. Palaeocurrent directions and facies changes in the sequence all suggest a dominant flow towards the south-east with minor north-eastward and south-westward axial flows. Clasts of feldspar, high-grade metamorphic rocks and granitic rocks are more abundant than in earlier Dalradian turbidites, suggesting a newly emerged, less mature source area (Harris and Pitcher, 1975; Harris et al., 1978), and Plant et al. (1984) and Anderton (1985) suggested that this could be due to stripping of a cover sequence in a north-western source area to reveal a granitoid basement.

The Loch Avich Lavas, the ‘green beds’ and the ‘brown beds’ represent a final phase of magmatism in the Dalradian basins, probably erupting from volcanic centres in the Loch Awe area, close to the source of the earlier Tayvallich lavas.

1.4.5 Units of uncertain stratigraphical affinity

Two metasedimentary units of uncertain affinity crop out on the Isle of Islay, where they are separated by the NNE-trending Loch Gruinart Fault, a possible splay from the Great Glen Fault (see Chapter 2). Both the Colonsay Group and the Bowmore Sandstone Group are thought to have been deformed and metamorphosed during the Grampian Event but they are separated tectonically from the Dalradian succession and direct correlations on lithological grounds are equivocal. There are significant differences between the early deformation histories of the Colonsay Group and the Dalradian, adding to speculation that western Islay and Colonsay might represent a separate terrane (e.g. Bentley et al., 1988; Muir et al., 1992).

1.4.5.1 Colonsay Group

The Colonsay Group crops out on Colonsay, Oronsay and the Rhinns of Islay. It consists of a 5.0–5.5km-thick sequence of greenschist-facies, strongly deformed metasandstone and phyllitic metamudstone, with minor calcareous beds. The succession youngs generally towards the north-west and Stewart (1962a) and Stewart and Hackman (1973) divided the sequence into 18 lithostratigraphical units; the lower 10 on Islay and the upper 8 on Oronsay and Colonsay. It is possible that there is a slight overlap between the successions on Islay and Oronsay but Bentley (1988) thought that there might be a stratigraphical gap of up to a kilometre in thickness. Subsequently, Muir et al. (1995) have recognized that four formations in the lower part of Stewart and Hackman’s Islay succession are in fact two formations repeated by folding, but they also identified a previously unrecorded formation (the Octofad Sandstone Formation) on the south-east coast of the Rhinns, which is the lowest preserved part of the group.

The lowest 650m of the succession on Islay consist of feldspathic metasandstone and metamudstone, interpreted as representing delta-top sheet sands and interdistributary muds. These pass upwards on Islay into quartz-rich metagreywacke and metamudstone suggesting deeper water, delta-slope turbidite accumulation. These become increasingly distal as the succession youngs on to Oronsay and Colonsay, possibly reflecting basin deepening. The upper part of the succession on Colonsay shows a change back towards shallow-water, marine, mainly siliciclastic sedimentation but with several calcareous developments, notably the 1–5m-thick (dolomitic) Colonsay Limestone. The topmost units possibly reflect the onset of renewed deepening. Several thin volcaniclastic layers have been identified from mineralogical and geochemical criteria, mainly in the upper part of the succession on Colonsay, by Batchelor (2011).

On Islay, the Colonsay Group is in sheared contact with gneisses of the Rhinns Complex. A perceived coarsening of facies and the presence of local conglomerates towards the contact led Wilkinson (1907) and Bentley (1988) to suggest that the contact is a sheared unconformity. However, for much of its length in the western and central Rhinns, the contact is marked by a 3–4 m-wide mylonite zone that cuts discordantly across four formations in the lower part of the group. It is therefore possible that an unknown thickness of the original sequence might be missing and Stewart and Hackman (1973) interpreted the contact as a major tectonic dislocation, which they termed the Bruichladdich Slide. However, on the south-east coast of the Rhinns, Muir et al. (1995) identified complex basement–cover interslicing, which extends into the Bruachladdich area, and hence they proposed that the simpler, single plane of dislocation in the central and western Rhinns would be more appropriately named the Kilchiaran Shear-zone. The contact at the north-eastern end of Colonsay was interpreted as a sheared unconformity by both Bentley (1988) and Fitches and Maltman (1984). However, the conglomeratic cover rocks there belong to the Kilchattan Formation, about 4km above the base of the Colonsay Group succession on Islay, which would require a highly uneven basement topography.

For many years the Rhinns basement was thought to be Lewisian, and consequently the overlying Colonsay Group had been correlated with the Torridonian, until this was challenged by Stewart (1969, 1975). The presence of metalimestone in the upper part of the group suggested to Litherland (in Stewart and Hackman, 1973) that there could be a correlation with the Appin Group and this was supported by Rock (1985), who suggested that the chemistry of the Colonsay Limestone is similar to that of the Ballachulish Limestone. In fact the uppermost formations on Colonsay are lithologically similar to the laterally persistent Leven Schist to Appin Quartzite sequence of the Ballachulish type area. Such a correlation would imply that the underlying parts of the Colonsay Group are lateral equivalents of the Lochaber Subgroup and the Grampian Group. The main objection to a correlation with any part of the Dalradian has been on structural grounds, since the first phase of deformation seen in Colonsay Group rocks is not apparent in the undoubted Dalradian rocks of Islay (Borradaile, 1979; Fitches and Maltman, 1984). The matter is still not fully resolved but the distribution of U-Pb ages in detrital zircon and titanite does suggest a correlation with the Grampian Group (McAteer et al., 2010).

1.4.5.2 Bowmore Sandstone Group

The Bowmore Sandstone Group is separated from both the Colonsay Group of western Islay and the Dalradian succession of eastern Islay by tectonic dislocations. The predominantly grey-brown, feldspathic sandstones have been divided into two formations, each exceeding 2km in thickness (Amos, 1960). The lower, Laggan Sandstone Formation, consists of fine- to medium-grained sandstone with silty partings, and the upper, Blackrock Grit Formation, is mainly coarse-grained sandstone with pebbly beds. The rocks are tightly folded but the locally developed tectonic fabrics are weak and metamorphism is slight. However, younging indicators are rare.

The group has been correlated variously with the Moine, the Torridonian and the Dalradian (Fitches and Maltman, 1984, table 1). It is separated from an overlying Dalradian succession by the Loch Skerrols Thrust, and those workers who have regarded this as a major structure, possibly equivalent to the Moine Thrust, have correlated the Bowmore Sandstone mainly with the Torridonian, by analogy with the structure of the North-west Highlands (e.g. Johnstone, 1966; Stewart, 1969, 1975). Others have regarded the Loch Skerrols Thrust as a structure of local importance only and hence have proposed that the Bowmore Sandstone is more likely to be part of the Dalradian succession. Fitches and Maltman (1984) argued that it is the lateral equivalent of the Crinan Grits on the lower, inverted limb of the Islay Anticline.

1.5 Structure of the Grampian Highlands

D. Stephenson

Current structural interpretations of the Grampian Highlands are still based upon those proposed by C.T. Clough and E.B. Bailey, which drew upon the results of the primary mapping by the Geological Survey. In a series of papers, from 1910 to 1938, Bailey demonstrated that the rocks of the South-west and Central Grampian Highlands are disposed in large, Alpine-scale recumbent folds. He proposed that the long limbs of many of these folds are partly replaced by low-angled faults, termed ‘slides’, with postulated movements of several kilometres. Most of the major fold structures originally identified and named by Bailey are still recognized, and many of his slides exist as complex low-angled high-strain zones, some with tectonic dislocation and/or stratigraphical attenuation, but some have lost their original significance as major tectonic and stratigraphical boundaries.

Subsequent refinements to the Bailey model have involved careful consideration of the relative ages of the various structures, their relationships with each other, and hence the overall structural history of the area, which involves several phases of deformation. This history has been painstakingly pieced together and refined by a multitude of workers. Several alternative structural models have emerged but no single hypothesis can account satisfactorily for all of the observed features (see Stephenson and Gould, 1995 for examples). Various controversies have raged over the years but there has also been a remarkable amount of agreement and consensus on many aspects.

The overall structure of the Grampian Highlands is best illustrated by the dissected block diagram of (Figure 1.7), which is based upon an original published by Peter Thomas in 1979 and not bettered since, though it was modified slightly by Stephenson and Gould (1995) and reproduced by Strachan et al. (2002).

The large-scale structures are most easily described and understood with reference to those of the South-west Grampian Highlands ((Figure 1.7), block A), where the relatively simple fold geometry remains essentially as described first by Bailey (1910, 1922, 1934), despite subsequent re-interpretation by Roberts and Treagus (1977c), Hickman (1978) and Litherland (1982). There, the major folds are seen to diverge on either side of a central Loch Awe Syncline which, although composite, is characterized by open to close, upright, upward-facing folds that are generally regarded as F1. On either side of this syncline, the early folds become progressively overturned and verge to the north-west and south-east. To the south-east, the tight Ardrishaig Anticline is interpreted as the core of a large SE-facing nappe, the Tay Nappe, which dominates the overall structure of the south-eastern part of the Grampian Highlands throughout its strike length.

Throughout the Central Grampian Highlands, the core of the Tay Nappe is obscured within a zone of steeply dipping strata resulting from the refolding of co-axial F1 and F2 folds by near-upright F3 structures (the Ben Lui Fold-complex and the Tummel and Cairnwell steep belts). To the north-west of this zone, generally SE-dipping Appin and Argyll group strata are separated from the structurally underlying Grampian Group by a high-strain zone, which includes several planes of dislocation and has been termed the Boundary Slide. Beneath the Boundary Slide, major early folds still face south-east (e.g. those formerly known as the Glen Orchy and Atholl nappes) but farther to the north-west, a fundamental change in facing direction lies in the region of a 4km-wide zone of upright, isoclinal folds known as theGeal-charn–Ossian Steep Belt ((Figure 1.7), blocks C and D).

To the north-west of the Loch Awe Syncline is a stack of NW-facing early folds, including the Islay Anticline, Appin Syncline, Beinn Donn Syncline and Ballachulish Syncline (Bailey, 1934; Roberts and Treagus, 1977c). Similar NW-facing folds occur to the west of the Geal-charn–Ossian Steep Belt (Key et al., 1997) and there is reasonable continuity of overall structure between Islay and Glen Roy. To the west, these folds of mainly Appin and Argyll group rocks overlie the Grampian Group.

In the following sections, each major structural domain of the Grampian Highlands is described firstly in the south-west and then progressively through to the north-east, with reference to the labelled blocks of (Figure 1.7). The structure of the Dalradian terrane in the Shetland Islands is described in the introduction to Chapter 7.

1.5.1 The Tay Nappe

The Tay Nappe dominates the south-eastern part of the Grampian Terrane, in a 15–20 km-wide outcrop that extends from Kintyre to Stonehaven parallel to the Highland Boundary Fault (Figure 1.7), see also (Figure 4.4). Throughout much of its outcrop, the nappe is parallel-sided and flat-lying. The erosion level is such that most of the outcrop constitutes part of the inverted limb, so that stratigraphical sequences are inverted. Structures underlying this inverted limb are seen only in the area of the Angus glens, where a largely right-way-up sequence has been interpreted as a separate Tarfside Nappe (Harte, 1979). Close to the Highland Boundary Fault, the Tay Nappe is bent downwards by a large monoform known as the Highland Border Downbend to form the Highland Border Steep Belt (see below). The hinge-zone of the nappe thus becomes downward facing as a synformal anticline, recognized in the South-west Grampian Highlands as the Aberfoyle Anticline (Shackleton, 1958). The south-eastern limb of this anticline is the only place where the original upper limb of the Tay Nappe is preserved.

In the South-west Grampian Highlands the inverted limb of the Tay Nappe has the general form of a broad arch, known as the Cowal Antiform ((Figure 1.7), block A), although in some areas, at least, this late (?D4) antiform is more of an open chevron (see (Figure 4.4), section A–A′). To the north-west of the Cowal Antiform, the core of the nappe is brought down below the level of erosion to crop out as the Ardrishaig Anticline. Here the fold limbs, primary axial planes and associated cleavages all dip steeply to the north-west and constitute part of the Knapdale Steep Belt (Roberts, 1974). This is the only area in which the core of the Tay Nappe is exposed.

North-east from Cowal, the crest of the late antiform flattens and the inverted lower limb of the Tay Nappe forms the Flat Belt which dominates so much of the South-eastern Grampian Highlands ((Figure 1.7), blocks B, C and D). The width of the Flat Belt varies and is up to 18 km wide in Perthshire. There, Krabbendam et al. (1997) and Treagus (1999) have identified kilometre-scale isoclines and right-way-up sequences within the dominantly inverted limb. The north-western limit of the Flat Belt is defined by a zone of F1 and F2 folds with steeply SE-dipping axial surfaces, which must refold the hinge-zone of any major syncline originally underlying and complementary to the Tay Nappe (Roberts and Treagus, 1979). This zone is termed the Ben Lui Fold Complex in (Figure 1.7) block B (after Cummins and Shackleton, 1955) but the zone widens north-eastwards in blocks C and D with the addition of many tight, upright folds of later generation (F2 and F3) to form the Tummel Steep Belt (see below).

Farther to the north-east, the amplitude of the Tay Nappe is considerably reduced ((Figure 1.7), block E). In the Glen Esk area a broad late antiform, the Tarfside Culmination, exposes a wide zone of right-way-up strata, apparently beneath the Tay Nappe that has been assigned to the Tarfside Nappe. According to Harte (1979) the axial zone of the fold separating the two nappes has been replaced by a slide, the Glen Mark Slide. On the coast section north of Stonehaven, the Highland Border Downbend is well developed and a generally flat-lying inverted sequence is bent down and becomes overturned to the south-east, with downward-facing D1 structures (Booth, 1984; Harte et al., 1987) (see (Figure 4.33)).

To the north of Aberdeen the Tay Nappe can no longer be identified with any certainty ((Figure 1.7), block F). The rocks are mostly the right way up, possibly beneath the Tay Nappe, and it is not clear whether the inverted limb has been cut out by major thrusting, or whether, more simply, the nappe structures have a much smaller amplitude in this area (Harte et al., 1984). However, in the coast section around Collieston the beds are regionally inverted and there are abundant small- to large-scale recumbent, isoclinal F1 folds with a maximum amplitude of about 1km (Read and Farquhar, 1956; Mendum, 1987) (Figure 6.40). (Figure 6.41). These folds plunge gently to the north, face eastwards and may represent a subdued equivalent of the nappe (Ashcroft et al., 1984).

Later (F3 and/or F4) NE-trending folds and other structures on both a major and a minor scale were imposed subsequently on the Tay Nappe. In addition to the Highland Border Downbend, broad, upright folds affect the north-western part of the Flat Belt ((Figure 1.7), blocks B, C and D). Of these, the best developed are the Ben Lawers Synform (Treagus, 1964b, 2000) and its complementary, lower amplitude Loch Tay (or Ben More) Antiform to the south-east (Watkins, 1984; Harte et al., 1984; Treagus, 2000). The somewhat tighter Sron Mhor Syncline may be of the same generation but it is closely associated with steeper structures to the north and was probably initiated as an earlier structure. The Collieston F1 folds are also refolded by a set of near-coaxial, tight folds which post-date porphyroblast growth and hence are assigned to D3.

1.5.1.1 The Highland Border Steep Belt and The Hinge-zone of the Tay Nappe

The existence of a belt of steeply dipping rocks along the Highland Border has long been known and was implicit in the early discussions of the area (Henderson, 1938; Anderson, 1947a; Stringer, 1957). However, it was Shackleton (1958) who used both sedimentary structures and cleavage-bedding relationships to demonstrate that the Aberfoyle Anticline is a downward-facing synformal anticline and interpreted it as the closure of the Tay Nappe. Subsequently, many detailed local studies have been made of key areas (from south-west to north-east: Simpson and Wedden, 1974; Paterson et al., 1990; Mendum and Fettes, 1985; Stone, 1957; Harris, 1962, 1972; Harris and Fettes, 1972; Harte et al., 1984; Booth, 1984).

(Figure 4.4) shows a series of cross-sections along most of the Highland Border region. North-eastwards from the Cowal and Bute area, the Cowal Antiform passes into the Highland Border Downbend (or Monoform), a sharp monoclinal flexure, over which the Flat Belt of the Tay Nappe gives way to the Highland Border Steep Belt. North-east from Dunkeld, the steep belt is overturned so that sequences of formerly inverted rocks dip at around 60° to the north-west and are once more the right way up (Treagus et al., 1972; Harris and Fettes, 1972). The hinge-zone of the Tay Nappe was originally described as a single downward-facing anticline, but it has been shown subsequently to be more complex. For example, in the Ben Ledi area the hinge-zone consists of two major synforms, the Aberfoyle Anticline and the Benvane Synform, separated by the Ben Ledi Antiform (Mendum and Fettes, 1985). These structures are all inferred to be of D1 age, since minor D2 structures are not seen near the Highland Border in this area and only become overprinted on D1 farther to the north-west.

In the rocks of low metamorphic grade in the steep belt and in much of the adjacent parts of the Flat Belt, minor structures are very distinctive and are much influenced by lithology. Pelites develop axial planar slaty cleavages but in the psammites, which constitute most of the Southern Highland Group, spaced S1 pressure-solution cleavages fan around fold closures and are strongly refracted across lithological boundaries (see (Figure 4.26)). Subsequent deformation during D2 results in characteristic microlithons of S1 in the hinge-zones of minor F2 folds but flattens the spaced cleavage on more-highly strained limbs and induces a finely striped rock, so that it is difficult to recognize the multiple origin of the dominant foliation (Harris et al., 1976) (see (Figure 4.29)).

The Highland Border Downbend is not generally seen to refold D3 structures. However, Crane et al. (2002) described gentle NE-trending warps near Kirkmichael that do refold F3 folds and hence the downbend is confirmed as regional F4. Where downbend-related folds are present, they are usually accompanied by a strong crenulation cleavage, best seen in the more pelitic or strongly foliated rocks. This cleavage dips north-westwards at moderate to subvertical angles.

1.5.1.2 The Ben Lui Fold-complex, Tummel Steep Belt and Cairnwell Steep Belt

From Dalmally north-eastwards to Strathtummel, a zone of steeply-inclined strata intervenes between the Flat Belt of the Tay Nappe and the Boundary Slide. Since the axial plane of the anticlinal Tay Nappe appears to be above the level of erosion throughout this area, early workers considered that the underlying major syncline (the ‘righting fold’) must lie within this zone. An antiformal syncline around Ben Lui was termed the ‘Ben Lui Fold’ by Bailey (1922) who proposed that it is the F1 syncline beneath the Tay Nappe.

Subsequent work by Cummins and Shackleton (1955) supported Bailey’s view, but a more detailed study by Roberts and Treagus (1964, 1975) showed that the folding is more complex. Beneath the F1 Ardrishaig Anticline, the Ben Lui Fold-complex consists of three recumbent folds with SE-dipping axial surfaces, the Dalmally Syncline, the Ra Chreag Anticline and the Ben Lui Syncline, all of which have been shown to be later, D2 (or possibly D3) structures. This tripartite structure has been traced north-eastwards, via the Balquhidder area (Watkins, 1984), to Glen Lyon (Roberts and Treagus, 1979), beyond which the individual folds lose their identity and the structure is probably represented by the F2 Ruskich Antiform (Nell, 1984; Treagus, 1987, 1999). Many folds in the Schiehallion area and associated folds in Strathtummel with SE-dipping axial surfaces appear to belong to the same generation of F2 folds, which in places are seen to fold the axial planes of major F1 folds such as the Beinn a Chuallich Folds and the Creag na h’Iolaire Anticline (Roberts and Treagus, 1979; Treagus, 1987) see (Figure 3.3)b.

A similar co-axial arrangement of steeply inclined F1, F2 and F3 folds occurs east of the Loch Tay Fault in the Tummel Steep Belt (Bradbury et al., 1979; Treagus, 1999, 2000) and can be traced north-eastwards towards Braemar, where it is known as the Cairnwell Steep Belt (Upton, 1986; Crane et al., 2002) ((Figure 1.7), blocks C and D). These two steep belts are offset along the complex 5 km-wide NW-trending zone of large-scale F2 and F3 folds termed the Carn Dallaig Transfer Zone (Crane et al., 2002, figure 19). They are described in more detail in the Introduction to Chapter 6.

In the area around Schiehallion and Strathtummel ((Figure 1.7), block C) the strata undergo a dramatic swing of strike which is a major feature of even small-scale maps (e.g. (Figure 1.1)). This displacement is caused by two large-scale, steeply plunging, north-trending, relatively late folds termed the Errochty Synform and the Bohespic Antiform (see (Figure 3.43)). Earlier authors assigned these folds to D3 or D4 (Rast, 1958; Thomas, 1980; Treagus, 1987), but Treagus (2000) gave them a local designation of De, being unable to state categorically whether they pre- or post-date the regional D3 phase. The tight synform and broad, open antiform have contrasting geometry. Such changes in fold geometry commonly occur at the junction of materials of contrasting competence; in this case the thick psammites of the Grampian Group and the multilayered pelites and quartzites of the Appin Group. A similar geometry occurs in other fold pairs that have symmetrical spatial relationships to the major NNE- to NE-trending Loch Tay, Bridge of Balgie and Tyndrum faults, and Treagus (1991) considered that there might be a relationship between ductile shearing, the late fold pairs and later brittle faults. The latest folds in this area, termed the Dt phase by Treagus (2000), strongly deform the Errochty Synform along NW-trending axes, the most important being the Trinafour Monoform.

1.5.2 The Boundary Slide-zone and possible related structures

The Boundary Slide can be traced almost continuously, as a zone of high strain and attenuation along the boundary between the Grampian and Appin groups, from Dalmally to Glen Tilt (Roberts and Treagus, 1977c) and then with less certainty, intermittently to Glen Rinnes (Stephenson and Gould, 1995). This zone of highly schistose or platy rocks varies in thickness from a few metres upto 2000m, and includes several planes of dislocation. Significant displacement or tectonic excision of strata is difficult to prove in most areas but a considerable stratigraphical hiatus can be demonstrated. Indeed the whole succession from the Ballachulish Subgroup to the lower Easdale Subgroup is missing over some distance, between Dalmally and Glen Lyon (A.G. Leslie, pers. comm., 2007). Such a large hiatus most likely represents a major unconformity due to non-deposition, possibly over a fault-bounded basement ‘high’. However, overlap can be demonstrated locally and attenuated sequences of the Lochaber and Ballachulish subgroups in the Schiehallion area mark the return to a complete succession towards the north-east (Treagus and King, 1978; Roberts and Treagus, 1979; Treagus, 1987, 1999, 2000). The sharp change in competence between the psammitic and quartzitic Grampian Group and lower parts of the Lochaber Subgroup and the more-variable overlying strata then probably acted as a locus for high strain, with or without localized shearing and displacement.

Originally interpreted as a major dislocation separating two major tectonostratigraphical units, the Ballappel Foundation and the Iltay Nappe (e.g. Bailey, 1922), the Boundary Slide has lessened in significance as a structural boundary in most current interpretations. However, there is no doubt that it does transgress the stratigraphy on a regional basis and local planes of dislocation within the slide-zone might result in a total displacement of up to several kilometres (Treagus, 1987). Individual slides truncate early isoclinal folds and there is no doubt that the slide-zone is folded by later structures. The strong platy or schistose fabric present throughout the zone was regarded by Treagus (1987) as an intense development of the main regional schistosity (S2) developed during nappe formation. Most of the slides within the zone occur on the short limbs of F2 folds and result in an overall movement, in a thrust sense, towards the north-west. This fact is clearly contrary to the overall concept of the dislocations as extensional slides (lags), but it may be that they were initiated as extensional structures during basin development (Soper and Anderton, 1984; Anderton, 1988) or during the formation of the primary nappes (Thomas, 1980) and were then re-activated in a reverse sense as thrusts during subsequent tectonism. The present sinuous trace of the slide reflects large-scale folding by later structures.

A Boundary Slide has been traced north-eastwards from Glen Tilt, through the Braemar area to the eastern end of the Cairngorm Granite (Upton, 1986), where its overall effect is much reduced. Farther north, zones of high strain, locally accompanied by slides, are common at or below the Grampian Group–Appin Group junction but towards the north coast, the boundary appears to represent a relatively undisturbed, rapid stratigraphical passage (see Chapter 6, Introduction). However, zones of shearing around the Grampian–Appin group boundary in Glen Rinnes can be projected north-eastwards towards a major shear-zone that can be traced for some 30 km at higher stratigraphical levels, to reach the coast between Sandend and Portsoy. This Keith Shear-zone appears to have excised parts of the Ballachulish Subgroup in places and between its multiple branches are several pods and lenses of deformed muscovite-biotite granite dated at c. 600 Ma (Barreiro, 1998). Hence this shear-zone, at least, must have been in existence soon after sedimentation, during the extensional events that permitted emplacement of the Ben Vuirich Granite, the Tayvallich volcanism and the associated rifting heralding the opening of the Iapetus Ocean (e.g. Kinny et al., 2003b). Subsequent Grampian deformation of the granite sheets resulted in a top-to-north-west, thrust sense of movement.

1.5.3 Folds beneath the Boundary Slide: the former Glen Orchy and Atholl nappes and the Gaick Fold-complex

In the area around Glen Orchy and Dalmally, Thomas and Treagus (1968) and Thomas (1979) recognized three major isoclinal, recumbent folds beneath the Boundary Slide, the Glen Lochy Anticline, the Beinn Udlaidh Syncline and the Beinn Chuirn Anticline, which they regarded as F1((Figure 1.7), block B). However, Roberts and Treagus (1975, 1977c) interpreted the uppermost, Beinn Chuirn Anticline, as a secondary F2 fold, associated with strong deformation along the Boundary Slide. More-recent fieldwork on the Beinn Udlaidh Syncline by Tanner and Thomas (2010) failed to find evidence of any earlier folding but thin sections clearly revealed a cleavage and metamorphism pre-dating the major folds, all of which were therefore re-assigned to D2. All of the isoclinal folds, together referred to by some authors as the Glen Orchy Nappe, face towards the south-east (i.e. in the same direction as the major nappes above the Boundary Slide). These folds are arched across the broad dome, of the Glen Orchy Antiform, so that they face upwards on the north-west side of the dome and downwards beneath the Boundary Slide on the south-east side. Tanner and Thomas (2010) refer to this late structure as the Orchy Dome and observe that it post-dates late crenulation cleavages and hence is of D4 age or younger. A similar structure to the Glen Orchy Antiform refolds three early isoclines farther to the north-east, in the Glen Lyon area ((Figure 1.7), block C; Roberts and Treagus, 1979).

Still farther to the north-east, around Strathtummel and along the A9 road section, a stack of isoclinal recumbent folds face downwards to the south-east (Thomas, 1979, 1980). Their axial surfaces, limbs and dominant cleavage all steepen to the south-east and dip beneath the Boundary Slide ((Figure 1.7), block D). These folds, most notably the Meall Reamhar Synform, Clunes Antiform and Clunes Synform, were originally interpreted as components of a large-scale, isoclinal F1 fold termed the Atholl Nappe, which was thought to lie beneath the Tay Nappe, the intervening syncline having been cut out by the Boundary Slide. However more-recent work has shown that the dominant major folds and cleavage belong to the D2 regional phase (Treagus, 2000). They all verge towards the north-west and hence are developed on what is essentially an extension of the lower, inverted limb of the Tay Nappe, from which they are separated by the Tummel Steep Belt of later, D3 and possibly D4, folding. The Boundary Slide marks the north-western limit of that steep belt but does not seem to disrupt either the stratigraphy or the overall structure to any significant degree. In places, an earlier, bedding-parallel S1 cleavage is recognized, rare tight, metre- to 10 metre-scale F1 folds do occur on similar axes to the F2 folds and the possibility of large-scale F1 folds cannot be excluded (Treagus, 2000). The later Errochty Synform (Fe phase, see above) refolds the F1 and F2 folds and becomes very tight in this area, where the hinge-zone passes into a high-strain zone of flaggy rocks, at least 3 km wide, called the Dalnacardoch Banded Zone (Thomas, 1979; Treagus, 2000).

Farther north, neither F1 nor F3 folds are recognized across a wide area dominated by a stack of flat-lying, isoclinal, kilometre-scale, south-facing F2 folds and termed the Gaick Fold-complex (Leslie et al., 2006). Around the Pass of Drummochter, the F2 axial planes and cleavages of the flat belt are folded across the broad Drummochter Dome, now generally accepted as a late structure (D3 or D4 depending upon the author) analogous to the domes of Glen Orchy and Glen Lyon (e.g. Lindsay et al., 1989). On the north-western limb of the dome, some complex F1/F2 interference patterns are observed at Crubenmore (Thomas, 1988). Beyond Drummochter, a progressive increase in D3 deformation leads to steepening north-western dips and eventually to a zone of open to close, upright F3 folds on the south-eastern flank of an inlier of ‘basement’ Badenoch Group rocks termed the Glen Banchor High (Robertson and Smith, 1999).

1.5.4 The Geal-charn–Ossian Steep Belt

To the north-west of the Glen Banchor High is a 2–4km-wide zone, in which all the fold limbs and axial planes are steeply dipping (over 60°). This is the Geal-charn - Ossian Steep Belt of Thomas (1979), which can be traced for some 50km from south-west of Loch Ossian to Kinlochlaggan (Robertson and Smith, 1999). It is most clearly defined in the south-west, where it affects upper Grampian Group and Appin Group strata ((Figure 1.7), block D). To the north it widens and becomes more diffuse, affecting lower strata of the Grampian Group and the Glen Banchor Subgroup ‘basement’ succession. On its north-western side, the steep belt includes the Kinlochlaggan Syncline, which has a core of Appin Group strata and has long been regarded as a major isoclinal primary fold (Anderson, 1947b, 1956; Smith, 1968; Treagus, 1969).

Within the steep belt, upright, near-isoclinal coplanar F1 and F2 folds and fabrics (see (Figure 5.4)) are associated with ductile shear-zones and/or slides. Major recumbent folds face in opposite directions on either side, and the intense focussing of deformation into such a narrow zone led Thomas (1979) to interpret the steep belt as a fundamental root-zone from which the large-scale F1 nappes of the Grampian Highlands originally diverged. He therefore regarded it as comparable to, but at a lower structural level than, the Loch Awe Syncline in the South-west Grampian Highlands. The concept of a root-zone is no longer popular but more-recent work has confirmed the steep belt as a zone of primary upright deformation that separates two contrasting structural domains: the upright to steeply inclined NW-facing folds of the Loch Laggan–Glen Roy area to the west, and the gently inclined, SE-facing folds of the Drummochter–Strathtummel area to the south-east (Robertson and Smith, 1999). It probably formed as a result of its location adjacent to an original faulted rift-basin margin, where the strata became compressed against the rigid upstanding buttress of the Glen Banchor High (see (Figure 5.32)). F3 folds in the steep belt are comparable in orientation to F3 folds in the immediately adjacent domains and are characterized by an upright crenulation cleavage.

To the north of Kinlochlaggan, a zone of steeply inclined strata, thrust slices and intense deformation extends to the north-north-east through the Monadhliath mountains (Piasecki and van Breemen, 1983). However, many of the upright folds in this zone are interpreted as later, possibly F3 structures (Smith, 1968) and their relationship to the steep belt is uncertain.

1.5.5 NW-facing folds between Islay and Glen Roy

This area is dominated by large-scale recumbent NW-facing early folds that have been refolded by largely upright open to close later folds. The area between Appin and Glen Roy in particular is arguably the most intensively studied in the whole of the Scottish Dalradian, particularly in the well-exposed, cross-strike section along Loch Leven and Loch Eilde that is represented by several GCR sites in Chapter 3. The original interpretation of the regional structure by Bailey (1910) was based largely upon that section. It underwent several modifications, including a complete reversal of the order of stratigraphical succession and consequently the facing direction of the major folds. The final interpretation by Bailey (1934, 1960) has remained a sound basis for all subsequent work.

The two principal re-interpretations (Roberts and Treagus, 1977c; Hickman, 1978) both accepted much of the near-surface fold geometry proposed by Bailey (1934). However, they differed considerably in the way in which the folds were projected to depth and correlated across strike. There were also highly significant differences in the assignation of relative ages to individual folds, much of the evidence for which depends on the detailed observation and interpretation of minor structures. These contrasting models have been summarized by Stephenson and Gould (1995) and are illustrated in (Figure 1.8). They are all commented upon in Chapter 3, but the descriptions there follow the Roberts and Treagus (1977c) model, which was based largely upon detailed studies in the Ardsheal Peninsula (Treagus and Treagus, 1971), around Ballachulish (Roberts, 1976) and around Kinlochleven (Treagus, 1974). It also drew upon the authors’ experience in adjoining areas of the South-west Grampian Highlands (Thomas and Treagus, 1968; Roberts, 1974; Roberts and Treagus, 1975).

In the Loch Leven area, Bailey recognized three NW-facing recumbent isoclinal folds, each about 15 to 20km in amplitude. From north-west to south-east and progressing structurally upwards these are the Appin Syncline, the Kinlochleven Anticline and the Ballachulish Syncline. On the lower limb of the Kinlochleven Anticline, demonstrably inverted strata of the Lochaber Subgroup crop-out over a cross-strike width of some 7km to form what is known as the Kinlochleven Inversion. Bailey also recognized the existence of near co-axial upright ‘secondary folds’, such as the Stob Ban Synform, which refolds both the Kinlochleven Anticline and the Ballachulish Syncline, and additional F2 folds such as the Kinlochleven Antiform and Blackwater Synform that deform rocks of the Kinlochleven Inversion were recognised by Roberts and Treagus. The lower limbs of the two major synclines were considered by Bailey to be largely replaced at an early stage in the deformation by tectonic slides: the Fort William Slide beneath the Appin Syncline and the Ballachulish Slide beneath the Ballachulish Syncline. However, the apparent excision of strata along the Fort William Slide, at least in this area, has been shown subsequently to be due to an unconformity in the lower part of the Appin Group (Glover, 1993).

Any fundamental departures from Bailey’s regional interpretation have been refuted by contemporaneous and later researchers. In particular, Weiss and McIntyre (1957) concentrated entirely on small-scale structures, and by taking no account of the larger-scale structures or overall stratigraphy, failed to recognize any earlier recumbent folds. The concept of large-scale nappes was also completely rejected by Voll (1964) who explained the observed outcrop pattern in the critical Loch Creran area in terms of major facies changes. However, a later paper did recognize the Kinlochleven Inversion, and hence accepted the presence of recumbent folds (Kruhl and Voll, 1975).

Hickman (1978) re-interpreted many of the large-scale folds identified as primary by Bailey and by Roberts and Treagus, as secondary, F2 structures. Large-scale recumbent F1 folds were recognized only in the eastern part of the section, and in the north-western part of the section, the Appin Syncline and the Tom Meadhoin Anticline (the latter formerly regarded as the upward-facing hinge of the Kinlochleven Anticline) were re-interpreted as separate upright F2 folds. In many respects this model followed the interpretation of Bowes and Wright (1967, 1973), based upon detailed mapping in the Ardsheal peninsula, and was contested by Roberts and Treagus (1980).

A similar major fold geometry has been identified farther to the north-east, around Glen Spean and Glen Roy (Key et al., 1997). There, large-scale recumbent F1 folds, correlated with the Treig Syncline and Kinlochleven Anticline, are refolded by co-axial upright F2 folds, including the Appin Synform, the Stob Ban Synform, the Inverlair Antiform and the Blackwater Synform (Figure 1.9). In this area, the Fort William Slide does seem to exist as a high-strain zone and has been traced around the closure of the Appin Syncline, but it dies out eastwards, where the Lochaber Subgroup appears to rest conformably on the Grampian Group.

Correlation of the major folds to the south-west of Loch Leven, into the area of Loch Creran and Benderloch, and onwards to the Islay Anticline and Loch Awe Syncline, has been the subject of much controversy. The Appin Syncline can be traced from Glen Roy through Ardsheal to Lismore where, according to Hickman (1978), the equivalent Balygrundle Syncline refolds an early isoclinal fold, confirming its F2 status. The Tom Meadhoin Anticline was correlated with the Airds Hill Anticline of Appin and Benderloch by both Roberts and Treagus (1977c) and Hickman (1978), but whereas the former regarded it as the hinge of the Kinlochleven Anticline, and hence an F1 fold, the latter regarded it as F2. The Ballachulish Syncline was correlated by Roberts and Treagus with the Beinn Donn Syncline of Loch Creran but its main trace is subsumed into a complex stack of slides farther to the east in Glen Creran (Bailey, 1960).

The Loch Creran area was described by Litherland (1982) almost entirely in terms of relatively upright F1 folds in a simple ‘mushroom-like’ structure. He recognized no secondary folds of regional extent, whereas Roberts and Treagus (1977c) concluded that several of the F1 folds are refolded by a continuation of the F2 Stob Ban Synform and Hickman (1978) interpreted some of the upright folds as F2.

Both the Islay Anticline and the Loch Awe Syncline were regarded by Bailey as secondary folds but almost all subsequent workers have interpreted them as primary (e.g. Cummins and Shackleton, 1955; Shackleton, 1958; Rast, 1963; Borradaile, 1973; Roberts, 1974; Roberts and Treagus, 1977c; Litherland, 1982). Roberts and Treagus correlated the Islay Anticline with the Airds Hill and Kinlochleven anticlines, and the Loch Awe Syncline with the Beinn Don and Ballachulish synclines. However, those correlations involve considerable deflections of the fold axial traces and Litherland (1982) suggested alternative, more-direct correlations. He correlated the Islay Anticline with the Glen Creran Anticline, and the Loch Awe Syncline with the Beinn Sgulaird Syncline of upper Glen Creran.

The correlation of major slides in the west of the Central Grampian Highlands has been the subject of much speculation, especially when they were thought to separate fundamental tectonostratigraphical units. On Islay, the Islay Anticline over-rides the Bowmore Sandstone along the Loch Skerrols Thrust, which was formerly equated with the Moine Thrust (Bailey 1917, 1922; Kennedy, 1946). Later authors projected the Loch Skerrols Thrust north-eastwards into the Benderloch Slide at Loch Creran, which Bailey (1922) had correlated directly across the Etive Pluton with the Boundary Slide of the Central Grampian Highlands, to form what was later to become known as the ‘Iltay Boundary Slide’ (MacGregor, 1948; Rast, 1963). Subsequently, Rast and Litherland (1970) proposed that the Benderloch Slide could be extended north-eastwards beyond the Ballachulish Pluton to continue as the Ballachulish Slide ((Figure 1.7), block B). The latter is folded around the closure of the F2 Stob Ban Synform and trends back southwards to Glen Etive, from where Roberts and Treagus (1977c), following Bailey, linked it across the Etive Pluton with a slide in the Dalmally area, to propose once again a single, continuous Boundary Slide. Whatever their connections and correlations might be, the slides in this western area are clearly folded by F2 folds. Hence their initiation and main movement must have been early, and Litherland (1982) suggested that the Benderloch Slide could have been a re-activated synsedimentary fault. However, intense local re-activation of slides in the Dalmally area is compatible with the dominant D2 movements elsewhere along the Boundary Slide (Roberts and Treagus, 1975).

1.5.6 Corrieyairack, Strathspey and the Monadhliath Mountains

Metasedimentary successions in this area of the Northern Grampian Highlands consist almost entirely of rocks of the Grampian Group and the Badenoch Group.

To the south-east and east of the Geal-charn–Ossian Steep Belt, structures are comparable to and share a common history of structural development with those described above in the overlying Atholl Nappe (Lindsay et al., 1989). This structural continuity was used as an argument for including the Grampian Group in the Dalradian, in accord with the proposal of Harris et al. (1978), based largely on stratigraphical continuity.

To the west and north-west of the steep belt and its projected continuation, many of the major structures are downward continuations of those that affect higher structural and stratigraphical levels of the Dalradian to the south and south-west, and hence the phases of deformation are broadly comparable.

Semipelites and micaceous tops to graded psammitic beds in the Grampian Group commonly show an early schistosity, usually subparallel to or steeper than bedding, which is clearly folded by the major folds and is assumed to represent S1. In the area around Loch Killin and the Corrieyairack Pass, Haselock et al. (1982) also recognized stratigraphical repetitions and minor structures, which suggested the presence of major early isoclines, refolded by the major more-upright regional folds. A similar structural pattern was recorded in the area between the Corrieyairack Pluton and Loch Laggan by Okonkwo (1988). However, major F1 folds cannot be identified anywhere north-east of Glen Roy and the Strath Ossian Pluton (Key et al., 1997).

These areas are dominated by near-upright NE-trending F2 folds that face to the north-west, such as the Tarff Antiform-Synform pair, the Corrieyairack Synform and the Creag Mhor Anticline (Haselock et al., 1982; Key et al., 1997; May and Highton, 1997). Their axes are typically gently inclined to subhorizontal, curvilinear and are associated with a prominent lineation. Farther east, in the upper Findhorn area, the major folds, such as the Loch Laggan Anticline and the Loch Laggan–Monadhliath Syncline-complex, become tight to isoclinal and, together with the Corrieyairack Syncline, were originally interpreted as primary structures (Anderson, 1956; Piasecki, 1975).

The earliest recognizable folds to affect most of the Grampian Group (F2) are associated locally with high-strain zones and slides, which are particularly abundant close to the contact with the ‘basement’ of the Badenoch Group (e.g. the Lochindorb Shear-zone). In the west, between Loch Lochy and Loch Tarff, the Eilrig Shear-zone brings younger Grampian Group rocks to rest upon the Glenshirra Subgroup (Phillips et al., 1993; Key et al., 1997). Microstructures indicate a consistent NW-directed sense of shear and the metamorphic state changes from amphibolite facies above to greenschist facies below the shear-zone. Phillips et al. (1993) suggested that this shear-zone might represent the surface expression of a major basal decollement or floor thrust to many of the structurally higher slides and shear-zones.

In places, the D2 structures are indistinguishable from those of similar-style F3 folds. However, in the area between Killin, Farr and upper Glen Kyllachy, D3 structures consist of near-upright open folds with axes generally trending around north–south (Piasecki, 1975; Haselock et al., 1982) see (Figure 5.2)b. Minor structures and fabrics of this phase dominate the area, overprinting and commonly obliterating the earlier structures.

A major structural feature of the Northern Grampian Highlands is ta 3–4 km-wide zone that extends almost normal to the Caledonian regional trend from the Foyers Pluton, through Kincraig, to the eastern end of the Cairngorm Pluton. Within this zone, bedding, foliation and F1 and F2 axial surface traces become south-east trending and the upright to sideways-facing folds verge towards the south-west. The change in trend of the major folds from north-east to south-east see (Figure 5.2)b was recognized and investigated in the area between Kincraig and Newtonmore by Whitten (1959) in a detailed petrofabric study of the minor structures, which concluded that the two trends represent a single generation of folds. Later NW-trending open domes (?F4) have a considerable influence on the local outcrop pattern, around Kincraig for example, and expose inliers of Badenoch Group basement. Hence Smith et al. (1999) have suggested that the SE-trending zone, which they termed the Foyers–Cairngorm Lineament, is founded upon a series of basement ‘highs’ see (Figure 5.2)a, which influenced the orientation of fold axes within the shallow cover of Grampian Group strata.

To the south-west, between the Allt Crom Complex and the Corrieyairack Pluton, the broad arch of the post-D2 (?F3) Glenshirra Dome has a trace that swings from east to north-east (see (Figure 5.19)). The dome exposes an inlier of the Glenshirra Subgroup and both Haselock et al. (1982) and Okonkwo (1988) recognized a zone of high strain between it and the overlying Corrieyairack Subgroup, which they termed the Gairbeinn Slide and attributed to D1. However, recent investigations have concluded that the high strain is focused at the boundary between contrasting lithologies within a conformable succession and can be attributed to D2 (Smith et al., 1999).

1.5.6.1 Structures affecting the Badenoch Group

The tectonic relationships between the Grampian Group and the Badenoch Group have been discussed in detail elsewhere in this chapter. The current view advocates that the largely migmatitic Badenoch Group constitutes an older basement beneath a cover of Grampian Group metasedimentary rocks and consequently that they have undergone more phases of deformation and metamorphism (Piasecki and van Breemen, 1979a, 1979b, 1983; Piasecki, 1980; Piasecki and Temperley, 1988a; Smith et al., 1999).

1.5.6.1.1 The Grampian Shear-zone

The boundary between the Grampian Group and the Badenoch Group was formerly taken at a complex zone of shearing and dislocation, up to 200m in thickness, known then as the ‘Grampian Slide’ or ‘Slide-zone’ ((Figure 1.7), blocks E and F). Minor slides occur locally both above and below this zone and the slides appear to anastomose on a regional scale. It is now thought that the major slides, and certainly those that contain syntectonic pegmatitic veins dated at c. 806 Ma, are all within the Badenoch Group. However, the effects of later shearing are thought to have extended structurally downwards from the Grampian Group into the basement rocks and hence the Neoproterozoic ductile slides in the Grampian Shear-zone were reworked strongly in places during the Grampian Event. The main reworking is regarded as D2 in the structural sequence identified in the Grampian Group, since the shear fabrics are parallel to the axial surfaces of near-isoclinal folds that fold an earlier fabric assumed to represent D1.

Individual shear-zones range from a few centimetres up to several tens of metres thick and are characterized by intense grain-size reduction, destruction of pre-existing gneissose and migmatitic foliations, attenuation of minor folds, and muscovite growth. They commonly follow lithological boundaries and movement is concentrated locally within pelitic units. A variety of kinematic indicators indicate a mainly top-to-the-north or -north-east sense of shear.

1.5.6.1.2 Structures below the Grampian shear-zone

Most parts of the Badenoch Group are characterized by a penetrative gneissose fabric, which is folded by tight to isoclinal, recumbent folds. The gneissose fabric is generally concordant with broad compositional banding, which might represent original bedding. No major F1 folds have been recognized but rare intrafolial folds and local low-angle discordance between the gneissose fabric and compositional banding suggest that early large-scale recumbent structures do exist. This deformation is considered to be broadly coeval with middle- to upper-amphibolite-facies metamorphism and regional migmatization.

The early gneissose fabric of the Badenoch Group successions is crenulated and transposed by the S1 schistosity that affects the Grampian Group rocks, which is associated with a later generation of lit-par-lit migmatites in lithologies of suitable composition.

The outcrop pattern is controlled largely by intermediate-scale folds. In areas of low strain these structures are recumbent to reclined, with a well-developed axial planar crenulation cleavage. In areas of intense deformation the earlier gneissose fabric becomes transposed into the crenulation fabric to produce a new banding. A larger-scale recumbent fold in the Kincraig area, has a NNW-trending outcrop of Glen Banchor Subgroup rocks in its core; its axis trends north-east and it closes to the north-west. All of these folds were regarded by Piasecki (1980) and Piasecki and Temperley (1988a) as affecting only the basement rocks, but are now considered to be part of the D2 phase in the Grampian Group structural sequence. Several phases of later near-upright folding recognized locally in the basement by Piasecki (1980) were correlated with the later deformations in the Grampian Group.

1.5.7 The Banff and Buchan area

The Banff and Buchan area of the North-east Grampian Highlands is bounded to the west and south by major shear-zones and it is distinguished from the remainder of the Scottish Dalradian by different stratigraphical, metamorphic, igneous and geophysical features. The western boundary of this Buchan Block is marked by the Portsoy–Duchray Hill Lineament, a major tectonic and stratigraphical boundary with a long history of extensional, transtensional, compressional and transpressional activity, which can be traced from the north coast to the Glen Shee area (Fettes et al., 1986, Goodman, 1994). Major stratigraphical discontinuities occur at the lineament (Fettes et al., 1991) and marked differences in metamorphic history on each side indicate major westward overthrusting during the regional D3 event (Baker, 1987; Beddoe-Stephens, 1990). The lineament also appears to have acted as a pathway for magma at various times, influencing the emplacement of both basic and silicic sheet intrusions. The southern margin of the Buchan Block is more difficult to define. It extends at least as far as the granitic plutons of Deeside but farther south the stratigraphical and structural features merge with those of the Highland Border region through an imbricated zone of east–west-trending shear-zones.

Various zones of shearing and dislocation on the western margin of the Buchan Block have been used to define its limit (see Chapter 6, Introduction). On the eastern edge of the Cowhythe Psammite Formation outcrop, a zone of highly deformed rocks marks the position of the Boyne Line of Read (1955), which was interpreted as a major slide underlying his proposed Banff Nappe. From the western margin of the Cowhythe Formation, a sub-vertical zone of shearing and faulting extends through Portsoy westwards for over 1km. This Portsoy Shear-zone, which can be traced inland in a general southerly direction to the Cabrach area, is the main surface expression of the northern section of the Portsoy–Duchray Hill Lineament.

To the west of the Portsoy–Duchray Hill Lineament, Appin and Argyll group rocks are involved in a series of NW-facing folds, which can be traced down sequence into the underlying Grampian Group. There would appear to be no equivalent to the Boundary Slide in the coast section, although high strain along this boundary does occur in Glen Rinnes, 40 km to the south-west. On the east side of the Buchan Block, recumbent folds which might be correlated with those of the Tay Nappe have been identified in sections around Collieston (see above). However, the Buchan Block is dominated by late, open, broad, upright folds with gently NNE-plunging axes, principally the Turriff Syncline and Buchan Anticline, within which the upper parts of the Dalradian succession occur in right-way-up sequences (see Chapter 6 for a full discussion).

Early interpretations regarded the Banff and Buchan Dalradian succession as part of an allochthonous, recumbent, SE-facing ‘Banff Nappe’ (Read, 1923, 1955; Read and Farquhar, 1956) and other authors have suggested that the nappe and its underlying gneissose ‘basement’ are all allochthonous (Sturt et al., 1977; Ramsay and Sturt, 1979). However, current interpretations suggest that the structures are essentially autochthonous and that successions can be correlated, at least at subgroup level, with those farther to the south-west (see Chapter 6, Introduction).

Along the north coast the rocks exhibit locally complex folding with steep dips over much of the section, but regionally they are disposed in the form of a broad, open syncline, the Turriff Syncline. The western limb of the syncline is steep due to a related monoform, termed the Boyndie Syncline and to the east it passes into the broad Buchan Anticline. An inlier of gneissose rocks in the hinge of the anticline was thought by Read and Farquhar (1956) to be in the core of the Banff Nappe, which others have equated tentatively with the Tay Nappe (Treagus and Roberts, 1981; Ashcroft et al., 1984).

The attitude and style of the consistently upward-facing F1 folds varies according to structural level, being generally upright and open to close at the highest levels in the centre of the late Turriff Syncline and generally close to tight on both limbs, where they face to the north-west, contrary to Read’s (1955) model of a SE-facing nappe. Post-D1 structures are restricted to lower structural levels and hence are seen only on the western and eastern limbs of the Turriff Syncline. D2 folds and fabrics are dominant locally, as in the area around Portsoy. The D3 deformation, which post-dates the peak of metamorphism, is represented by westerly verging, small- to medium-scale folds with limits that almost coincide with the andalusite isograd (Figure 6.3)a. Larger-scale F3 folds are characteristically monoclinal, as in the Boyndie Syncline. The major Turriff Syncline and Buchan Anticline have been tentatively attributed to the D3 phase by most authors, although their overall open and upright character is more comparable with major D4 structures elsewhere.

One of the most striking features of both stratigraphical and structural maps of the Grampian Highlands is the 20 km-amplitude ‘S‘-shaped ‘knee-bend’ in the strata in the area between Braemar and Tomintoul (e.g. (Figure 1.1), (Figure 1.7), block E, 6.1). The reasons for this major feature are obscure, since there are no obvious associated minor folds or cleavages developed. It is clearly a late structure, as all of the folds and dislocations in the area, including the Portsoy–Duchray Hill Lineament, are folded around it. Such a large-scale structure must reflect deep crustal weaknesses and might reflect crustal block movements in the later stages of the Caledonian Orogeny. It is curious that both axial traces of the ‘Knee-bend‘ appear to be marked by E–W-trending lines of late-Caledonian granitic plutons, the southern one roughly coinciding with the Deeside Lineament.

1.5.8 Structural development of the Grampian Highlands

In most areas of the Grampian Highlands the structural development of the Dalradian has been explained in terms of three or four major episodes or phases of deformation (D1–4), all of which occurred during the Caledonian Orogeny. Pre-Caledonian events are recorded only in the Badenoch Group and the Rhinns Complex. During the first widespread Caledonian deformation (D1), close to tight major folds were initiated with a north-east–south-west axial trend. The D1 structures are widespread but were commonly transposed during later events and hence are difficult to detect except in the south-west and south-east. Metamorphism at this stage did not exceed greenschist facies. D2 structures affect almost all Dalradian strata except for those at the highest structural levels, close to the Highland Boundary Fault. During D2 the F1 folds were extensively modified, either by near-co-axial refolding or, at deeper levels, by simple shear on a regional scale. Much of the observed high strain and/or displacement along major slides can be attributed to the D2 phase, although the slides might have been initiated earlier, during D1 or as synsedimentary growth faults. A separate D3 phase of generally more-upright folding can be identified in most areas, except within about 5–10 km of the Highland Boundary Fault. It is broadly coincident with the peak of regional metamorphism and associated igneous intrusion, although it might be a little later in some areas, particularly in the north-east. Various late-tectonic phases overprint the composite foliations of the early nappes and the F2 and F3 fold-complexes. The most widespread, D4 structures are within the Highland Border Steep Belt and the Buchan Block but still-later phases occur locally. They are all separated from the earlier movements by a significant time gap and seem to be the result of a change in tectonic regime from ductile folding to basement fracture and block uplift, possibly coeval with but not necessarily resulting from the Scandian (Baltica–Laurentia) collision.

1.5.8.1 Early crustal shortening (D1)

The early regional syntheses of E.B. Bailey and others established that the Dalradian strata are disposed in large-scale recumbent folds or nappes that originated early in the deformational sequence. The mechanism of nappe development was not considered at that time, apart from an ‘eddy’ theory in which NW-directed movements in the lower structural levels are compensated by movements in the opposite direction at higher levels (Bailey, 1938). South-eastward gravity sliding of the higher nappes was proposed by Cummins and Shackleton (1955) and has subsequently been invoked by several authors in conjunction with various models (e.g. Bradbury et al., 1979; Shackleton, 1979; Anderton, 1988) (Figure 1.10)c. However, the first overall models involved the concept of a root-zone, from which nappes were expelled laterally in opposing directions (Sturt, 1961; Harris, 1963; Rast, 1963; Thomas, 1979, 1980) (Figure 1.10)a. These were refined into models in which essentially upright major folds fanned outwards as ‘mushroom’ or ‘fountain’ structures above the root-zone and then ‘collapsed’ under gravity in a continuous process to produce the recumbent structures (Roberts and Treagus, 1977c, 1979; Nell, 1986; Treagus, 1987) (Figure 1.10)b. The later models emphasized the importance of north-westerly directed movements, particularly on such structures as the Ballachulish and Boundary slides, possibly rising from a fundamental ‘floor thrust’ (Bradbury, 1985; Mendum and Fettes, 1985; Nell, 1986; L.M. Hall in Fettes et al., 1986; Treagus 1987) (Figure 1.10)d. Thus, despite the apparent dominance of the SE-facing Tay Nappe, overall movement in the Grampian Highlands was considered to be to the north-west, and hence the structural development of the whole Scottish Caledonides could be modelled as a complete entity from the Highland Border to the Moine Thrust Belt, where Caledonian structures over-ride the foreland (e.g. Coward, 1983; Fettes et al., 1986). All of the above models have been summarized by Stephenson and Gould (1995, pp. 102–107), from where (Figure 1.10) has been taken.

The overall concept of a root-zone and a ‘mushroom’ or upright ‘fountain’ of primary folds has fallen out of favour. Most authors find it difficult to imagine major nappes being expelled in this manner from such a narrow zone, either laterally or vertically, and root-like F1 folds are difficult to identify with any certainty in the complex steep belts (e.g. Shackleton, 1979; Coward, 1983). There is no direct evidence regarding the original attitude of the F1 folds (i.e. upright or recumbent) and current conflicting views are dependent upon how they are seen to relate in particular to the development of the Tay Nappe (see below).

1.5.8.2 Peak deformation (D2)

The history of development of the Tay Nappe is central to all interpretations of the overall evolution of the Grampian Highlands and is a matter of continuing discussion. It is possible that SE-facing folds, such as the Tay Nappe, originated as primary (D1) structures, in which case the F1 folds were most likely to have been recumbent and tight to isoclinal (e.g. Mendum and Fettes, 1985; Mendum and Thomas, 1997). However, other authors have considered that their current geometry and facing direction is a result of subsequent deformation during the D2 phase. Both backward gravitational collapse of an upright ‘nappe fountain’ (Roberts and Treagus, 1977a, 1977c; Treagus, 1987; (Figure 1.10)b) and large-scale isoclinal backfolding of an original NW-facing Tay Nappe (L.M. Hall in Fettes et al., 1986; (Figure 1.10)d) have been suggested. New fold hinges formed by either of these mechanisms became F2 folds, such as the the Ben Lui Fold-complex, which has the overall effect of ‘righting’ the inverted limb of the Tay Nappe.

However, most authors now agree that the D2 deformation was almost continuous with D1 as part of a single progressive event and had a high simple-shear component. Nell (1986), Treagus (1987) and Anderton (1988) all suggested that the F2 folds were overturned towards the north-west at deep structural levels where they were associated with NW-directed simple shear along the Boundary Slide and related dislocations. At higher levels, the upright F1 folds became translated by SE-directed simple shear, also of D2 age, to produce the intense deformation of the inverted limb of the Tay Nappe in the Flat Belt as described by Harris et al. (1976) (echoes of the ‘eddy’ theory of Bailey, 1938).

Most current models for the development of the Tay Nappe invoke near-upright compressional F1 folds, the lower levels of which were subjected to top-to-the-south-east simple shear as a result of north-westward subduction beneath the Laurentian margin (e.g. Krabbendam et al., 1997; Treagus, 1999; Rose and Harris, 2000). At higher levels the upright F1 folds were little affected by the D2 deformation, except that they might have developed a south-east vergence (P.W.G. Tanner, personal communication, 2008). They are preserved only in the Aberfoyle Anticline and equivalent folds in the Highland Border Steep Belt. At lower levels, the originally SE-facing limbs became the dominant long limbs of shear-folds, giving the impression of being part of a single inverted lower limb of a large-scale nappe (see (Figure 1.11)). Krabbendam et al. (1997) estimated the structural thickness affected by the shear to be 4–5 km, with a displacement of 10–50 km.

In areas to the north-west of the Tay Nappe, the F1 folds were considerably modified by generally co-axial F2 folding, which is particularly well developed above the Boundary Slide and in the various steep belts. To the north-west of the Geal-charn–Ossian Steep Belt, NW-facing nappes, such as the Ballachulish Nappe, were refolded into upright folds such as the Stob Ban Synform. However, most of the F2 folds were originally recumbent and verged to the north-west. They are typically close to isoclinal and are almost always associated with the dominant foliation (S2), which varies from pressure-solution striping to crenulation, schistosity or gneissosity, depending on lithology, metamorphic grade and amount of strain. Extremely strong fabrics, some submylonitic, in high-strain zones on the attenuated limbs of large-scale F2 folds are continuous with the penetrative S2 fabrics, confirming the D2 affinities of major slides, most of which moved in a top-to-north-west thrust sense during this phase. The stacking produced by the D2 refolding and thrusting resulted in considerable crustal thickening that eventually led to the classic Barrovian metamorphism.

1.5.8.3 Waning deformation and peak metamorphism (D3)

The D3 deformation phase is dominated by more-upright structures that are particularly well represented in the various steep belts such as the Tummel Steep Belt. They formed under lower to middle amphibolite-facies conditions, and were associated with locally high temperatures producing minor granitic melts and pegmatitic segregations. The F3 folds clearly fold the dominant S2 fabrics and vary from open to tight, with moderately inclined to vertical axial planes. Axial planes and fold axes display highly variable orientations; in many places the F3 folds are co-axial with F2 folds, but elsewhere they developed at right angles, as in the Carn Dallaig Transfer Zone near Kirkmichael, and spectacular non-co-axial interference structures are seen in some areas.

1.5.8.4 Post-nappe deformation and uplift (D4 and later)

The later phases of Caledonian deformation were, with notable local exceptions, generally less intense and produced more-open, near-upright structures trending between east-north-east and north-east. These take the form of broad upright folds, chevron-style minor folds, crenulations and brittle structures; well-developed axial planar fabrics are generally seen only in hinge-zones. The numbering of later phases is frequently inconsistent and is complicated by local variations (see for example the scheme adopted by Treagus, 2000 to identify late phases in the area north of Schiehallion, which uses local letters rather than regional numbers). However, most authors are agreed upon the existence of a late, D4 phase that is associated with retrograde metamorphism.

The D4 phase probably included more than just one type of tectonic mechanism but most authors have attributed the initial, most widespread D4 deformation to the commencement of late-orogenic isostatic uplift (Watson, 1984; Harte et al., 1984; Mendum and Fettes, 1985; Dempster, 1985a). The scale, monoformal nature and lateral continuity of many of the earlier F4 folds, especially the Highland Border Downbend, suggest that they are controlled by a parallel series of major basement lineaments. Between these lineaments episodic uplift of crustal blocks at different rates generated the pattern of contrasting flat and steep belts which now dominates the south-eastern Grampian Highlands.

The Highland Border Downbend is the largest and most obvious D4 structure and is responsible for the downturning of the hinge-zone and part of the lower limb of the Tay Nappe into the Highland Border Steep Belt. The D4 deformation is also probably responsible for the steepening of major F2 and F3 fold limbs in the Knapdale, Tummel and Cairnwell steep belts. In the North-east Grampian Highlands, the Turriff Syncline, Buchan Anticline and other late NE- to NNE-trending open to close folds are regarded as D4, although they are often difficult to separate from D3 structures. Renewed movements at this time on major shear-zones such as the Portsoy Shear-zone, probably resulted in further disruption and re-orientation of intrusions of the North-east Grampian Basic Suite.

Minor folds and crenulation cleavages that overprint the earlier D4 major flexures locally, can be traced north-westwards across the Flat Belt, where they are seen to be related to broad, upright folds such as the Ben Lawers Synform (termed F3 by Treagus, 1987, 2000). These are taken to indicate a later D4 compressional phase which post-dates the earlier D4 uplift event. Roberts and Treagus (1977c, 1979) and Treagus (1987) also considered that the Drummochter and Glen Orchy domes are late-D4 compressional structures (i.e. their D3).

Other late major folds, such as the Bohespic Antiform and Errochty Synform, north of Schiehallion, which have a more north–south trend, were attributed to D4 (his D3) by Thomas (1979, 1980). They were considered to belong to a post-D4 phase (i.e. post his D3) by Treagus (1987) but were then re-interpreted as pre-dating D4 (his D3) by Treagus (2000). Whatever their relative age might be, they form part of a set of major north- to NNE-trending folds between the Ericht–Laidon and Loch Tay faults, which Treagus (1991) related to the initiation of sinistral movement on the faults. Refolding and overprinting by more-definitely later fabrics are generally only of local extent and tend to be small-scale, brittle open box folds and conjugate kink-zones. However, significant NW-trending flexures and monoforms affect the limbs of the Errochty Synform (Thomas, 1980; Treagus, 2000), and Roberts (1974) described a complex sequence of late structures from the South-west Grampian Highlands.

1.6 Metamorphism of the Grampian Highlands

D.J. Fettes and A.G. Leslie

The metamorphic grade expressed in the Dalradian rocks of the Grampian Highlands was initally referred to index minerals in pelitic rocks. Barrow (1893, 1912), working in the south-east of the Grampian Highlands, was the first to establish a zonal sequence in such rocks, indicative of progressive metamorphic grade. This scheme, slightly modified by Tilley (1925), became the classical Barrovian zones (chlorite → biotite → garnet → staurolite → kyanite → sillimanite). Subsequently, Read (1952) recognized a different style of metamorphism in the North-east Grampian Highlands, characterized by a progressive mineral sequence, which he termed the Buchan zones (biotite → cordierite → andalusite → sillimanite). The overall pattern of metamorphic zonation across the Grampian Highlands was established principally by Elles and Tilley (1930), Kennedy (1948), Chinner (1966), Dewey and Pankhurst (1970) and Winchester (1974) and comprehensive summaries of the metamorphic reactions that define the boundaries of the metamorphic zones were provided by Atherton (1977), Harte and Hudson (1979) and Hudson (1980). Although these workers all made use of zonation in pelitic and calc-silicate rocks, the metamorphic grade is more-generally illustrated by the broader concept of metamorphic facies, which is applicable to all lithologies (Fettes et al., 1985; Harte, 1988). The basis of the correlation between the mineral zones and the range of facies was given by Fettes et al. (1985).

Read (1952) regarded the Barrovian and Buchan metamorphisms as quite separate events. However, Johnson (1962, 1963) demonstrated that the two metamorphisms were coeval relative to the deformation sequence, a conclusion subsequently supported by radiometric dates (Oliver et al., 2000). Fettes et al. (1976) demonstrated that the Buchan zones are part of a progressive decrease in the pressure of metamorphism (or increase in the geothermal gradient) from the South-west to the North-east Grampian Highlands. The transition from the Barrovian to the Buchan areas was detailed by Harte and Hudson (1979), who defined a series of four zonal sequences reflecting decreasing pressure, namely: Barrovian (biotite → garnet → staurolite → kyanite → sillimanite), Stonehavian (biotite → garnet → chloritoid + biotite → staurolite → sillimanite, West Buchan (biotite → cordierite → andalusite → staurolite → kyanite) and East Buchan (biotite → cordierite → andalusite → sillimanite → sillimanite + K-feldspar). That is, the transition from Barrovian zonal sequences to Buchan sequences occurs both northwards along the Stonehaven coast and eastwards along the Banffshire coast. Strictly, the Buchan sequence as defined by Read (1952) is restricted to eastern Aberdeenshire. However, the term Buchan metamorphism is commonly used for the area of low-P/T type, that is the West Buchan and East Buchan sequences of Harte and Hudson (1979). Similarily the type area of Barrovian metamorphism is the Angus glens and eastern Perthshire, although the term is commonly used to describe the wider area of intermediate-P/T type (see below).

The above studies clearly indicate that the Buchan and Barrovian sequences are expressions of the variable pressure-temperature conditions, in both space and time, of the Grampian Event; the two sequences are broadly synchronous and are not separate or dissociated metamorphisms.

The metamorphic history of the Dalradian rocks on the Shetland Isles and its timing relative to Caledonian deformation is difficult to correlate with that of the Grampian Highlands and is discussed separately in the introduction to Chapter 7.

1.6.1 Distribution of facies

The distribution of metamorphic facies is shown in (Figure 1.12). This shows the general increase in grade from greenschist facies in the South-west Grampian Highlands, north-eastwards through the epidote-amphibolite facies to predominantly amphibolite facies; greenschist-facies assemblages wedge out as two arms running, respectively, along the Highland Boundary and Great Glen faults. The greater part of the higher grade rocks lie in the lower amphibolite facies, characterized by kyanite + staurolite and andalusite + cordierite assemblages. In the Northern and North-eastern Grampian Highlands the rocks reach middle and upper amphibolite-facies conditions, characterized by sillimanite + muscovite and sillimanite + K-feldspar assemblages. Superimposed on the main metamorphic facies from Loch Tay to Kintyre is a regional zone marked by the development of late-stage albite porphyroblasts (Watkins, 1983).

Some earlier zonal maps differentiated areas of ‘migmatites’ (e.g. Dewey and Pankhurst, 1970; Fettes, 1979). These rocks are characterized by the presence of quartzofeldspathic lenses, pods or stringers and are generally associated with sillimanite-bearing amphibolite-facies rocks. Barrow (1912) originally considered the migmatites as the heat source for the metamorphism, a view echoed by Kennedy (1948) and Read (1955). However subsequent studies have shown that the ‘migmatites’ encompass a variety of products including coarse sillimanite gneisses and true anatectic melts (e.g. Atherton, 1977; Ashworth, 1976, 1979). These products were derived by a range of relatively high-grade metamorphic and metasomatic processes, and the presence or absence of ‘migmatites’ at the local scale is largely controlled by the chemical composition of the host lithologies. As such, ‘migmatites’ can be regarded as by-products of high-grade metamorphism with no direct significance in terms of the pattern of metamorphic facies.

The boundaries of the facies, or the regional isograd surfaces, are broadly flat lying over the greater part of the Northern and Central Grampian Highlands but steepen markedly against the Highland Boundary Fault, particularly in the east towards Stonehaven. This pattern of facies and facies boundaries led Kennedy (1948) to propose the concept of a ‘thermal anticline’ whose core is marked by the higher grade rocks and which plunges south-westwards. The distribution of the metamorphic facies is now generally considered to represent the peak conditions during the Grampian Event. Although there is evidence of a pre-Grampian (Knoydartian) regional metamorphic event in the Badenoch Group, no evidence of overprinting consistent with polyorogenic metamorphism has been identified in the facies pattern in Dalradian strata. No regionally significant areas of retrogressive metamorphism are present.

1.6.2 Age of metamorphism and relationship to ductile deformation

Elles and Tilley (1930) suggested that the metamorphism was broadly coeval with the early deformation, but that no metamorphism accompanied the later folds, which demonstrably fold the isograd surfaces. Later workers placed emphasis on the textural relationship between porphyroblast growth and deformational fabrics (e.g. Rast, 1958; Sturt and Harris, 1961; Johnson, 1962, 1963; Harte and Johnson, 1969). This work broadly showed that growth started during the early deformation and that the peak took place after the main nappe-forming or crustal thickening phases but before the later uplift phases, that is syn-D2 to late-D3 (Strachan et al, 2002). The late deformation was accompanied by only limited and localized retrogression, due to relatively rapid uplift of the succession.

A maximum age for the metamorphism is given by the age of the youngest rocks that have been affected, namely the top of the Trossachs Group, which Tanner and Sutherland (2007) gave as early Arenig (c. 477 Ma). The later stages of metamorphism in the North-east Grampian Highlands are given by a series of dates on syn- to late-metamorphic intrusions, such as the Strichen Granite (473 Ma, Oliver et al., 2000), the Insch Pluton of the North-east Grampian Basic Suite (470 Ma, Dempster et al., 2002) and the Aberdeen Granite (470 Ma, Kneller and Aftalion, 1987). A minimum age is given by dates from post-metamorphic granites such as Kennethmont at 457 Ma (Oliver et al., 2008). This accords with dates of 473-465 Ma for garnet growth (roughly from syn-D2 to syn-D3) in Glen Clova given by Baxter et al. (2002). Baxter et al. further proposed that the peak conditions in the kyanite, sillmanite and garnet zones were effectively synchronous.

In the Northern Grampian Highlands, the age of the metamorphism has been given as 470–450 Ma based on U-Pb monazite ages by Barreiro (quoted by Phillips et al., 1999).

Although conditions and exact timing varied across the Dalradian belt, in general metamorphism, that is porphyroblast growth, lasted about 10 Ma from c. 474 to c. 464 Ma, with peak conditions around 470 Ma. This accords with a date for peak conditions for Grampian tectonothermal events in Connemara in western Ireland of c. 470 Ma (Friedrich et al., 1999).

In the Grampian Highlands the main metamorphism was rapidly followed by cooling and uplift, which had occurred by at least c. 460 Ma (Dempster, 1985a; Dempster et al., 1995; Soper and Evans, 1997; Soper et al., 1999; Oliver et al., 2000). Dempster (1985a) further argued for significant uplift and cooling episodes at 460-440 Ma and 410-390 Ma.

1.6.3 Facies variations in space and time

Although the pattern of the Grampian metamorphism is broadly uniform throughout the Grampian Highlands there are significant variations. Fettes et al. (1976) documented an increase in geothermal gradient from the South-west to the North-east Grampian Highlands and P-T estimates have subsequently quantified those variations. Thus temperatures of c. 550°C were recorded for lower amphibolite-facies conditions across the Northern and North-east Grampian Highlands but at markedly different pressures in different areas: namely 9–10 kbar in the Schiehallion area (Baker, 1985), 7–8 kbar in western Aberdeenshire (Baker, 1985) and in the Monadhliath (Phillips et al., 1999), 5–6 kbar in Angus (Dempster, 1985a; Zenk and Schulz, 2004), and 3–4 kbar in Banffshire (Beddoe-Stephens, 1990). In addition, Graham (1983) recorded pressures of 8–10 kbar for epidote-amphibolite-facies rocks in the South-west Grampian Highlands, whereas similar facies rocks in Buchan record pressures of 2–3 kbar (Hudson, 1985; Beddoe-Stephens, 1990). Although these figures indicate systematic variations across the Grampian Highlands there are also significant variations in style and timing at a local scale with marked kinks in the P-T-t loops. Harte (1988) categorized these variations into six regions characterized by different metamorphic and tectonic styles with their boundaries coincident in part with major lineaments (Ashcroft et al., 1984; Fettes et al., 1986).

1.6.3.1 South-west Region

This region encompasses the area south-west of the Cruachan Lineament (Graham 1986). This lineament separates a higher density block to the south-west, dominated by the thick succession of the mafic Tayvallich Volcanic Formation, from a lower density block to the north-east with a markedly lower abundance of mafic intrusive rocks. The lineament also marks the south-western limit of granitic plutons.

The region is characterized by greenschist-facies metamorphism with a thin spine of garnet-bearing epidote-amphibolite-facies rocks running through the centre. The biotite isograd swings through 90° across the Cruachan Lineament from a north-west trend to the north to a south-west trend to the south (Figure 1.12). The metamorphism is characteristically of a low-temperature–high-pressure type compared to that in the rest of the Grampian Highlands, with reported temperatures ranging from 410 to 530°C (in garnet-bearing rocks) and pressures in the range 8–10 kbar (Graham and Harte, 1985). A later retrogressive phase is associated with lower greenschist-facies conditions and pressures of c. 6 kbar. Graham (1986) attributed this distinctive style of metamorphism, particularly the absence of high-temperature assemblages, to the relatively low conductivity and low heat-production potential of the mafic rocks. He also noted that the peak of metamorphism in the region was closely associated with the early deformation and thus relatively earlier than in the other regions.

1.6.3.2 South Perthshire Region

This region lies between the Cruachan Lineament and the Portsoy–Duchray Hill Lineament and extends northwards from the Highland Boundary Fault to the Tummel Steep Belt. It encompasses the Highland Border Steep Belt and the Flat Belt of the Tay Nappe. The metamorphism is represented in classical Barrovian zones, reflecting a regional increase in grade to the north. The main porphyroblast growth took place after the nappe-forming movements, essentially post-D2 to syn/post-D3 (Bradbury, 1979; Watkins, 1985). Dempster (1985a) contrasted the presence of chloritoid + biotite assemblages in Perthshire with their absence from the Angus glens to the east (South-east Region), indicating slightly lower pressures of metamorphism than in the latter.

The zone of late-stage albite porphyroblast growth extends from this area into the South-west Grampian Highlands. Watkins (1983) noted that the albite porphyroblasts occur in the crests of regional F3 folds and attributed their origin to a late-stage metamorphic phase driven by dehydration fluids trapped in the fold crests. However, Dymoke (1989) suggested that the albite porphyroblasts were a prograde product initiated by a phase of D3 movements in rocks at temperatures still close to their maximum.

Elles and Tilley (1930) noted inverted metamorphic zones in the Balquidder area with garnet-zone overlying biotite-zone rocks. Watkins (1985) argued that, since porphyroblast growth post-dated the major nappe-forming movements in this region, the inversion could not be tectonic. He suggested that the inversion represented a negative thermal gradient, attributed to emplacement of relatively hot rocks at higher levels during the nappe formation. It is also possible however, that garnet crystallization was either inhibited in the biotite zone by bulk-rock compositional control or was subsequently retrogressed by late-stage fluid movement. Watkins (1985) argued though that retrogression was not a significant factor.

Dempster (1985a) studied the cooling ages recorded by a number of mineral species along two north–south transects of the Highland Border. In this region he noted differential and spasmodic uplift; over most of the transect the rocks show an initial slow cooling phase from the peak metamorphic conditions followed by a period of rapid uplift and cooling.

1.6.3.3 South-east Region

This region extends from the Highland Boundary Fault northwards to the River Dee, east of the Portsoy–Duchray Hill Lineament, and is the classical area of Barrovian metamorphism (Barrow, 1893, 1912; Tilley, 1925). However, in the east of the region, lower pressure chloritoid + biotite assemblages are present, both transitional to, and within, the area of Stonehavian metamorphism (Harte and Hudson, 1979, fig. 2).

The main porphyroblast growth occurred from syn-D2 to syn-D3 with a late phase of sillimanite growth slightly post-D3 (Harte and Johnson, 1969; McLellan, 1985). Throughout the region, the isograd surfaces trend broadly parallel to the Highland Boundary Fault and steepen markedly against it. Chinner (1978) suggested that the cause of the steepening was underthrusting by cold material, but Harte and Hudson (1979) attributed it partly to late folding but also to the presence of a tectonic boundary, roughly coincident with the Highland Boundary Fault; a subsiding basin to the south brought cold rocks into contact with the northern sequences at the time of metamorphism.

In a transect of this region, Dempster (1985a) deduced that the northernmost or structurally deepest rocks recorded the oldest cooling ages, indicating relatively rapid uplift. In contrast, rocks rotated into the Highland Border Steep Belt by late folds show evidence of very slow cooling. He suggested that the rapid uplift of hot rocks from depth might have resulted in heat transfer to adjacent blocks, thus prolonging or promoting metamorphism in those blocks.

1.6.3.4 North Perthshire Region

This region encompasses the Tummel Steep Belt and extends northwards across the Central Grampian Highlands to the Boundary Slide. The rocks in this region are steeply dipping, in contrast to the flat-lying aspect of those to the south in the Flat Belt of the Tay Nappe. Dempster and Harte (1986) noted that the Barrovian zones, which show a progressive increase in metamorphic grade across the Flat Belt, are poorly developed within the Tummel Steep Belt. This they attributed to continued porphyroblast growth in the northern region; porphyroblast growth in the Flat Belt took place between D2 and D3 and syn-D3, whereas in the Tummel Steep Belt growth continued post-D3. They further calculated that the later porphyroblast growth in the north was accompanied by a significant increase in ambient pressure conditions, rising from 7 kbar (at 550°C) in the earlier growth phase to 9 kbar (at 550°C) in the later. This pressure increase was attributed to D3 rotation and deeper burial of originally flat-lying strata in the Tummel Steep Belt, thus promoting continued porphyroblast growth.

Dempster (1985a) suggested that the region cooled rapidly to c. 300°C in the period 460–440 Ma; the fact that the rocks in the Tummel Steep Belt do not record a temperature rise consequent upon their burial was attributed to uplift shortly afterwards.

1.6.3.5 Buchan Region

This is essentially the area of Buchan metamorphism (Read, 1952), encompassing the East Buchan and West Buchan sequences of Harte and Hudson (1979). The area lies east of the Portsoy–Duchray Hill Lineament and north of the River Dee. The Buchan region hosts a significant number of syn- to late-metamorphic granites as well as intrusions of the 470 Ma North-east Grampian Basic Suite, a unique occurrence in the Scottish Dalradian. The area is relatively simple in structure; the D1 and D2 phases have not given rise to major nappe structures, zones of inversion or crustal thickening such as are seen in other regions. In addition, the only other deformational phase identified (D3) is manifested by a late crenulation cleavage that demonstrably post-dates the main porphyroblast growth. The area is characterized by high-temperature–low-pressure metamorphic conditions; Hudson (1985) estimated temperatures of c. 430°C, 490°C and 510°C for the cordierite, andalusite and staurolite isograds respectively. Pressure estimates are uncertain but were reported by Hudson as 2–3.5 kbar in the staurolite zone.

The North-east Grampian Basic Suite was intruded after the early deformation, broadly coeval with peak metamorphic conditions (Johnson and Stewart, 1960; Johnson, 1962, 1963; Fettes, 1970; Pankhurst, 1970). Droop and Charnley (1985) estimated inner aureole conditions of 700–850°C and 4–5 kbar (granulite-facies hornfelses), indicating an emplacement depth of c. 15–18 km. The main regional porphyroblast growth is, in general, difficult to separate from the thermal effects. However, the intrusions undoubtedly boosted ambient regional temperatures and gave rise, for example, to the upper amphibolite-facies sillimanite gneisses around the Huntly–Knock Intrusion (Ashworth, 1975). Harte and Hudson (1979) discussed the possibility of two generations of sillimanite crystallization, the first of regional metamorphic origin, the second related to the gabbroic intrusions, although concluding that both were most probably part of a regional prograde event.

The current position of the kyanite/andalusite isograd is close to the Portsoy–Duchray Hill Lineament. West of this however, there is a zone up to 10 km wide indicative of a pressure increase, where andalusite has inverted to kyanite (Chinner and Heseltine, 1979; Baker, 1985; Beddoe-Stephens, 1990). Beddoe-Stephens noted that pressure conditions east of the Portsoy–Duchray Hill Lineament in Buchan did not exceed 4.5 kbar but that rocks to the west record significantly higher pressures of 8–9 kbar in structurally lower strata. He calculated that a pressure increase of c. 2 kbar occurred across the lineament during porphyroblast growth, and that it was this increase that led to the inversion of the regional andalusite to kyanite; the pressure increase was attributed to overthrusting on the Portsoy Shear-zone. Dempster et al. (1995) subsequently suggested that the pressure increase might have reflected magmatic loading induced by intrusion of the basic rocks. However the question arises as to whether the intrusions had sufficient mass to produce the necessary overpressures.

Considerable movement occurred on a regional system of shear-zones subsequent to the main porphyroblast growth, deforming the basic igneous rocks, their aureoles and the Dalradian country rocks that were unaffected by gabbro intrusion (Ashcroft et al., 1984). Kneller and Leslie (1984) demonstrated that shearing occurred in rocks at or close to their peak metamorphic conditions and led locally to crystallization of sillimanite (fibrolite) in rocks previously carrying andalusite + cordierite mineral assemblages; this crystallization was attributed to the percolation of hot fluids along the shear-zones.

1.6.3.6 Monadhliath Region

This region covers the remainder of the Grampian Highlands, coinciding broadly with the Northern Grampian Highlands as defined in this volume. It includes rocks assigned to the Badenoch Group and to the unconformably overlying Grampian and Appin group strata. The region contains such notable structural features as the Geal-charn–Ossian Steep Belt, the Eilrig Shear-zone, and the Grampian Shear-zone. The greater part of the area features amphibolite-facies assemblages, the exception being a strip of greenschist-facies rocks lying below the Eilrig Shear-zone ((Figure 1.12); Phillips et al., 1993). Phillips et al. (1999) reported a coherent metamorphic history across the region, with growth of biotite during D1 and kyanite early in D2. The general conditions for kyanite growth were given as 7–8 kbar and 500–600°C. In the northern part of the region, there was significant decompression during the later part of D2 resulting in conditions of 5–6 kbar and 585–695°C. This moved the rocks from the kyanite stability field into that of sillimanite. However, in the south of the region, no such changes have been observed and the rocks remained in the kyanite field. Phillips et al. (1993) suggested peak temperatures of 250°C for the greenschist-facies rocks below the Eilrig Shear-zone in contrast to 550°C for rocks above, suggesting that considerable crustal shortening affected the Grampian Group strata juxtaposed along the shear-zone.

Late-Neoproterozoic syntectonic porphyroblast growth and pegmatitic segregation is recorded in ductile shear-zones within the Badenoch Group and, locally, close to the contact with the younger Dalradian rocks (Hyslop, 1992; Hyslop and Piasecki, 1999). U-Pb monazite analyses from the pegmatites have provided high-precision ages of 808 +11/−9 Ma and 806 ± 3 Ma, and a concordant age of 804 +13/−12 Ma has been obtained from the host mylonite matrix (Noble et al., 1996). More-recent U-Pb dating of single zircon grains within kyanite-bearing migmatites has yielded an age of 840 ± 11 Ma (Highton et al. (1999). Together, these data have been interpreted by Highton et al. as the effects, in the Badenoch Group rocks, of the high-grade metamorphism and migmatization that is associated with Knoydartian orogenesis farther west in the Moine Supergroup of the Northern Highlands Terrane.

In contrast, radiometric evidence for Neoproterozoic tectonothermal events has yet to be recorded in the succeeding Dalradian rocks. There is evidence of progressive overstep of various lithologies onto the older strata (Smith et al., 1999; Robertson and Smith, 1999), and this, combined with ⁸⁷Sr/⁸⁶Sr isotopic signatures from the lowest Dalradian strata (Thomas et al., 2004), provides evidence for a significant stratigraphical and tectonothermal break at the base of the Grampian Group. The Badenoch Group rocks have thus been interpreted as a Moine-like metasedimentary basement that was affected by the Knoydartian tectonothermal event prior to deposition of the overlying Grampian Group, which records only Grampian Event deformation and metamorphism.

1.6.4 Metamorphic models

A great variety of models has been proposed to explain the pattern of metamorphism in the Grampian Highlands. These include the thermal effects of ‘older granites’ (Barrow, 1912), burial (Elles and Tilley, 1930), thermal zonation around a mountain root (Kennedy, 1948), uprising migmatite domes (Read and Farquhar, 1956), and self-generating heat in tectonically heated crust (Richardson and Powell, 1976). The latter is now generally accepted as an important cause, although it does not explain the very high heat flows in the area of least crustal thickening in Buchan. The high heat flow has been ascribed to deep-seated igneous masses (Harte and Hudson, 1979), and to lithospheric stretching that might have occurred during thermal relaxation immediately following Grampian arc accretion (Kneller, 1985). The latter concept is consistent with the extension necessary for intrusion of the gabbros and with the contrasting structural architecture.

Also within the Barrovian sequences of the south-east Grampian Highlands, several workers have noted the limited time-span of the metamorphism, the requisite rate of heating and the virtual synchroneity of porphyroblast growth in the different mineral zones, and have concluded that advective heat sources are essential (Baxter et al., 2002; Ague and Baxter, 2007; Lyubetskaya and Ague, 2010; Viete et al., 2011a, 2011b; Vorhies and Ague, 2011). Viete et al. (2011a, 2011b) argued for mid-crustal extensional shear-zones that focussed heat sources such as magmas and hot fluids, as well as generating heat through mechanical working; they suggested crustal thickening was not a significant factor. Vorhies and Ague (2011) argued for thermal relaxation of a thickened crust as a general factor in the Dalradian metamorphism but also for significant advective heat input in the Barrovian (sensu stricto) and Buchan regions, the heat input being facilitated by the numerous igneous intrusions and associated fluid flow. They noted the variation across the Dalradian belt, with the metamorphism in the South-west Grampian Highlands and western Perthshire being the product of a thickened crust with little or no additional heat source, whereas in the eastern Grampian Highlands (essentially the areas where the greatest temperatures were attained) crustal loading was significantly less and advective heat from igneous intrusions was the major factor in the metamorphism. Viete et al. (2011a, 2011b) and Vorhies and Ague (2011) noted spikes in the peak metamorphic conditions, for example the phases of sillimanite growth. This they ascribed to thermal pulses associated with the igneous intrusions.

These systematic changes along the Dalradian outcrop reflect variations in tectonism during the Grampian Event. In the south-west, compression and crust-thickening deformation was greater or lasted longer than in the north-east, where the later phases might have included a period of extension.

Following peak metamorphic conditions at around 470–467 Ma, the entire region was subject to differential and spasmodic uplift, resulting in rapid exhumation and cooling. Ague and Baxter (2007) also suggested that the cooling might have been hastened by the ingress of relatively cold fluids.

1.7 Dating the Dalradian sedimentation

D. Stephenson

The age of the sediments that became the Dalradian Supergroup is poorly constrained. Fossils are rare, are poorly preserved, and where species have been identified they have a wide stratigraphical range. Tillites and other possible glacigenic deposits raise various possibilities of correlation with global glacial events. As yet few radiometric age determinations have been made on the undoubted interbedded volcanic rocks and most available dates relate to intrusions or later tectonothermal events.

1.7.1 Palaeontology

The palaeontology of the Dalradian was reviewed by Downie et al. (1971) and Downie (1975) and there have been few significant developments since then.

In the Appin Group, worm burrows recorded from quartzites by Peach and Horne, (1930) have subsequently been shown to be dewatering structures (Tanner, 1998a) but algal stromatolites have been demonstrated in the Lismore and Lossit limestones (Spencer and Spencer, 1972). The Lossit Limestone has also yielded oncoliths. None are of biostratigraphical value.

In the Argyll Group, the Bonahaven Dolomite Formation has yielded acritarchs and algal stromatolites (Hackman and Knill, 1962; Spencer and Spencer, 1972), including some which suggest a late-Neoproterozoic age (Downie, 1975). Fairchild (1977) recorded possible glauconitized microfossils, but a report by Brasier and McIlroy (1998) of metazoan faecal pellets was subsequently retracted (Brasier and Shields, 2000). The Easdale Slate contains long-ranging, Ediacaran to Cambrian acritarchs (Downie, 1975). Limestone clasts in the Selma Breccia contain oncoliths, catagraphs and other calcareous fossils, which have been assigned to the Ediacaran (Litherland, 1975). However, calcareous and burrowing algae that resemble Early Cambrian and younger forms are also present (Downie, 1975). Acritarchs from the Tayvallich Limestone, which were originally thought to be Early Cambrian (Downie et al., 1971), are now known to have a longer range, extending back into the Ediacaran.

In the Southern Highland Group the search for fossils has concentrated upon the weakly deformed, greenschist-facies metamudstones and metalimestones of the Highland Border area and Turriff Syncline. The Macduff Formation has yielded rare burrows, a few acritarchs and more-widespread microfossils, which resemble highly altered chitinozoa. Downie et al. (1971) tentatively identified chitinozoa of early Ordovician, probably Llanvirn, age, and a single specimen of the Ordovician acritarch, Veryhachium lairdii, was identified by Molyneux (1998). However, failure to replicate this find and its unusually pristine state of preservation despite metamorphism to biotite grade cast doubt upon its origin. In fact all of these identifications remain controversial and have not gained general acceptance. Similarly, a record of Silurian graptolites by Skevington (1971) is now attributed to a sample labelling error. Possibly one of the most striking features of these generally low-grade rocks is their lack of macrofossils and trace fossils which has led several workers to suggest that they must all be of Neoproterozoic age.

The fossiliferous limestone of Leny Quarry near Callander, which contains latest Early Cambrian Pagetid trilobites (Pringle, 1940; Rushton et al., 1999), is undoubtedly the most reliable palaeontological indicator of the age of the youngest Dalradian strata in Scotland. This outcrop was once regarded as part of the Highland Border Complex, in faulted contact with the Southern Highland Group (Bluck et al., 1984; Curry et al., 1984). However, a detailed examination of key sections, and in particular that in the Keltie Water GCR site, near Callander, has resulted in a radical re-appraisal of the relationship between the Highland Border Complex and the Southern Highland Group (Tanner, 1995; Tanner and Pringle, 1999; Tanner and Sutherland, 2007). It is now accepted that there is both stratigraphical and structural continuity between the Southern Highland Group and the Keltie Water Grit Formation, which includes the Leny Limestone. The south-easterly younging sequence continues through rocks containing mid-Arenig fossils (Rushton et al., 1999), but reports of supposed chitinozoa of younger, Caradoc–Ashgill age by Burton et al. (1983) and Curry et al. (1984) have now been discounted (Tanner and Sutherland, 2007). Thus, Dalradian sedimentation continued to at least Early Cambrian times and possibly through to Arenig times. The succession of Dalradian Supergroup plus at least part of the Highland Border Complex was then deformed as one, during the Llanvirn-Age Grampian Event.

1.7.2 Correlation of the Dalradian tillites and other possible glacigenic deposits

The tillites at the base of the Argyll Group constitute one of the most persistent stratigraphical markers throughout the length of the Dalradian outcrop. Traditionally they have been correlated with the Varanger tillites in Scandinavia, which had been dated by Rb-Sr methods on associated metasedimentary rocks at c. 653 Ma (Pringle, 1972). However, revised estimates for the age of the Varanger tillites (see below) now suggest that they are probably younger than the Tayvallich volcanic rocks at the top of the Argyll Group, necessitating a re-appraisal of the age of all possible glacigenic events represented in the Dalradian successions of Scotland and Ireland.

Prave (1999), Brasier and Shields (2000), Condon and Prave (2000) and McCay et al. (2006) have all proposed that the basal Argyll Group tillites can be correlated with the Ghubrah glaciation in the Oman, dated at 723 +16/−10 Ma (Brasier et al., 2000), which is part of the Sturtian global glacial event. If however, as is argued below, the base of the Grampian Group is unlikely to be older than c. 750 Ma and might even be as young as c. 700 Ma, this correlation would allow little or no time for the accumulation of the Grampian and Appin groups. Even the localized Kinlochlaggan Boulder Bed, interpreted as containing glacially-rafted dropstones and thus recording the earliest glacial influence in the Dalradian, is assigned to the Lochaber Subgroup, is hence almost certainly younger than 720 Ma and cannot be correlated with the Sturtian event. The most likely scenario, based upon currently available worldwide data and proposed by Leslie et al. (2008), is that the basal Argyll Group tillites should be equated with the Ghaub glaciation in Namibia at c. 635 Ma (Hoffmann et al., 2004), which is regarded as part of the Marinoan global event.

The younger, Southern Highland Group glacigenic dropstone deposits in Inishowen (Donegal, Ireland) and Macduff (North-east Grampian Highlands) could then probably be correlated with the Varanger tillites (now 620–590 Ma, Gorokhov et al., 2001; Bingen et al., 2005) and the Gaskiers Formation of Newfoundland (c. 580 Ma, Bowring et al., 2003).

1.7.3 Radiometric dating and isotopic evidence

Age constraints on the initiation of Dalradian sedimentation are poor. The youngest detrital zircons in Dalradian rocks have yielded ages of 900 Ma (Cawood et al., 2003; Banks et al., 2007) and the Grampian Group must be younger than the 800 Ma pegmatites within the Badenoch Group that constitutes its basement. The youngest deformation event recorded within the Moine rocks to the north-west of the Great Glen Fault is c. 730 Ma (Tanner and Evans, 2003; Emery, 2005) and it is possible that this might also have affected the lithologically similar Badenoch Group rocks. However, there is no radiometric evidence for Neoproterozoic tectonothermal activity in the Grampian Group or higher parts of the Dalradian Supergroup, which record only Caledonian events. ⁸⁷Sr/⁸⁶Sr whole-rock isotope data from the oldest metacarbonate rocks of the Grampian Group are consistent with a global late-Neoproterozoic strontium sea-water signature younger than 800 Ma and possibly as young as c. 670 Ma (Thomas et al., 2004). Given the mounting evidence for a significant stratigraphical and tectonothermal break at the base of the Grampian Group (Smith et al., 1999), the current consensus is that the base is unlikely to be older than c. 750 Ma and sedimentation could have been initiated as late as c. 730–700 Ma (Leslie et al., 2008).

Later parts of the Grampian Group have been shown to be stratigraphically continuous with the Lochaber Subgroup in several areas and in general there would then appear to be stratigraphical continuity throughout the Appin Group, with no obvious major tectonic breaks. There are no radiometric dates that cover this interval, which continued at least until the major glacial event at the base of the Argyll Group that could be correlated with global glaciations at c. 635 Ma (see above). The time interval represented by this marine glacial interval is not known, but no major unconformity is seen.

At the top of the Argyll Group, the Tayvallich magmatism has been reliably dated by high-precision U-Pb methods on zircons at 600–595 Ma (Halliday et al., 1989; Dempster et al., 2002).

Estimates of minimum radiometric ages for younger parts of the Dalradian depend upon the dating of post-sedimentary tectonothermal events, in particular granites that were intruded into the Dalradian succession and then experienced the same deformation and metamorphism.

Much attention has centred upon the Ben Vuirich Granite Pluton, near Pitlochry. For many years it was believed that this granite was intruded after at least some of the earlier phases of deformation of its Dalradian host rocks (Bradbury et al., 1976; Rogers et al., 1989; Tanner and Leslie, 1994). However, detailed investigations of hornfelsed country rocks and xenoliths within the granite have now suggested that the pluton was emplaced into a previously undeformed sedimentary sequence of the Blair Atholl Subgroup (Tanner et al., 2006). S1 and S2 fabrics, previously thought by some to have been preserved only in the aureole and xenoliths, and hence to be related to Neoproterozoic events, can now be assigned to the mid-Ordovician Grampian Event. The 590 Ma age of intrusion, obtained by U-Pb dating of carefully selected zircons (Rogers et al., 1989) and confirmed by ion-microprobe dates on individual zircons (Pidgeon and Compston, 1992), is now thought to be much closer to the age of sedimentation of the country rocks but prior to any significant tectonic deformation (Tanner et al., 2006). Farther north, in the North-east Grampian Highlands, the Keith–Portsoy Granite was intruded as a number of separate lenticular sheets into Appin Group and lower Islay Subgroup rocks. Zircons from two separate lenses have yielded precise U-Pb intrusion ages of c. 600 Ma (Barreiro, 1998).

The Ben Vuirich and Keith–Portsoy granites, together with several other intrusions in both the Grampian and Northern Highlands terranes, comprise the 600–590 Ma Vuirich Suite, which has been shown to have an A-type chemistry (Tanner et al., 2006). This rift-related granitic suite was broadly coeval with the 595 Ma Tayvallich basic magmatism that marks the top of the Argyll Group, and both suites were probably related to a major extensional event that was part of the break-up of the supercontinent of Rodinia, leading eventually to the formation of the Iapetus Ocean (Cawood et al., 2001). Sedimentation of the upper parts of the Dalradian succession took place in marginal basins and on continental slopes resulting from this ‘Iapetan Event’.

1.7.4 Conclusions

The detailed chronological interpretation is constantly changing as more and better radiometric dates become available and as other methods are developed, but a reasonable consensus is beginning to emerge.

The Badenoch Group successions must be older than the c. 800 Ma pegmatitic veins and mylonites that cut them and they might have been affected by the same 730 Ma tectonothermal event that has been recorded in the Moine successions of the Northern Highlands Terrane. The earliest Grampian Group sedimentation probably post-dates that late-Knoydartian Event, and strontium sea-water signatures of the oldest metacarbonate rocks suggest an age within the range 800–670 Ma. A current best estimate for the age of the basal Grampian Group is 730–700 Ma.

If the tillites at the base of the Argyll Group are correlated with the Marinoan glaciations at 635 Ma, close to the start of the Ediocaran Period, the Grampian and Appin groups together cover a maximum interval of c. 95 Ma in the late Cryogenian Period. As yet there is no evidence to enable any chronological subdivision of those groups. The top of the Argyll Group is well defined by the Tayvallich basic magmatism at 600–595 Ma, implying that the group spans some 40 Ma, all well within the Ediocaran.

Granites of the Vuirich Suite were intruded into lower Argyll Group successions at 600–590 Ma, whilst the uppermost Argyll Group and lowest Southern Highland Group sediments were being deposited and prior to any significant tectonic deformation or metamorphism.

It is now accepted by most workers that there is stratigraphical and structural continuity in the Southern Highland Group of the Highland Border region from the Ben Ledi Grit Formation through into the Keltie Water Grit Formation, and the Lower Cambrian Leny Limestone in particular (c. 515 Ma on current timescales). Therefore, the base of the Cambrian (at 542 Ma) must lie within the Southern Highland Group, though there is no obvious stratigraphical horizon where it may be located. Although this part of the succession has a total thickness of several kilometres, the general turbiditic nature of the sediments suggests relatively rapid accumulation, and it is difficult to imagine it taking some 80 Ma for them to accumulate. There are no obvious major tectonic breaks and significant stratigraphical hiatuses are difficult to recognize in these facies. It is also possible that sedimentation continued through to mid-Arenig time in the Trossachs Group (c. 477 Ma on current timescales).

However. the above model does allow time for the deformation and metamorphism to reach its peak at c. 470 Ma in the mid-Ordovician Grampian Event.

1.8 Tectonic evolution of the Dalradian basins

D. Stephenson

For much of Proterozoic time, the crustal foundations of Scotland were part of a continental block that also incorporated parts of present-day North America and Greenland. At the beginning of Neoproterozoic time, this block was included within the large supercontinent of Rodinia, where it lay adjacent to blocks that later became the Baltic shield and South America (Figure 1.13). It was only towards the end of Neoproterozoic time that Rodinia became fragmented and the three blocks became respectively the separate continents of Laurentia and Baltica and part of Gondwana. However, many authors have used those names to refer to the respective blocks of continental crust, even prior to their development as separate continents.

Sedimentation of the lower parts of the Dalradian Supergroup recorded events that took place within Rodinia, with later sedimentation recording increased instability and eventual break-up that left Laurentia separated from Gondwana and Baltica by the newly formed Iapetus Ocean. The latest Dalradian sedimentation and volcanism took place in basins on the Laurentian margin of the ocean. Later plate movements, during Early Palaeozoic time, resulted in closure of the Iapetus Ocean and the Caledonian Orogeny. Much evidence for the sequence of orogenic events is recorded by the deformation and metamorphism of the Dalradian strata.

Good, recent overviews of the tectonic setting and evolution of this area are given by Holdsworth et al. (2000), Strachan et al. (2002) and Leslie et al. (2008), on which much of this account is based.

1.8.1 Palaeoproterozoic and Mesoproterozoic events

Little is known about the nature of the crust that was stretched and rifted to form the Dalradian basins. The few areas of basement rock that are exposed within the Grampian Terrane are mostly fault bounded and their tectonostratigraphical affinities are unclear (see above). Despite the present relatively close proximity of the Lewisian Gneiss Complex in the Northern Highlands and Hebridean terranes, there is no evidence for the presence of Archaean crust beneath the Dalradian. Archaean detritus is present throughout the higher parts of the succession (Cawood et al., 2003), but the source has not been identified and could well be distant. However, large amounts of juvenile crust are known to have been generated by Palaeoproterozoic arc magmatism to the south of major Archaean cratons, in both the ‘Laurentian’ and ‘Baltic’ crustal blocks. Examples include the Ketilidian Belt (1900–1800 Ma) of Laurentia and the Svecofennian Province (1900–1850 Ma) of Baltica. The Rhinns Complex of Islay and Colonsay (c. 1780 Ma) is probably representative of a similar Palaeoproterozoic belt that might underlie much of the Dalradian.

Towards the end of the Mesoproterozoic Era, from c. 1100 to 950 Ma, a series of mountain belts was formed around the globe. It was this mountain building event, generally referred to as the Grenvillian after its type area in North America, that led to amalgamation of the supercontinent of Rodinia. Palaeomagnetic reconstructions generally place the crustal blocks that were to become Baltica, Laurentia and the South American sector of Gondwana adjacent to each other, with Scotland positioned close to the triple junction (e.g. Soper, 1994a, 1994b; Torsvik et al., 1996; Dalziel, 1997; Holdsworth et al., 2000) (Figure 1.14)a. A Grenville Front is well established in eastern Canada (e.g. Indares and Dunning, 1997) and the effects of the Sveconorwegian Orogeny, of similar age, are widespread in south-western Sweden (Möller, 1998). However, in Scotland, evidence for Grenvillian orogenesis is sparse. It is restricted to the Glenelg–Attadale Inlier, in the Northern Highlands Terrane and the Outer Hebrides Fault-zone, in the Hebridean Terrane (Mendum et al., 2009). The only pre-Grenvillian rocks in the Grampian Terrane, in the Rhinns Complex, show no evidence of Grenvillian reworking and hence it is concluded that the Grenville Front lay farther to the north, buried beneath the extensive cover of post-Grenville metasedimentary rocks.

1.8.2 Pre-Dalradian Neoproterozoic sedimentation and tectonothermal events

The Badenoch Group successions of the northern Grampian Highlands are now generally accepted as having been derived from sedimentary rocks that experienced a tectonothermal event at 840–800 Ma and hence pre-date the earliest Dalradian sediments, which show no evidence of such an event. The generally gneissose and migmatized lithologies yield little evidence of their origin but they undoubtedly comprise one of several thick sedimentary successions that were deposited on the ‘Laurentian’ crustal block of Rodinia towards the end of and immediately following the Grenvillian Orogeny. Representative successions in the Northern Highlands Terrane are the Torridon Group (fluviatile) and the Moine Supergroup (fluviatile and shallow marine); sediments of both contain very few Archaean detrital zircons and it has been suggested that they were derived largely from erosion of the Grenvillian mountain belt (Rainbird et al., 2001; Krabbendam et al., 2008).

Outcrops of the Badenoch Group now lie close to those of the Moine Supergroup and they are comparable in lithology to parts of the Moine succession. Their present proximity is largely as a result of terrane assembly during the Caledonian Orogeny and hence direct correlations across the Great Glen Fault are not possible. However, their detrital zircons (Cawood et al., 2003) have a similar age distribution to those of the Morar Group of the Moine and the Torridon Group (Karabbendam et al., 2008) and their broad tectonic setting and age are probably comparable.

The age of the Moine is constrained by its youngest detrital zircons at c. 950 Ma (Kinny et al., 1999; Friend et al., 2003), and by later intrusions at c. 870 Ma (Friend et al., 1997; Millar, 1999; Rogers et al., 2001). The intrusions, of granitic sheets and tholeiitic basic rocks, constitute a bimodal igneous event, typical of an intracontinental rift setting (Ryan and Soper, 2001), and it probably represents the development of a failed rift between the ‘Baltic’ and ‘Laurentian’ crustal blocks.

The earliest known tectonothermal event in the Badenoch Group is recorded by a U-Pb age of c. 840 Ma on zircons within migmatites (Highton et al., 1999) and from U-Pb ages of c. 806 Ma on monazites from pegmatites and their host mylonites (Noble et al., 1996). [The pegmatites had previously yielded less-precise ages of c. 750 Ma by Rb-Sr methods (Piasecki and van Breemen, 1979a, 1983).] Those dates correlate well with numerous dates in the range 820–780 Ma on pegmatites and metamorphic minerals within the Moine which, together with a separate cluster of dates at 750–730 Ma, have been attributed to a major Knoydartian Orogeny (e.g. Rogers et al., 1998; Tanner and Evans, 2003). The extent and nature of those events have been the source of much controversy. Soper and England (1995) envisaged that the Neoproterozoic evolution of the Scottish part of the Laurentian crustal block was dominated by rifting and extension, without any contractional orogenic events. However, by linking radiometric age dating with metamorphic pressure and temperature data, recent work has shown that at least some of the Knoydartian events are contractional (Vance et al., 1998; Zeh and Millar, 2001; Tanner and Evans, 2003). Orogenic igneous activity occurred at c. 840–800 Ma in northern Norway (Daly et al., 1991) and sinistral strike-slip motion between the ‘Baltic’ and ‘Laurentian’ crustal blocks at c. 800–750 Ma, as suggested by Park (1992), might have severed an original close association with the Knoydartian events of Scotland.

1.8.3 Late-Neoproterozoic Dalradian basins

The lithostratigraphy and sedimentology of the Dalradian Supergroup indicate periods of basin deepening and shallowing which, by analogy with Phanerozoic basins, have been attributed to multiple periods of lithospheric stretching, rifting and thermal subsidence. Early phases of rifting and subsidence, possibly from c. 730 Ma, took place within the supercontinent of Rodinia. Increasing instability, attributed to progressive lithospheric stretching, first produced a series of fault-bounded basins and eventually culminated in continental rupture, volcanicity and development of the Iapetus Ocean at c. 600 Ma (Figure 1.14)b. The Grampian Terrane then became part of an extensive passive margin to the new continent of Laurentia, where turbiditic sedimentation continued for at least 85 Ma, and possibly until the early Ordovician (Figure 1.14)c.

Several studies have addressed basin architecture and its influence on both the composition and shape of the sediment pile, factors that in turn have affected the morphology of some regional folds. In the Northern Grampian Highlands, basin margins within the Grampian Group and lower parts of the Appin Group can be identified from thickness and facies changes and overstep onto older strata (Glover et al., 1995; Goodman et al., 1997; Smith et al., 1999; Robertson and Smith, 1999). In the South-west Grampian Highlands, Knill (1963), Borradaile (1979), Litherland (1980) and Anderton (1985) identified very rapid lateral facies and thickness changes in the Appin Group and lower parts of the Argyll Group, which imply that syndepositional faulting accommodated the deposition of the essentially shallow-water shelf deposits. Anderton (1985), in particular, envisaged this area of deposition to be divided into a series of NW-dipping fault blocks bounded by ‘scoop-shaped’ listric faults, which delimited individual basins (Figure 1.15). The faults might also have acted as controls for later sub-marine volcanic activity, and were eventually draped by rapidly deposited sub-marine fan deposits on a subsiding continental margin in Southern Highland Group times.

More-recent studies have concentrated on tracking the evolution and denudation history of the hinterland, through plotting the age distribution and frequency of detrital zircons obtained by Sensitive High-Resolution Ion Micro Probe (SHRIMP) analysis (e.g. Cawood et al., 2003, 2007).

1.8.3.1 Early rifting

Rifting of the ‘Laurentian’ crustal block was probably initiated during the middle of the Cryogenian Period at some time after 730 Ma and lasted for about 60–70 Ma. Turbiditic sands and muds of the Grampian Group were derived from a hinterland to the west and north-west (Banks, 2005), with at least one basin receiving detritus from the east (Banks et al., 2007). Early models that envisaged a broad ensialic rift, opening north-eastwards to form a marine gulf (Winchester and Glover, 1988), were refined following the recognition of discrete intrabasinal highs separating a series of localized basins (e.g. Robertson and Smith, 1999; see (Figure 5.2)a). Those early (presumably failed) rift basins had been infilled by the time that deltaic and shelf sands prograded from the south and east across their margins in late Grampian Group and early Appin Group times. The sand- and mud-dominated sedimentary facies associations culminated with diminished sediment input into marginal offshelf to lagoonal, locally emergent, even evaporitic environments in Lochaber Subgroup time (Stephenson, 1993). Ages of detrital zircons show that the source areas throughout Grampian Group and earliest Appin Group sedimentation were dominated by late-Palaeoproterozoic to earliest-Neoproterozoic rocks, possibly from the Ketillidian and Grenvillian mountain belts, such as those that had supplied earlier Neoproterozoic successions in the Northern Highlands Terrane (Cawood et al., 2003; Banks et al., 2007).

1.8.3.2 Post-rift thermal subsidence

Shallow-marine shelf sedimentary facies associations that dominate the remainder of the Appin Group are characteristic of post-rift thermal subsidence, which might have extended for c. 30–40 Ma until the end of the Cryogenian Period at c. 635 Ma. The base of the Ballachulish Subgroup is marked by a major transgression across a flooding surface, which saw the onset and development of wide-ranging and uniform sedimentary lithofacies (Figure 1.15)a. Typically, deposition of dark anoxic limestone and mud is followed by shallowing upwards cycles of progradational clean-washed sands and shallow-water muds and limestones. Marine basins at this time were probably wide and shallow as evidenced by remarkably similar successions extending along strike across Ireland and Scotland. Renewed flooding resulted in further deposition of muds and limestones during Blair Atholl Subgroup time, before glacial diamictites were deposited during a major lowstand at c. 635 Ma. In the Central Grampian Highlands, some parts of these basins were interspersed with sediment-starved areas, now expressed by major stratigraphical omissions and subsequent overstep. The flooding at the base of the Ballachulish Subgroup introduced, or just preceded, the arrival of Archaean zircons in detritus and an apparent absence of Grenvillian ones.

1.8.3.3 Renewed rifting

Vigorous extension is recorded by renewed rifting in early-Ediacaran time. Sharply-defined thickness changes become apparent in the Islay Subgroup and are most-likely to have been fault controlled (Anderton, 1985). Instability is recorded by influxes of pebbly sands, debris-flow breccias and slump deposits in the Easdale Subgroup. Localized basaltic volcanic centres became a widespread feature for the first time (e.g. Goodman and Winchester, 1993; Fettes et al., 2011) and synsedimentary exhalative mineralization affected some small basins. Within this increasingly unstable environment the upward-shallowing cyclical behaviour, recorded previously in the Appin Group, recurred briefly in late Easdale Subgroup time, with deposition of shallow-marine shelf sands, limestones and muds. It was only following this stage that the sediment-starved sectors in the Central Grampian Highlands were overstepped as a more rapidly foundering rift system evolved during Crinan Subgroup time (Figure 1.15)b. A newly established trough received a deluge of immature siliciclastic sediment that became increasingly dominated by Archaean detritus.

1.8.3.4 Rift–drift transition; the Iapetan extensional event

The rifting of Rodinia during early-Ediacaran (Argyll Group) time eventually led to complete rupture of the continental crust and the formation of the Iapetus Ocean, as the new continents of Baltica and Gondwana drifted away from Laurentia (Soper, 1994b; (Figure 1.14)c). [Iapetus was the father of Atlas in Greek mythology.]

The timing of this rift–drift transition is generally considered to be tightly constrained by a sudden increase in magmatism at c. 600 Ma. Localized episodes of igneous activity had punctuated periods of basin deepening throughout Argyll Group time, but more-voluminous basic volcanism in the South-west Grampian Highlands dominated the Tayvallich Subgroup and eruption continued more-sporadically during deposition of the Southern Highland Group (Figure 1.15)c. All of the Dalradian lavas have tholeiitic affinities similar to those formed at accreting oceanic plate margins (Graham and Bradbury, 1981), but the more-voluminous later eruptions show evidence of a more-enriched, deeper mantle source that was able to rise to higher levels as continental rupture progressed (Macdonald et al., 2005; Fettes et al., 2011). Anderton (1985) observed that the thickest volcanic sequences overlie areas of supposed greatest crustal attenuation and it might also be significant that the most-enriched magmas were erupted close to major structures such as the Cruachan and Portsoy lineaments. The Tayvallich basic volcanic rocks have been dated precisely at around 600–595 Ma (Halliday et al., 1989; Dempster et al., 2002), coeval with the foliated, mildly alkaline granites of Keith–Portsoy at 600 Ma and Ben Vuirich at 590 Ma (Rogers et al., 1989). Together they define a major bimodal magmatic event, typical of an extensional rift environment (Tanner et al., 2006).

The onset of the Iapetan extensional event is also recorded in the Northern Highlands Terrane, where a variety of igneous rocks were intruded into the Moine Supergroup between 610 and 590 Ma (e.g. Kinny et al., 2003b). As in the Grampian Terrane, this magmatism was distinctly bimodal, with both basic and silicic intrusions, some with alkaline affinities, and all compatible with a continental rift setting.

The rifting was also heralded by basic (and possibly ultrabasic) volcanism in Shetland which, according to Flinn (2007), was followed very shortly afterwards by burial metamorphism. The metamorphism was boosted by high heat-flow through the stretched and thinned crust, prior to oceanic opening. As the Laurentian passive margin developed, the Dalradian strata become tilted (in the manner of present day ‘seaward-dipping seismic reflectors’), were eventually rotated into a vertical orientation and suffered further metamorphism and granitic veining locally under the influence of permeating hydrous fluids. The upper part of the vertical ‘slab’ was then rotated further to form the inverted limb of a huge monocline that is the main structure of the Shetland Dalradian. Flinn’s interpretation places this whole sequence of events in an extensional regime, related to Iapetus opening, some 100 Ma before the Grampian Event affected the Grampian Highlands, possibly overprinting and obscuring any earlier tectonothermal events in that region.

A major change to more-immature sediment, increasingly dominated by Archaean detritus, had already occurred at the base of the Crinan Subgroup. From that time onwards, Palaeoproterozoic zircons became less-abundant in the sediment load, suggesting that either any c. 1800 Ma Ketillidian/Rhinnian source had become isolated from the Dalradian basins by the Iapetan rifting or, more likely, that it had been buried by sedimentation (Leslie et al., 2008). During Tayvallich Subgroup time, volcaniclastic aprons built out from the volcanic centres and carbonate build-ups on the shelf were reworked into deeper water as redeposited metacarbonate rocks (Thomas et al., 2004). In Southern Highland Group time, as the Iapetus Ocean began to widen and the continental margin foundered, turbiditic submarine fans prograded from the margin into the trough and possibly extended along the trough axis, spilling out onto adjacent marginal platforms. Localized volcanic activity continued and volcaniclastic debris was reworked along the margin as ‘green beds’ (Pickett et al., 2006).

1.8.4 Cambrian to early Ordovician sedimentation

It is now generally accepted that deep-water turbiditic sedimentation on the Laurentian side of the Iapetus Ocean continued throughout the later parts of the Ediocaran Period at least until latest Early Cambrian time (cf. the Leny Limestone) and possibly into the early Ordovician (mid Arenig) (Tanner, 1995; Tanner and Pringle, 1999; Tanner and Sutherland, 2007; see the Introduction to Chapter 4). The younger rocks, currently assigned to the Highland Border Complex, occur only in fault-bounded slivers along the Highland Boundary Fault. They include pillow lavas and remnants of a fragmented ophiolite-complex that must have been obducted onto the margin of the Grampian Terrane at some stage during the Grampian Event (Tanner, 2007; see below).

Coeval with this upper Southern Highland Group–Highland Border Complex succession is the undeformed, non-metamorphosed Cambrian to mid-Ordovician succession of the Hebridean Terrane, described in GCR Volume 18 (Rushton et al., 1999). This sandstone and carbonate sequence (the Ardvreck and Durness groups) rests upon a remarkably planar basal unconformity and represents a major marine transgression onto a wide, shallow shelf on the subsiding Laurentian margin.

The two successions, together with their correlatives in East Greenland and north-west Newfoundland, formed subparallel sedimentary belts located along the continuous passive margin of Laurentia (Figure 1.14)c. The shelf sequences were deposited on the landward side of the generally deeper water Dalradian-type lithologies that accumulated on the continental slope and rise. Oceanic crust is presumed to have existed to the south-east of the present Dalradian outcrop, where it might ultimately have been covered by progradation of the youngest sediments. All along the former margin, these two successions are now in much closer proximity than they were at their time of deposition due to the effects of crustal shortening during the Caledonian Orogeny. Whether or not there was originally a continuous transition between them has been a matter of debate (e.g. Dewey, 1969; Bluck et al., 1997; Leslie et al., 2008). However, the ages of their detrital zircons are distinctly different (Cawood et al., 2007), reflecting derivation from different sectors of the Laurentian margin. Palaeocurrent data suggests that the Dalradian sediments were supplied by currents from the south-west, along troughs parallel to the Laurentian shoreline (Anderton, 1985). These could have sampled Mesoproterozoic rocks of the Grenvillian mountain belt currently exposed in eastern Canada, which are not represented by zircon ages in samples from the Ardvreck Group. The sediments of the latter group were transported by currents flowing from the north-west, across the shelf and transverse to the shoreline (McKie, 1990), from a hinterland to the west composed almost entirely of Palaeoproterozoic and Archaean rocks.

1.8.5 The Caledonian Orogeny

The Iapetus Ocean reached its greatest width, of about 2000 km, during late Cambrian time at around 500 Ma. It then began to close rapidly due to the renewed convergence of Laurentia with Baltica and with a microcontinent called Avalonia that separated from Gondwana (Soper and Hutton, 1984; Pickering et al., 1988; Soper et al., 1992) (Figure 1.14)d. The combined tectonic and magmatic effects of that convergence, leading eventually to continent–continent collision and subsequent uplift, constitute the Caledonian Orogeny. Prior to the more-recent opening of the North Atlantic Ocean, the resultant Caledonian mountain belt stretched continuously from the Appalachians, through Newfoundland and the British Isles to East Greenland and north-west Scandinavia (Figure 1.3). The western Caledonian Front can be followed from East Greenland, through the Moine Thrust Belt of north-west Scotland to Newfoundland, whilst the eastern front runs through Sweden and Norway. In southern Britain a complex Caledonian Front lies buried beneath younger strata.

Within this broad orogenic framework, many separate events have been identified, several of which have been given specific names. Of most relevance to the Dalradian are the mid-Ordovician peak of deformation and metamorphism, accompanied by localized basic magmatism and crustal melting, termed the Grampian Event, and a period of later folding, uplift and generation of huge volumes of largely silicic magma, during the Silurian, termed the Scandian Event. The Early- to Mid-Devonian Acadian Event, which was responsible for most of the deformation in northern England and Wales, might have been the cause of re-activation of the major bounding faults of the Scottish Highlands (Mendum and Noble, 2010).

The GCR sites in this volume represent Caledonian deformation and metamorphism in the Grampian Terrane. GCR volumes 3 (Treagus, 1992) and 34 (Mendum et al., 2009) represent deformation and metamorphism elsewhere in Britain, and Volume 17 (Stephenson et al., 1999) represents Caledonian magmatism throughout Britain.

1.8.5.1 The Grampian Event

The mid-Ordovician Grampian Event was the first collisional event to affect the Scottish part of the Laurentian margin, and was the first phase of the Caledonian Orogeny (Lambert and McKerrow, 1976). This event is now widely regarded as having been caused by the collision of an arc terrane against the passive margin (Dewey and Shackleton, 1984; van Staal et al., 1998; Oliver, 2001; see (Figure 1.14)d) and (Figure 1.16). In Scotland, there is only indirect evidence that an arc is buried beneath post-Ordovician sedimentary cover in the Midland Valley (Bluck, 1983, 1984) and might also occur at depth beneath part of the Highland Border region of the Grampian Highlands. However, parts of an early Cambrian to early Ordovician island arc, together with a suprasubduction ophiolite, are exposed in western Ireland (Dewey and Ryan, 1990). Most Grampian deformation and metamorphism in Scotland occurred in the Grampian Terrane. In the Northern Highlands Terrane, most deformation appears to be Knoydartian (see above) or Scandian (see below), but some significant Grampian tectonic and metamorphic effects have also been proved (Kinny et al., 1999; Dallmeyer et al., 2001; Rogers et al., 2001; Emery, 2005). A comparable mid-Ordovician orogenic event in the Appalachians of eastern North America is known as the Taconic Event (Dewey and Shackleton, 1984).

In Cambrian and early Ordovician times, the Scottish sector of the Laurentian margin consisted of a shallow-water shelf that developed on the foreland of the Hebridean Terrane. This passed south-eastwards into the deep-marine turbidite basins of the Southern Highland Group (Figure 1.16)a, where sedimentation was terminated in the mid Ordovician by crustal flexure and/or uplift associated with the Grampian Event (e.g. Soper et al., 1999). The arc–continent collision is thought to have resulted in the obduction or overthrusting of exotic ophiolitic nappes (Figure 1.16)b, accompanied by regional deformation and Barrovian metamorphism of the Dalradian rocks (Dewey and Shackleton, 1984; Dewey and Ryan, 1990; Chew et al., 2010). The best-preserved ophiolite-complex occurs in north-east Shetland in two nappes that structurally overlie Dalradian rocks (Flinn and Oglethorpe, 2005). Another ophiolite-complex occurs in the Grampian Terrane at Clew Bay in Ireland and ophiolitic fragments form a discontinuous outcrop along the Highland Border from Bute to Stonehaven (Tanner, 2007). Ophiolites of similar age occur in the Midland Valley Terrane at Ballantrae and Tyrone.

Several authors have attributed the main Grampian deformation to the obduction of ophiolites (e.g. Dewey and Shackleton, 1984; Dewey and Mange, 1999; Dewey, 2005) and it has been common practice to regard their age of emplacement as marking the start of the Grampian Event. Although relatively precise radiometric dates are now available for the crystallization ages of some of the obducted material, attempts to date the time of obduction have so far proved more difficult and, in some cases, contradictory.

In Shetland, an Ar-Ar step-heating age of 498 ± 2 Ma from hornblende in the metamorphic sole of the lower nappe (Flinn et al., 1991) is older than a U-Pb zircon crystallization age of 492 ± 3 Ma for a leucotonalite vein within the ophiolite (Spray and Dunning, 1991). Nevertheless, Flinn and Oglethorpe (2005) concluded that the lower nappe was obducted at about 500 Ma, i.e. early in the Grampian Event. However, U-Pb monazite ages from pelites in the footwall of the ophiolite nappes have been considered to date the regional metamorphism at 462–451 Ma (Cutts et al., 2011). Leucotonalites associated with the Ballantrae ophiolite have yielded a U-Pb zircon age of 483 ± 4 Ma with a marginally younger K-Ar hornblende age of 478 ± 8 Ma on an amphibolite that is considered to date obduction (Bluck et al., 1980). Two Cambrian dates obtained from an amphibolite on Bute that have been claimed to date obduction of the Highland Border ophiolite (Dempster and Bluck, 1991) have been questioned by Tanner (2007). The latter, expanding upon a model first proposed by Henderson and Robertson (1982), argued that the Highland Border Ophiolite was obducted at a relatively late stage in the Grampian Event onto the upper limb of an already recumbent Tay Nappe (F1) and hence could not have been responsible for significant early deformation of the Dalradian. However, subsequent work has shown that the ophiolite is relatively early in age and was emplaced prior to D1 (Henderson et al., 2009), at around 490 Ma (Chew et al., 2010), and prior to the metamorphic peak at c. 470 Ma.

It seems reasonable to conclude that fragments of ophiolite were probably obducted at various times in different places during the Grampian Event. Possibly more-reliable indicators of the onset of arc–continent collision in the west of Ireland are U-Pb zircon dates of 489 ± 3.1 Ma and 487 ± 2.3 Ma obtained from leucotonalite boulders unequivocally derived from the Cambro-Ordovician arc (Chew et al., 2007). Nd isotope evidence indicates that these leucotonalites had assimilated significant amounts of Laurentian margin sediment and hence that subduction was underway by c. 490 Ma. The timing might have varied slightly along the Laurentian margin, particularly if the convergence was markedly oblique, and some authors have suggested that events occurred earlier in Ireland than Scotland.

In the Scottish sector, intra-oceanic obduction and accretion was probably under way by 480 Ma (Tremadoc–Arenig) time, with the Midland Valley Arc beginning to encroach upon the Laurentian margin (Bluck, 2001) (Figure 1.16)a. The youngest rocks affected by Grampian deformation in the Highland Border region are at most c. 520 Ma (latest Early Cambrian; the Leny Limestone) and possibly as young as c. 475 Ma (mid Arenig; Tanner and Sutherland, 2007). But, by Arenig–Llanvirn time (470 Ma), several lines of evidence indicate that orogenesis was well under way and that the arc had been accreted onto the continental margin (Figure 1.16)b.

The Grampian deformation culminated in the formation of major fold stacks or nappe complexes and associated zones of structural attenuation (Figure 1.16)c. Structures associated with at least the major phases D1, D2 and D3 were formed at this time. The gross lateral continuity of the Dalradian lithostratigraphy precludes the existence of any large-scale thrusting at the present exposure level but a major zone of top-to-the-SE simple shear in the lower levels of the SE-facing and southerly directed Tay Nappe might reflect underthrusting of the arc beneath the Laurentian margin (Krabbendam et al., 1997).

Several suites of plutonic rocks were intruded into the Dalradian rocks of the North-east Grampian Highlands at c. 470 Ma, demonstrably during regional deformation and metamorphism. Precisely dated by U-Pb methods on minerals, these include the North-east Grampian Basic suite and two suites of diorite and granite (Kneller and Aftalion, 1987; Dempster et al., 2002). Similar basic and silicic plutons were emplaced at c. 470 Ma during the peak of metamorphism in the west of Ireland (Friedrich et al., 1999). These dates are consistent with U-Pb monazite ages of c. 470–450 Ma obtained from Grampian Group rocks of the Northern Grampian Highlands that are thought to date the peak of metamorphism (Phillips et al., 1999).

General consensus is that the main collisional event was relatively short-lived, and its culmination is constrained in both Western Ireland and Scotland to an interval of c. 10 Ma between 475 and 462 Ma (Dewey and Mange, 1999; Friedrich et al., 1999; Soper et al., 1999; Oliver, 2001). It was followed by uplift and cooling as the metamorphic belt was unroofed, starting at c. 460 Ma and continuing through to early Silurian time. A range of mineral ages, determined by various methods on muscovite and biotite, suggest relatively rapid cooling through 500°C and 350°C between 470 and 440 Ma and slower cooling through 300°C between 460 and 430 Ma, depending on location within the orogen (Dempster et al., 1995; Soper et al., 1999; Oliver et al., 2000).

Uplift and erosion dispersed Grampian-age metamorphic detritus across the accreted arc (or arcs) and into an accretionary wedge facing the narrowing Iapetus Ocean. In Scotland, such detritus appears in the sedimentary record of the Southern Uplands in the Caradoc Stage (Oliver, 2001). Later strike-slip displacements, which moved structural blocks and possibly whole terranes for considerable distances along the orogenic belt (see below), mean that detailed sediment-dispersal pathways cannot be determined. However, more complete data in the west of Ireland record progressive unroofing of an ophiolite complex during Arenig time, followed by exhumation of the metamorphic belt during the Llanvirn Age (Dewey and Mange, 1999).

1.8.5.2 The Scandian Event

The second main phase of the Caledonian Orogeny, termed the Scandian Event, occurred during early Silurian to Early Devonian time, between c. 435 and 400 Ma. It coincided with the final closure of the Iapetus Ocean, and the docking of Avalonia against Laurentia, but resulted mainly from the collision of Baltica with Laurentia (Coward, 1990; Dewey and Mange, 1999; Dallmeyer et al., 2001; Kinny et al., 2003a) (Figure 1.14) e and f. The effects of this event are widespread in the Northern Highlands Terrane and both west and east of the Walls Boundary Fault in Shetland, but are very limited in the Grampian Terrane, which must have been located some distance away from, and to the south of, the main collision zone (Figure 1.17)a. An alternative view is that the Grampian Highlands might have formed a relatively rigid block entrained between contractional deformation zones at the leading edges of the obliquely colliding Laurentian and Baltican plates (Leslie et al., 2008).

There is no specific evidence of regional deformation and metamorphism in the Grampian Terrane during the Scandian Event. Almost all of the major regional folds, including the Tay Nappe, had formed by now and the metamorphic peak was well past. However, the uplift and cooling at the end of the Grampian Event might have overlapped in time with Scandian deformation elsewhere and it is not clear how the major regional D4 deformation and subsequent localized phases are related to the overall tectonic pattern.

In contrast, Scandian nappe stacking dominates the structure of the Northern Highlands Terrane of Scotland and the East Greenland Caledonides. The most prominent structure ascribed to the Scandian Event in Britain is the Moine Thrust Belt, which forms the western Caledonian Front in Scotland. Thrusts within the belt are cut by syenites of the Loch Borralan Pluton, which have been dated at 430 ± 4 Ma (van Breemen et al., 1979). However, dating of micas in the Moine Supergroup and mylonites that overlie the thrust belt has indicated that ductile deformation may have been occurring until c. 410 Ma (Dallmeyer et al., 2001). Ductile deformation and thrusting also occurred farther to the east within the Moine rocks, and hence, in the Northern Highlands Terrane, the Scandian Event resulted in significant overall crustal shortening, probably in excess of 150 km. West-verging shears to the west of the Walls Boundary Fault in Shetland, including the Virdibreck Shear-zone responsible for emplacement of possible Dalradian rocks of the Queyfirth Group, have long been equated with the Moine Thrust Belt and hence can be regarded as Scandian.

East of the Walls Boundary Fault, the Upper Nappe of the Shetland Ophiolite-complex was emplaced by westward thrusting over the Lower Nappe, from which it is separated by an imbricate zone. Within the imbricate zone are siliceous metasedimentary rocks of Ordovician to Silurian age, derived by erosion of the Lower Nappe and Dalradian ‘basement’. The thrusting was broadly coeval with emplacement of the Skaw Granite, which has been dated by Ar-Ar step heating of muscovite in its aureole at c. 425 Ma (Flinn and Oglethorpe, 2005). Hence the emplacement of the Skaw Granite, the westward thrusting of the Upper Nappe and the metamorphism of the Ordovician–Silurian sedimentary rocks can all be regarded as part of the Scandian Event. In a further development, Cutts et al. (2011) have attributed a marked contrast in P–T conditions between pelites directly below the Lower Nappe and those from deeper structural levels to an extensional tectonic break that excised at least 10 km of crustal section at some time after the peak of metamnorphism.

Flinn and Oglethorpe (2005) also attributed the west-verging Quarff Nappe to the Scandian Event. The presence of Scandian structures in eastern Shetland further emphasises the separation between the Scottish and Shetland Dalradian sequences. Prior to later sinistral movement along terrane boundaries (see below), Shetland was probably sited more-directly opposite the Norwegian sector of Baltica, where a Scandian fold-and-thrust belt with an opposite, easterly vergence marks the eastern Caledonian Front.

The final stages of closure of the Iapetus Ocean was achieved by subduction to the north-west beneath the Laurentian margin (e.g. Dewey and Ryan, 1990), reversing the polarity of oceanic subduction, which had been towards the south-east beneath the Midland Valley Arc (Figure 1.16). It is generally accepted that the ocean had closed and the three continents had collided by the end of the Wenlock Epoch (e.g. Soper et al., 1992). And it is at about that time, at c. 425 Ma, that large volumes of silicic magma were intruded into both the Grampian and Northern Highlands terranes, followed shortly afterwards by High-K calc-alkaline volcanism in the Grampian and Midland Valley terranes (424–410 Ma). The magmas were derived mainly from lower crustal sources, but there is also a significant background input of magma from the subcontinental lithospheric mantle, modified by contact with subducted oceanic crust, even after the continental collision (see the Caledonian Igneous rocks of Great Britain GCR volume; Stephenson et al., 1999). The emplacement of the magmas implies that by this time the Grampian Terrane at least was being uplifted in an extensional regime and hence could not have been experiencing Scandian orogenic (i.e. compressional) forces.

1.8.5.3 Late Brittle Faulting, Terrane Assembly and Extensional Collapse

The main phase of Scandian ductile thrusting and folding was followed by sinistral strike-slip displacements along an array of NE-trending steep faults that dissect the Northern Highlands and Grampian terranes (Watson, 1984) (Figure 1.17). Strike-slip faults had already developed prior to and during the collision of Avalonia and Laurentia. However, the major sinistral movement was associated with the lateral translation of Baltica and Avalonia along the Laurentian margin that was probably continuous throughout late Silurian and into Early Devonian time. Those movements eventually resulted in juxtaposition of the Northern Highlands and Grampian terranes in more-or-less their current positions. According to some authors (e.g. Bluck, 2002), the Highland Border Ophiolite-complex was part of an exotic terrane that ‘docked’ with the Grampian Terrane during this strike-slip motion, and the Midland Valley Terrane reached its present position only in latest Devonian time (Bluck, 1984).

The most prominant faults are those that form the terrane boundaries; the linked Great Glen and Walls Boundary faults and the Highland Boundary Fault, along which hundreds of kilometres of displacement might have occurred. Seismic reflection studies show that the Great Glen Fault is coincident with a subvertical structure that extends to at least 40 km depth (Hall et al., 1984), and isotopic signatures of mantle-derived magmas are different either side of the fault, suggesting that it has some expression in the upper mantle (Canning et al., 1996, 1998). Geophysical studies have shown that the Highland Boundary Fault is broadly coincident with a change in lower crustal structure (Rollin, 1994) and various workers have speculated that this may correspond to the edge of the Laurentian craton (e.g. Soper and Hutton, 1984).

The timing and amount of movement along these faults are important for reconstructing some of the tectonic events in the later part of the Caledonian Orogeny (Harris, 1995). Although little is known about its pre-Silurian history, sinistral movement on the Great Glen Fault certainly occurred during the period between c. 430 Ma and c. 400 Ma (Stewart et al., 1999). More specifically, Silurian displacement is indicated by the U-Pb zircon age of 428 ± 2 Ma of the Clunes Tonalite which is thought to have been emplaced during sinistral shear (Stewart et al., 2001). The fault was re-activated around 400–393 Ma, as shown by evidence of sinistral transpression affecting the c. 400 Ma Rosemarkie Inlier leucogranite intrusions (Mendum and Noble, 2010). Late Emsian sedimentary rocks within the fault-zone are relatively undeformed and it therefore seems certain that intense deformation fabrics seen in Moine and Dalaradian rocks within the 3 km-wide fault-zone pre-date most Old Red Sandstone deposition (Stewart et al., 1999). However, during Early and Mid Devonian time, alluvial fans were deposited along active fault scarps (Trewin and Thirlwall, 2002). Thus, lateral movement along the Great Glen and associated faults overlapped with, but outlasted, the Scandian Event. Post-Caledonian structures along the fault-zone (most notably associated with Mesozoic dextral movements) are invariably brittle in style (Stewart et al., 1999).

The magnitude of early displacement along the Great Glen Fault is uncertain because there is no unambiguous correlation of pre-Devonian features across the fault. Estimates of sinistral movement based on correlations of various igneous and metamorphic features have ranged from 104 km (Kennedy, 1946) to 160 km (Winchester, 1974; Piasecki et al., 1981). Latterly, the general consensus has been that sinistral displacements are unlikely to have exceeded 200–300 km. This is consistent with most palaeomagnetic estimates (Briden et al., 1984) and the inferred offset of reflectors within the mantle lithosphere (Snyder and Flack, 1990). A comparable early sinistral offset of at least 100 km on the Walls Boundary Fault is based on the onshore geology of Shetland (Flinn, 1961) but apparent offsets of various offshore features have produced a variety of ambiguous and controversial estimates (e.g. Ritchie and Hitchen, 1993; Underhill, 1993). However, larger displacements are implied by tectonic reconstructions that place the Northern Highlands Terrane opposite Baltica during the Scandian collision, and Dewey and Strachan (2003) have suggested that at least 700 km of sinistral displacement must have occurred in order to explain the absence of Scandian deformation from the Grampian Terrane.

The Highland Boundary Fault is a high-angle reverse fault that eventually juxtaposed Dalradian rocks against Upper Devonian and Lower Carboniferous rocks of the Midland Valley Terrane. Much of the movement was therefore post Caledonian, although it seems likely that the reverse fault might have re-activated an older and more fundamental structure (Tanner, 2008). The full tectonic significance of the fault is uncertain. Late Silurian to Early Devonian sinistral displacements have commonly been assumed (e.g. Soper and Hutton, 1984; Hutton, 1987; Soper et al., 1992) and studies of clasts within Old Red Sandstone sedimentary rocks led Bluck (1984) to conclude that major strike-slip movement continued until latest Devonian time. However, there is little direct evidence and other workers have argued against any major lateral displacements after the Grampian Event (e.g. Hutchison and Oliver, 1998; Oliver, 2001).

Within the Grampian Terrane, several related NE-trending faults occur mainly between the Great Glen and Loch Tay faults (see figures 1.1 and 3.1). Marked facies and thickness changes across these faults, particularly in the Central Grampian Highlands, suggest that they exerted an influence on Appin and early Argyll group sedimentation and defined basin margins in places. Treagus (1991) demonstrated an early phase of dip-slip movement with a cumulative downthrow to the east of c. 7 km, followed by sinistral strike-slip with a cumulative displacement of c. 23 km. Various lines of evidence suggest a late Silurian to Early Devonian age for the main strike-slip movements, which had a close temporal relationship with, and partly controlled the emplacement of, certain late-Caledonian plutons and dyke-swarms (Morris and Hutton, 1993; Jacques and Reavy, 1994). The intrusion of large volumes of silicic magma, starting at c. 425 Ma, suggests that by mid-Silurian time sinistral transtensional stresses had begun to replace oblique convergence and transpression.

Most of the strike-slip faults developed prior to the onset of Old Red Sandstone sedimentation and by Early Devonian time the Scottish sector of the Caledonian orogenic belt was undergoing extensional collapse (McClay et al., 1986). Extensional dip-slip movements on many of the faults then led to the development of localized intermontane basins during the Early Devonian and more-extensive regional basins in Mid to Late Devonian time. Syndepositional deformation of Old Red Sandstone sedimentary rocks in the Orcadian Basin has been proposed by Seranne (1992), and shear fabrics in the Upper Nappe of the Shetland Ophiolite-complex and its underlying imbricate zone might also be related to this extensional phase and hence represent the youngest deformation to affect Caledonian metamorphic rocks (Cannat, 1989).

References