Content and licensingview original scan buy a printed copy

Moffatdale and the upper Ettrick valley: description of the solid geology of parts of 1:25 000 sheets NT 10, 11, 20 and 21

By B C Webb A W A Rushton D E White. Contributor: R J Merriman

Bibliographic reference: Webb, B C, Rushton, A W A, and White, D E. 1993. Classical areas of British geology: Moffatdale and the upper Ettrick valley: a description of the solid geology of parts of 1:25 000 sheet NT 10, 11, 20 and 21. (London: HMSO for British Geological Survey.)

Classical areas of British geology

Moffatdale and the upper Ettrick valley Classical areas of British geology: description of the solid geology of parts of 1:25 000 sheets NT 10, 11, 20 and 21

Authors

• B C Webb, BSc, PhD Formerly British Geological Survey
• A W A Harrison, BA, PhD British Geological Survey, Keyworth
• D E White, MSc, PhD Formerly British Geological Survey

British Geological Survey London HMSO 1993 © Crown copyright 1993. First published 1993.

(Front cover)

(Rear cover)

Notes

In this book the word 'district' means the area covered by the accompanying 1:25 000 map. National Grid references are within square NT and are given in square brackets [1. Fossil identifications are of specimens housed in the Survey collections at Keyworth.

Preface

The elongate valley of Moffatdale with its steep sides, hanging valleys and spectacular waterfalls is a justifiably popular area for tourists and hillwalkers in the Scottish Borders. The main fault-controlled valley, with its prominent NE–SW Caledonoid trend so characteristic of the geological structure of northern Britain, has been deeply gouged by glacial action to form a dramatic topographic feature which is visible even from space.

In the latter part of the 19th century, the area around Moffat played a key role in the understanding of the hitherto intractable geology of the Southern Uplands culminating in the outstanding work by Lapworth who demonstrated the value of graptolites in stratigraphical correlation within geologically complex terranes. The fundamental and continuing value of this far-sighted work is demonstrated every year by the numerous geological parties who visit the classic Moffatdale locality of Dob's Linn which contains the world stratotype section for the base of the Silurian System.

The first systematic survey of Moffatdale by the Geological Survey of Scotland was carried out by H Skae and J Young and the map published in 1889. The subsequent revision by B N Peach and J Home was incorporated in their Southern Uplands Memoir of 1899 with their revised map published in 1924. The present booklet is based on a partial revision in 1981–82 by B C Webb and J D Floyd (Gameshope catchment) with contributions on the palaeontology by A W A Rushton and D E White and on the metabentonites by R J Merriman.

Peter J Cook, DSc Director, British Geological Survey, Keyworth, Nottingham. NGJ2 5GG. October 1992

Chapter 1 Introduction

The Moffatdale district here described, forms part of the Central Belt of the Southern Uplands in which greywacke sandstones (the Gala and Hawick groups) overlie black pelagic shales (Moffat Shale Group) of upper Ordovician and Llandovery age (Figure 1). Remapping in 1981–82 was undertaken primarily to clarify the structure and stratigraphy of the greywackes. The stratigraphy of the Moffat Shale Group is based on the work of the earlier researchers cited, together with detailed work during this survey at a few selected localities.

High, rounded hills covered by rough grassland and peaty, heather moorland dominate the district, in which sheep farming is the main occupation, though substantial areas have been turned over to forestry. The highest ground lies north-west of Moffatdale where White Coombe [NT 162 151] rises to just over 800 m above sea level. Here, glaciation has produced dramatic scenery which includes Grey Mare's Tail Waterfall [NT 183 148] where the Tail Burn, after leaving the moraine-dammed Loch Skeen [NT 174 160], plunges 200 m from its hanging valley into Moffatdale. South-eastwards the relief is more subdued but still lies above 600 m along the main watersheds.

There is a pronounced north-east–south-west grain to the country, subparallel to the strike of bedding and the major structures. The main valleys of Moffatdale and the Ettrick Water show this orientation. Watersheds tend to lie closer to the south-eastern sides of the main valleys so that tributaries draining south-eastwards are longer and commonly expose good sections across strike. Ridges between tributaries, particularly those of the Ettrick Water, are cut by numerous slacks, the larger of which mark major fault lines and outcrops of the Moffat Shale Group. South-east of the Ettrick Valley the hills become more rounded and featureless as the incidence of major faults and shale outcrops decreases. Over most of the area exposure is limited to the stream sections.

Moffatdale is justly a 'Classical area of British geology', for it was in this region that Lapworth (1878) conclusively demonstrated the stratigraphical value of graptolites, and thereby made the biggest single contribution to the understanding of the stratigraphy and structure of the Southern Uplands of Scotland.

Various early workers such as Nicol (1848), Moore (1849) and Harkness (1851) had established that (1) the prevailing strike in the Southern Uplands is north-east to south-west, (2) most of the rocks are unfossiliferous greywackes dipping steeply to the north-west, (3) there are numerous outcrops of dark shale that generally show complicated structure and contain many species of graptolites, (4) the rocks on the north-west margin of the Southern Uplands include fossiliferous limestone of Llandeilo age, and (5) the rocks to the south (for example, near Kirkcudbright) are of Wenlock age.

Accordingly, the theory of the Southern Uplands, evolved at that time and adopted by the Geological Survey of Scotland, postulated an anticlinal structure whose axis extended north-east to south-west roughly on a line from Hawick to Lochmaben. To the south-east of the anticline an unconformity was invoked to explain the presence of Wenlock rocks and the absence of Llandeilo strata, though no unconformity had actually been observed. The Llandeilo succession to the north-west of the anticline was thought to occupy a tract about 40 km wide and, notwithstanding the possibility that the strata (including the black shales) might be repeated by faults or folds, to be of immense thickness. The presence of rich graptolite faunas throughout the Llandeilo succession was judged to be typical of that series, and the point that some of the species had been recorded from the Caradoc or Llandovery of Wales was accounted for by means of Barrande's now discredited theory of 'colonies'.

Lapworth spent many years studying the black shales and their graptolites. He observed that the graptolites always occur in definite faunal associations, that each of these associations ranges through a limited thickness of black shale of distinctive lithology, and that where undisturbed the associations occur in a regular order. From these observations Lap-worth devised a lithostratigraphy and biostratigraphy for the black shale outcrops and showed them to represent a single lithostratigraphic group, 'the Moffat Shales', which always underlies the greywackes. By comparison with graptolitic records in Wales and abroad, he showed that the 'Moffat Shales', whose base is not seen, range in age from Llandeilo to Llandovery. Lapworth accounted for the disposition of the black shale outcrops by postulating an anticlinorium of concertina-style folds with Moffat Shale Group rocks'in the cores of isoclinal antiforms. Although his structural scheme has been superseded, his stratigraphical conclusions, and indeed his 'typical section' at Dob's Linn [NT 196 158], have remained undisputed standards of reference for more than a hundred years.

It is difficult now for us to grasp the magnitude of Lapworth's achievement; in complex ground, and without preexisting standards to draw upon, he devised a theory that not only removed all the difficulties of the orthodox integration of the structure of the Southern Uplands but also established the value, previously quite unsuspected, of graptolites in Lower Palaeozoic stratigraphy. Despite the success of his work, his conclusions were not immediately welcomed by the Geological Survey of Scotland (see discussion in Lapworth, 1878), just as the extension of Lapworth's methods to the rocks of Bohemia by Marr (1880) was unwelcome to Barrande, for it began the destruction of the theory of colonies.

By 1888, however, the Survey had begun to review their maps of the Southern Uplands in the light of Lapworth's theories, and in a remarkable 10-year survey of the whole Southern Uplands country, executed during the winter months (Wilson, 1977), Peach and Horne (1899) found them to be applicable far beyond the limits of Moffatdale.

Recent palaeontological work in the area has mainly been concerned with the detailed distribution of graptolites in the Moffat Shale Group sections in Dob's Linn (Toghill, 1968; Williams, 1982a,b, 1983). These sections serve as a standard not only for Scotland but for successions in the graptolitic facies in many parts of the world, and include the world stratotype for the base of the Silurian System (Cocks and Rickards, 1988), thus vindicating Lapworth's choice of Dob's Linn as a 'typical section'.

Lapworth's structural interpretation was accepted for eighty years until Craig and Walton (1959) showed it to be contradicted by the younging evidence of sedimentary structures in the turbidite sandstones of the Kirkcudbright area. They advanced a new theory in which the mesoscopic folds lay on a series of larger, north-facing monoclines of several kilometres wavelength, separated by major thrust faults. Rust (1965) working in the Whithorn area showed that the deformation was polyphase and this was confirmed by other workers (for example, Weir, 1968). Rust (1965) invoked three phases of Caledonian deformation. The first produced the major folds and strike-parallel thrust faults; the second refolded the first structures to produce the monoclines and the third produced steeply plunging folds and reactivated the early thrusts as sinistral wrench faults. Rust (1965) also identified later, minor structures which include a set of recumbent mesoscopic folds with gently inclined axial planes. The main cleavage was related by Rust to his third deformation since it was not axial-planar to the earlier folds.

The importance of the early thrusts in controlling the Moffat Shale Group outcrops was demonstrated by Toghill (1970c) in the upper Ettrick Valley, and an imbricate thrust structure rising from a sole thrust in the Moffat Shale Group rocks was proposed for the nearby Craigmichan Scaur area [NT 162 061] by Fyfe and Weir (1976). These observations led to a comparison of the Southern Uplands structure with that of modern accretionary wedges developed at destructive plate margins (McKerrow et al., 1977; Leggett et al., 1979). It was proposed that the Lower Palaeozoic rocks of the Southern Uplands represented sediments scraped off the floor of the Iapetus Ocean and accreted onto the Laurentian plate margin as the oceanic crust was subducted northwestwards beneath the continent. In this model deformation should be relatable to both the period of accretion and to the collision event when the ocean finally closed in end-Silurian to early Devonian time. Eales (1979) considered that all of Rust's (1963) Caledonian deformation phases could be related to accretion tectonics.

More recent structural work in the Southern Uplands has largely been concerned with testing the accretionary wedge hypothesis and seeking evidence for collision tectonics. Stringer and Treagus (1980, 1981) revised Rust's (1965) deformation sequence in Galloway. They dispute the existence of Rust's second folds, the major monoclines, and combine his first and third phases into a single phase. They argue that as a result of oblique subduction, variably plunging folds and non-axial planar cleavage may develop synchronously. They suggest that the late, mesoscopic folds identified by Rust were the product of the collision event. However, more wide ranging syntheses of Caledonian tectonics (Dewey and Shackleton, 1984; Soper and Hutton,1984) point to an important late phase of sinistral shear strain to which the main cleavage may be related. As research progresses it becomes increasingly clear that major shear movements in the Iapetus suture zone have juxtaposed terranes with different histories. Many researchers would now question the applicability of the accretionary wedge hypothesis to much, if not all of the Southern Uplands (Murphy and Hutton, 1986; Stone et al., 1987).

Chapter 2 Stratigraphy

Moffat Shale Group

The Moffat Shale Group comprises variegated grey and black mudstones ranging in age from mid-Ordovician (Llandeilo) to early Silurian (lowermost upper Llandovery) and represents a thin but laterally continuous sequence of hemipelagic deposits. Thin layers of bentonitic clay are commonly interbedded with the shales. Although the base of the Moffat Shales is not seen in the Moffatdale district, evidence from farther north indicates that the group was deposited over cherts and basalts which may represent oceanic crust. The uppermost beds pass upwards and also laterally into greywackes of the Gala Group. It is difficult to estimate the thickness of the Moffat Shale Group but at its fullest development it appears to be about 100 m thick. Because of the incompetence of the shale bands, they provide a level along which the overlying greywackes are apt to be faulted or thrust, and so the shales crop out only in tectonically deformed inliers.

Lapworth (1878) divided the Moffat Shale Group (or Moffat "Series") into three, the Glenkiln, Hartfell and Birkhill shales, each of which he further subdivided. Peach and Home adopted these divisions on their maps of the Moffat area and all the adjoining districts. In the present survey these units were not separately mapped because it proved impossible to represent them adequately at the scale used. Peach and Horne (1899) described both the major outcrops at Dob's Linn [NT 196 158] and Craigmichan Scaur [NT 162 061] and also many smaller outcrops in the district. References to recent, more detailed descriptions will be found in the text.

Glenkiln Shale

The Glenkiln Shale is the least well-known unit of the Moffat Shale Group. The base is not seen but the exposed thickness may reach about 20 m, consisting of grey mudstone and silt-stone. Some units are hard, flaggy and silicious and some beds are bioturbated. There are at least two horizons of cherty black graptolitic shale, the lower one referable to the Nemagraptus gracilis Zone and the upper to the 'Climacograptus peltifer' Zone.

The best exposures of the Glenkiln Shale lie outside the district but parts of the succession may be seen along the line of the Ettrick Valley, at Craigmichan Scaur, Entertrona Burn [NT 1850 0804] and Birnie Cleuch [NT 1925 0932]. The formation is exposed in Broomy Gutter [NT 1513 1365], north of Moffatdale, and there was formerly an exposure at the foot of the Main Cliff at Dob's Linn, but this is now covered by scree.

Hartfell Shale

The Hartfell Shale is about 50 m thick and falls into two divisions. The lower division is predominently of black mud-stone and is 22 m thick. At the base cherty grey mudstone is interbedded with soft black mudstone and is referred to the Climacograptus wilsoni Zone; above are 10 m of hard flaggy black mudstone referred to the Dicranograptus clingani Zone; and at the top are 5 m of softer black shale of the Pleurograptus linearis Zone with thin pale grey seams of metabentonite claystone. The upper Hartfell division is Lapworth's "Zone of barren mudstones" and consists mainly of grey mudstone with black graptolitic bands, together about 28 m thick; the lowest two black bands are, the only fossiliferous representatives of the Dicellograptus complanatus Zone. The lower of these is 40 mm thick and lies about 9 m above the base of the Upper Hartfell Shale, whilst the upper band, 0.4 m higher, is only 10 mm thick. The upper half of the division includes several black bands referable to the Dicellograptus anceps Zone and, near the top, are a single massive bed with trilobites and a black seam, 2–3 mm thick, with Climacograptus? extraordinarius.

The Hartfell Shale is well exposed at Dob's Linn and Craigmichan Scaur, and parts of the sequence are proved at several other localities such as Donald's Cleuch [NT 1505 1882], Rowantree Gutter [NT 1970 1245], Range Cleuch [NT 185 109] and Black Grain [NT 224 141].

Birkhill Shale

The Birkhill Shale is exposed at many localities and the lithological and faunal successions are well known. The lower part of the Birkhill Shale consists of black shale and hard massive black mudstone with thin metabentonite beds, and is nearly 20 m thick. Several graptolite zones are proved, from the Glyptograptus persculptus Zone (now relegated to the top of the Ordovician) to the lower part of the Coronograptus gregarius Zone (Llandovery). The upper part of the Birkhill Shale consists of grey mudstone and shale with black graptolitic mudstone and numerous beds of metabentonite, together more than 20 m thick. All the zones from the upper part of the C. gregarius Zone to the lower part of the Rastrites maximus Subzone are present.

The type section for the Birkhill Shale is at Dob's Linn where the whole sequence is exposed. A large part of the succession is exposed also at Craigmichan Scaur and smaller parts are seen at dozens of localities in the Moffatdale area, for example Swine Cleuchs [NT 171 135], Muchra Burn [NT 2266 1671], Fala Grain [NT 216 145] and Shorthope [NT 219 124] (Ettrick Water). The highest beds of the Birkhill Shale are seen in contact with the overlying Gala Group at Dob's Linn, Mid Craig (Loch Skeen) [NT 163 167], Nether Torr Gill [NT 139 122], Seavy Sike [NT 1775 1460], Black Grain and Pot Burn [NT 1805 0895]. This contact is older in the north-west, where the youngest Birkhill Shale is low in the Monograptus sedgwickii Zone (as at Mid Craig), than in the south-east where it lies well up in the R. maximus Subzone (as at Pot Burn).

Biostratigraphy

Graptolites are by far the commonest fossils in the Moffat Shale Group and are the only group that is used biostratigraphically; other groups such as brachiopods, conodonts and crustacea occur too rarely or sporadically to be of value. There is no doubt that graptolites were ocean-going animals, though to what extent they were automobile or passive plankton is debated (for example, Kirk, 1972). It is also uncertain at what depth they habitually lived: it is probable that different forms favoured different levels in the water-column (for example, Berry, 1962).

In the Moffat Shale Group graptolites are confined to the beds of black mudstone that are thought to have accumulated in anoxic conditions. Although they are distributed throughout the mudstones, they tend to be concentrated on certain bedding-planes, sometimes (according to Williams and Rickards, 1984) by a winnowing process rather than because of mass mortality.

Ever since the work of Lapworth (1878, 1879) the Moffat Shale Group has furnished a standard of reference for graptolite biostratigraphy in the upper part of the Ordovician and lower part of the Silurian, not only for southern Scotland but world-wide. Several of the graptolite zones used for international correlation were first recognised in the Moffat Shale Group, and it is a tribute to Lapworth's acumen that, despite many subsequent modifications to the zonal scheme in general use, the zones that he proposed remain the most easily distinguished.

The zones in the Moffat Shale Group are recognised mainly by the concurrence of particular species of graptolites: in other words by assemblages of species whose ranges overlap (see (Table 1), (Table 2), (Table 3)). Each zone is named after a particular species that may be especially characteristic of the zone but is not necessarily confined to it; a zone can be recognised in the absence of the nominal species, as for examples the convolutus Zone which is identified by its typical assemblage, but commonly in the absence of M. convolutus itself.

Biostratigraphical precision is attained by detailed fossil-collecting, and this in turn depends on the sections available. The biostratigraphical resolution in the Birkhill Shale and parts of the Hartfell Shale is accordingly good, thanks to the work of Toghill (1968, 1970a) and Williams (1982a, b, 1983), but is less good in the Glenkiln and lower part of the Hartfell Shale. Each zone is briefly discussed below.

Nemagraptus gracilis Zone

The N. gracilis Zone is well developed in the Glenkiln Shale but, because the graptolites are confined to a comparatively restricted thickness of black mudstone, the limits of the zone, which presumably lie within the sub- and superjacent barren beds, are conjectural. The zone includes a large fauna, characterised especially by N. gracilis itself, though other species of Nemagraptus are known both above (Strachan, 1960, p.111) and below (Toghill, 1970b, p.124) the limits of the zone. Didymograptus superstes is confined to the gracilis Zone and several other species are particularly common: Dicellograptus intortus, D. sextans, Climacograptus antiquus, Orthograptus whitfieldi, Hallograptus mucronatus and Thamnograptus typus. (Figure 2).

In the Moffatdale area the gracilis Zone is proved at Broomy Gutter on Carrifran Gans and in Entertrona Burn.

'Climacograptus peltifer'Zone

Generalised zonal schemes for the British Isles (for example, Williams et al., 1972) show a Diplograptus multidens Zone above the gracilis Zone and below the Dicranograptus clingani Zone. In the Moffat Shale Group this interval has been divided into two zones of only local value, the zones of Climacograptus peltifer and C. wilsoni. Riva (1976, p.595) showed that in large populations the nominal species C. peltifer is indistinguishable from the longer-ranging species C. bicornis, so that C. peltifer cannot be relied upon to identify the Zone. The 'peltifer'Zone fauna is much the same as that of the gracilis Zone, but Dicranograptus recites and D. ziczac are common and Dicellograptus patulosus is reported to be restricted to the 'peltifer' Zone; indeed, Elles and Wood (1904, p.148; 1906, p.196) had referred to this zone as the 'patulosus Zone'. The 'peltifer' Zone is known at Dob's Linn and Craigmichan Scaur.

Climacograptus wilsoni Zone

The C. wilsoni Zone contains a poorly diversified transition-fauna between the richer gracilis/'peltifer' faunas below and the clingani Zone fauna above. The zone is not identified beyond southern Scotland and further work is needed to assess its value. It is identifiable in the Moffat Shale Group because of its distinctive facies of soft shale, sometimes with pyritised graptolites, at the base of the Hartfell Shale. The zone is characterised by the presence of C. wilsoni (Figure 2) and the appearance of Orthograptus of the amplexicaulis ( = truncedus) species group, but, apart from Dicranograptus nicholsoni, species of Dicranograptus and Dicellograptus are rare.

In the Moffatdale area the wilsoni Zone is exposed only at Dob's Linn and Craigmichan Scaur. The type development is at Hartfell Spa, north-west of the present district.

Dicranograptus clingani Zone

The fauna of the D. clingani Zone contains many Dicellograptus (including D. caduceus, D. flexuosus, D. morrisi) and Dicranograptus (D. clingani, D. ramosus). Large Orthograptus of the calcaratus and amplexicaulis groups are common, with Climacograptus spiniferus, C. caudatus and species of Corynoides, Leptograptus and Neurograptus (Figure 3). Dicranograptus clingani, Dicellograptus caduceus and C. caudatus are most characteristic of the Zone.

The clingani Zone is well developed at Dob's Linn, and Williams (1982a) has shown in detail the distribution of species in the upper part of the zone there. It is also proved at Craigmichan Scaur.

Pleurograptus linearis Zone

Pleurograptus linearis is confined to levels within its own zone, and is accompanied by Dicellograptus (several species), Leptograptus, large Orthograptus (including the distinctive O. quadrimucronatus) and Climacograptus styloideus (Figure 3).

The type locality of the P. linearis Zone is at Dob's Linn, and Toghill (1970a) discussed the development in the Main Cliff there. Williams (1982a) showed that the faunal distribution in a section on the North Cliff is gradational with the fauna of the clingani Zone below. He took the base of the linearis Zone at an arbitrary level above where elements of the clingani Zone fauna die out.

Dicellograptus complanatus Zone

The boundary between the Pleurograptus linearis and Dicellograptus complanatus zones presumably lies at some level above the base of the pale 'barren' mudstone that overlies the P. linearis Zone. The complanatus Zone yields graptolites, chiefly D. complanatus (Figure 3) and Orthograptus socialis, but only in two thin black beds that lie 9 m above the base of the pale mudstone. The overlying 13 m of barren beds are arbitrarily assigned to the same zone (Toghill, 1970a, p.8; Williams, 1987)), and are overlain by the lowest black beds of the Dicellograptus anceps Zone.

In the Moffatdale district the complanatus Zone has been proved only at Dob's Linn, though the pale barren mudstones assigned to the zone are conspicuous in other sections, including those at Black Grain, Fala Grain, Rowan-tree Gutter and Craigmichan Scaur.

Dicellograptus anceps Zone

The D. anceps Zone is characterised by Dicellograptus (D. anceps, D. complexus, D. minor, D. ornatus), Climacograptus (C. supernus, C. latus, C. miserabilis) and swarms of Orthograptus abbreviatus (Figure 3). Dicranograptus is absent.

The development of the anceps Zone at Dob's Linn was discussed by Williams (1982b); he showed the graptolite distribution through the five black bands present and distinguished two subzones, that of Dicellograptus complexus below and Paraorthograptus pacificus above. The anceps Zone is proved also at Black Grain and Range Cleuch.

Climacograplus? extraordinarius Zone

At Dob's Linn the topmost black band of the Hartfell Shale, a mere 2–3 mm thick, yields C.? extraordinarius (Williams, 1982b). This is the zonal fossil for the extraordinarius Zone as developed in the uppermost Ordovician in Kolyma, USSR, and shows that the zone is present in the Moffat Shale Group, as also in Co. Cavan, Ireland (Siveter et al., 1980). In both Dob's Linn and Co. Cavan a small-eyed Dalmanitid trilobite occurs in beds adjacent to the extraordinarius Band.

Glyptograptus persculptus Zone

The G. persculptus Zone is the lowest zone identified in the Birkhill Shale and yields a restricted fauna of diplograptids such as G. cf. persculptus, Climacograptus normalis and C. miserabilis (Williams, 1983). The G. persculptus Zone was formerly regarded as the lowest zone of the Llandovery Series but the Ordovician–Silurian boundary is now defined by international agreement at the base of the overlying Parakidograptus acuminatus Zone (Cocks, 1985), with the stratotype at Dob's Linn.

In the Moffatdale district the persculptus Zone has been recognised only at Dob's Linn.

All the graptolite zones of the British Silurian were discussed thoroughly by Rickards (1976) and the notes below are a guide to their application in the Moffat Shale Group. The distribution of graptolites (see (Figure 4), (Figure 6)). All the figures are about twice natural size. 1 Pribylograptus leptotheca (Lapworth) 2 Monograptus argenteus (Nicholson) 3 Petalograptus ovatoelongalus (Kurck) 4 Rhaphidograptus toernquisti (Elles and Wood) 5 Glyptograptus tamariscus tamariscus (Nicholson) 6 Monograptus lobiferus (McCoy) 7 Rastrites longispinus Perner 8 Monograptus limatulus T�rnquist 9 Monograptus convolutus (Hisinger) 10 Monograptus clingani (Carruthers) 11 Cephalograptus cornea (Geinitz) 12 Monograptus decipiens Tornquist 13 Clinoclimacograptus retroversus Bulman and Rickards 14 Pristiograptus regularis (Tornquist) 15 Monograptus sedgwickii (Forelock)." data-name="images/P992267.jpg">(Figure 5), (Figure 6) in the sections at Dob's Linn, the only locality at which all the units of the Moffat Shale Group have been recognised, was described by Toghill (1968).

Parakidograptus acuminatus Zone

This zone is recognised by the appearance of P. acuminatus, Akidograptus ascensus, Climacograptus tr:fi lis and the first monograptid, namely Atavograptus ceryx. The zone was originally described at Dob's Linn and has been recognised at Black Grain and Fala Grain.

Atavograptus atavus Zone

This was originally defined at Dob's Linn as the vesiculosus Zone, characterised by the large and distinctive C. vesiculosus (Figure 4), accompanied by Dimorphograptus, Climacograptus spp. and slender monograptids. This zone is easily recognised in the Birkhill Shale because it is typically developed in hard flaggy shales that tend to form prominent exposures, and several of the records (especially in Peach and Home, 1899) are based on it. The more restricted view taken by Toghill (1968) makes the vesiculosus Zone equivalent to the Atavograptus atavus Zone (Rickards, 1976, p.160); besides A. atavus, the presence of Rhaphidograptus extenuatus and Coronograptus cyphus praematurus characterises the zone.

In the wider sense, the vesiculosus Zone has been recognised at Craigmichan Scaur and along the Ettrick Water, at Fala Grain and in Cossarhill Burn [NT 227 156], as well as at Dob's Linn.

Lagarograptus acinaces Zone

The L. acinaces Zone is equivalent to part of Lapworth's vesiculosus Zone and the base of the Coronograptus cyphus Zone as recognised by Toghill. It is characterised by Lagarograptus acinaces, Pribylograptus sandersoni, Atavograptus strachani and Dimorphograptus spp. The acinaces Zone can be difficult to recognise but has been proved at Black Grain and Dob's Linn.

Coronograptus cyphus Zone

Besides the presence of C. cyphus, this zone is distinguished by the appearance of Monograptus with hooked thecae (revolutus group). Several Climacograptus spp. are present in this zone but are rare thereafter. The cyphus Zone is recognised at Birnie Cleuch, Fala Grain, Black Grain, Dob's Linn and west of Bran Law [NT 1797 1558].

Coronograptus gregarius Zone

In the Birkhill Shale the gregarius Zone is easily recognised by the abundance of the zone fossil, together with the appearance of Monograptus with triangulate thecae and Petalograptus. The gregarius Zone has been recognised at several localities in the present area.

The gregarius Zone is divided into three subzones that have been identified at Dob's Linn but have not in general been distinguished in other sections, probably because insufficient detailed collecting has been undertaken. These are, in upward succession:

1.      The Monograptus triangulatus Subzone which is readily recognised by the appearance of the subzonal fossil with Petalograptus.

2.      The Diplograptus magnus Subzone which is recognised at a horizon with D. magnus and Monograptus fimbriatus.

3.      The Pribylograptus leptotheca Subzone which is recognised at a level where the subzonal fossil and Monograptus argenteus appear.

Monograptus convolutus Zone

The convolutus Zone yields a rich fauna characterised also by M. clingani, M. decipiens, M. limatulus, M. lobiferus and species of Cephalograptus (Figure 6)). All the figures are about twice natural size. 1 Pribylograptus leptotheca (Lapworth) 2 Monograptus argenteus (Nicholson) 3 Petalograptus ovatoelongalus (Kurck) 4 Rhaphidograptus toernquisti (Elles and Wood) 5 Glyptograptus tamariscus tamariscus (Nicholson) 6 Monograptus lobiferus (McCoy) 7 Rastrites longispinus Perner 8 Monograptus limatulus T�rnquist 9 Monograptus convolutus (Hisinger) 10 Monograptus clingani (Carruthers) 11 Cephalograptus cornea (Geinitz) 12 Monograptus decipiens Tornquist 13 Clinoclimacograptus retroversus Bulman and Rickards 14 Pristiograptus regularis (Tornquist) 15 Monograptus sedgwickii (Forelock)." data-name="images/P992267.jpg">(Figure 5). The zone is well developed at Dob's Linn and has been identified at Mid Craig, Swine Cleuchs, Muchra Burn, Trow Grain [NT 2155 1435], Craigmichan Scaur and along the Ettrick Water, for example at [NT 219 124].

Monograptus sedgwickii Zone

The M. sedgwickii Zone is characterised by the zone fossil accompanied by M. decipiens, Pristiograptus regularis, P. jaculum, Rastrites fugax and Petalograptus tenuis. The middle part of the zone is commonly rich in well preserved fossils. The type locality is at Dob's Linn and the zone has been recognised at many other localities in the district.

Monograptus turriculatus Zone (Rastrites maximus Subzone)

Only the base of the M. turriculatus Zone is developed in the Birkhill Shale, and that part is distinguished as the R. maximus Subzone by the occurrence of the subzonal fossil, accompanied by R. linnaei, R. distans, Monograptus halli'and Pristiograptus nudus (Figure 6.).

The subzone is best developed in the south-east of the district, as at Pot Burn (Toghill, 1970c), but is thinner or absent to the north-west of Dob's Linn.

Metabentonite

At Dob's Linn at least 11 per cent of the Birkhill Shale is formed of altered volcanic ash which now consists of soft white or pale green-grey mudstones, recorded by Lapworth (1878) as 'white lines' or 'white clay bands'. The petrography, mineralogy and geochemistry of these mudstones is characteristic of metabentonites (Merriman and Roberts, 1990). In excess of 130 metabentonites occur between the base of the Dicellograptus complanatus Zone, in the Harden Shale, and the lowest Gala Group greywackes at Dob's Linn, in beds ranging from a few millimetres to 0.5 m in thickness. Trace-element geochemistry suggests that the metabentonites were derived from glassy silicic ash representing trachyandesitic, dacite and rhyolitic magmas. The greatest volume of ash accumulated in the Monograptus sedgwickii Zone, where it forms at least 20 per cent of the succession, with a mean bed thickness of 8.6 cm and a frequency of 2.4 metabentonite beds per metre of shale.

Gala and Hawick groups

The Gala and Hawick groups comprise sandstones and siltstones with sedimentary structures indicative of their deposition by turbidity currents and other mass flow processes. The facies distribution in the Gala Group shows that it represents the deposits of a submarine fan complex which spread both south-eastwards and south-westwards across most of the district. The Hawick Group crops out only in the south-east of the district and shows little facies variation.

Gala Group (Queensberry Formation)

The base of the Gala Group is intermittently exposed between Gameshope Burn and the Ettrick Valley. Because the major thrusts generally pass downwards into the underlying Moffat Shale Group the contact between the shales and the Gala Group is commonly undisturbed and conformable. Decrease in thrust separation indicates that the group becomes thinner south-eastwards. It is some 900 m thick north-west of Moffatdale but less than 100 m thick near the Ettrick Water.

North-west of the Moffatdale district the Gala Group sandstones are subdivisible on petrological grounds (Walton, 1955; Floyd, 1982) but here they comprise only quartzose greywacke and are referred to a single formation. This, the Queensberry Formation, is subdivisible on sedimentological criteria but the complex boundaries between different facies cannot, in general, be represented on the map. A distinctive siltstone in the north-west of the district has been mapped and is traceable north-eastwards for a further 20 km, to beyond the Megget Water. South-westwards, however, there is transition towards the more common sandstone facies and the boundaries become increasingly difficult to represent. For this reason it is not, at present, considered wise to refer the sandstones above and below the siltstone to separate formations.

The Queensberry Formation predominantly comprises grey and greenish grey, quartzose greywackes in beds ranging from a few centimetres to several tens of metres thick, interbedded with grey and greenish grey, rarely purple, laminated, muddy siltstones. The siltstones normally constitute less than 50 per cent of the local succession but dark grey, laminated siltstone in units up to 250 m thick occurs, as noted, in the north-west of the district between Loch Skeen and Blackhope. At least one thin (80 mm) metabentonite horizon occurs locally within the basal few metres of the formation and is sporadically exposed in the Ettrick Water's tributaries (for example, [NT 2234 1385]). Interbedded carbonaceous shale containing graptolites is restricted to the extreme south-east of the outcrop.

The greywacke sandstones are dominantly of fine- to medium-sand grade but thicker units are commonly of coarse sand to granule grade. Mudstone rip-up clasts are sporadically present near the bases of beds but true conglomerates are rare. In Blackhope [NT 1317 1229] a very dark quartzofeldspathic greywacke with common pyrite crops out. It is up to 7 m thick and carries mudstone clasts and small lenses of conglomerate with variably rounded quartz and lithic clasts up to 100 mm across. Such distinctive lithologies are rare and subdivision of the formation is based on sedimentary features. Since the Bouma divisions typical of turbidites (Figure 7) are seldom clearly visible, subdivision was largely based on bed thickness and grain size (Figure 8). Six subdivisions were recognised and related to four turbidite facies, A, B, C and D described by Walker and Mutti (1973). These facies, or associations of them, can be related to a generalised, submarine fan model (Figure 9). They are annotated on the map and described below, starting with the thin-bedded, more distal facies.

Facies C

This facies comprises regularly bedded, classical turbidite sandstone and siltstone. The siltstone, commonly laminated, in units 50–150 mm thick makes up 25–50 per cent of the succession. The sandstone beds are in the range 50–500 mm in thickness and are laterally continuous. They are of fine- to medium-sand grade and commonly occur in coarsening-upwards cycles from a few metres to a few tens of metres thick. Towards the tops of such cycles thicker, amalgamated beds, more properly assigned to Facies B, may be present. The graded and parallel-laminated Bouma divisions are normally present although fine grain size may obscure grading. Ripple cross-lamination and convolute laminations are sporadically visible as are bottom structures. This is a middle fan facies typical of the smooth, distal portion of a suprafan lobe. It crops out in the lower and middle reaches of the Ettrick tributaries and from there across to Craigmichan Scaur.

Facies B/C

This facies is dominated by thicker and amalgamated turbidite sandstones (Facies B) although Facies C beds are still present. Interbedded turbidite siltstone ranges from a few millimetres to over a metre in thickness but commonly constitutes less than 25 per cent of the succession. Facies B sandstone ranges from medium-sand to granule grade, in beds commonly a metre or more thick and amalgamated. Lateral continuity is good although less so than in Facies C. Internal sedimentary structures are poorly visible and generally limited to graded bedding and faint parallel laminations. Internally structureless beds are probably the product of grain flow or fluidised sediments. Bottom structures including flute and groove casts are commonly well preserved and upper surfaces may be rippled. Facies B sandstones are the deposits of braided channels and the mixed B/C facies represents the upper, channelled portion of a suprafan lobe. It crops out from the upper reaches of the Ettrick tributaries across into the lower reaches of the Moffatdale tributaries.

Facies A/B

This facies is dominated by massive, commonly amalgamated, coarse-grained, locally conglomeratic sandstones. Thinner sandstone beds relatable to Facies C are rare and only very locally developed. Interbedded siltstone is local and makes up only a small percentage of the succession.

Facies A/B

Beds range from one to several tens of metres in thickness and are commonly internally structureless apart from local, poorly developed grading. They represent channel facies sandstones developed in the upper middle fan to inner fan region and crop out in the upper reaches of the Moffatdale tributaries and across the watershed into Gameshope Burn. South-eastwards there is transition into Facies B/C but where Facies A/B is well developed it is intimately related to Facies D, the interchannel siltstones (see below). On Mid Craig, south-west of Loch Skeen, two Facies A/B sandstone units, 50 m and 25 m thick respectively, alternate with rather thicker Facies D siltstones. Traced south-westwards the sandstone units thicken to 200 m and 150 m respectively and interbedded siltstone becomes more common. North-west of Loch Skeen only one Facies A/B unit, some 600 m thick, appears to be present.

Facies D

This facies is dominated by laminated siltstone but thin, fine-grained turbidite sandstone occurs sporadically. Intimately related to the channel sandstones (Facies A/B) Facies D represents overbank deposits laid down alongside active channels. The major outcrop extends north-east and south-west from Loch Skeen with minor outcrops in the Gameshope catchment.

Current directions inferred from bottom-structures such as flute and groove casts invariably indicate a northeast–south-west current orientation. The rare unidirectional structures indicate that currents flowed from the north-east (for example, Trow Grain [NT 2114 1447] and Raking Gill [NT 1998 1503]).

Hawick Group

The Hawick Group comprises turbiditic sandstones and siltstones very similar to the Queensberry Formation of the Gala Group. The boundary between these groups is drawn along a thrust which is not exposed within the map area but is seen in Master Grain [NT 2307 1190] just off the eastern edge of the map. Black shale in the hanging wall of the thrust yielded graptolites probably of the R. maximus Subzone. There is an exposure, possibly of the M. crispus Zone also in Master Grain [NT 2323 1166] some 300 m south-east of the R. maximus outcrop, and of the M. griestoniensis Zone in a forestry road cutting above Glendearg Burn [NT 2214 0971]. Since the beds young north-westwards a thrust must separate these localities. It is unexposed but probably gives rise to the feature west of Glendearg [NT 215 092] and that south-east of Black Knowe [NT 228 107]. This thrust defines the base of the first northernmost thrust slice of the Hawick Group some 100 m below the M. crispus occurrence with the top some 300 m above it. The base is probably at roughly the same horizon as the tops of the Gala Group slices to the north-west and is probably at a horizon of siltstone and silty mudstone of high M. turriculatus Zone age. Farther south-east the base of the Hawick Group is not seen and may lie at a lower stratigraphic level.

In the field, Hawick Group lithologies are subdivided in the same way as those of the Gala Group and the facies associations are referred to the same submarine fan model. Middle fan facies (B/C) predominates, with sandstone and siltstone beds commonly a metre or more thick. Sandstones are somewhat calcareous and generally appear more flaggy and rusty weathering than those of the Gala Group. Sedimentary structures are similar to those of the Gala Group and include sand volcanoes, typical of fluidised beds (for example, Glendearg [NT 2282 0628]). Siltstones are pale green or red in colour. The red colour is considered to be primary and its presence has previously been used to define the Hawick Group (Peach and Horne, 1899). Although it is difficult to draw a precise boundary using this criterion alone, the structural interpretation concurs. Black, carbonaceous shales a few millimetres thick occur sporadically within the turbidite siltstones and have yielded graptolites indicative of the M. crispus and Mci. griestoniensis zones. Evidence of the McL crenulata Zone which is the uppermost zone of the Llandovery Series and, possibly, of the C. centrifugus Zone of the Wenlock have been found immediately south of the district. These horizons lie, respectively, 100 m, 350 m, 650 m and 1100 m above the base of the group as defined along the southern side of the Ettrick Valley. The McL crenulata occurrence is in a locally more thin-bedded part of the succession, which acted as a roof thrust horizon in this part of the outcrop. Further work in adjacent areas may show the Hawick Group to be subdivisible at this horizon.

Current directions from bottom structures indicate a dominant north-east–south-west current orientation identical to that of the Gala Group.

Deposition of the Gala/Hawick submarine fans

With the possible exception of the Hawick Group, the onset of greywacke deposition in the Southern Uplands occurred at progressively later dates towards the south-east (McKerrow et al., 1977). In the Moffatdale area fan deposits of the Gala Group are entirely of middle fan facies. They range from the smooth, distal portion of a suprafan lobe (Facies C) to the proximal, channelised portion of the lobe, possibly the margin of the inner fan (Facies A/B and D). Outer fan deposits are absent and the middle fan facies were all laid down directly over pelagic muds of the Moffat Shale Group. This base-absent progradation suggests that a mature fan which, on the basis of flute marks, would have been distributing sediments south-westwards was gradually shifted laterally towards the south-east. Such migration is typical of foreland basins normally developed over continental crust in front of advancing thrust sheets. The accretionary model of McKerrow et al. (1977) offers an alternative explanation since progressive accretion and uplift of sediments against the continental margin would cause the depositional trench to migrate oceanwards.

By the end of the mid-Llandovery a major channel existed in the north-west of the Moffatdale area, part of a high-efficiency fan delivering sediment to areas farther southwest. Adjacent to the channel the pelagic sediments probably rose, diapirically, in response to loading by the channel deposits and restricted lateral sediment transport. Overbank deposits were laid down in this adjacent area and gradually extended some 10 km south-eastwards (Figure 10)a. Following deposition of the R. maximus Subzone shales an influx of coarse clastic sediment swept across the area south-east of the rise and built up a suprafan lobe some 25 km wide. The channelled portion of this lobe was developed predominantly along its north-western margin and the deposits thin out and become more distal south-eastwards towards the line of the Ettrick Valley Fault (Figure 10)b. The Hawick Basin, farther south-east, may also have been receiving clastic sediments at this time.

The deposition of coarse clastics on the Gala fan in this area ceased at least temporarily, towards the top of the M. turriculatus Zone. Since fossils of higher Llandovery zones occur in the Innerleithen district, farther north (Toghill and Strachan, 1970), sedimentation must have continued above the Gala fan but may have been dominated by finer-grained deposits. In the Hawick basin coarse clastic deposition continued (Figure 10c) and may have extended northwards above the Gala fan deposits.

Hawick Group sediments are of middle-fan facies, but the sequence is considerably thicker than the Queensberry Formation of the Gala Group and was deposited over a longer period of time with quiescent intervals when hemipelagic muds were laid down. Their outcrop is 4 km wide (across strike) in the Moffatdale area which, after removal of the effects of thrust faulting, represents some 10 km of a much wider depositional basin. The outcrop extends a further 16 km to the south-east where the Group passes beneath or, in part, changes laterally into the Riccarton Group of Wenlock age (Warren, 1964). The latter has lithological affinities with Wenlock strata in the English Lake District and suggests that both groups may have been deposited in a large basin representing the last remnants of the Iapetus Ocean prior to its final closure in the early Devonian.

Biostratigraphy

Greywackes, deposited from turbidity currents, represent much shorter intervals of time than corresponding thicknesses of Moffat shale. At times the turbidity currents waned or were diverted so that the supply of coarse sediment temporarily ceased and there was a return to a Moffat Shale Group type of environment, with deposition of black shale and accumulation of graptolite remains. These intervals were evidently of comparatively short duration and rare occurrence since the shales are generally very thin, of the order of 2 to 3 cm, and are separated by thicknesses of greywackes measured in hundreds of metres. Consequently exposures of graptolitic shale are difficult to find and although some were reported by early workers, such as Lapworth and Peach and Home, new discoveries are still being made, including several during the present revision. The reference sections for those Llandovery graptolite zones which these strata span are located in the Lake District and Central Wales. Nevertheless, the R. maximus Subzone and all overlying Llandovery zones have been proved within the groups (Table 3), although several of the locations lie outside the boundaries of the present district.

The base of the Gala Group becomes progressively younger south-eastwards. At the north end of Mid Craig, Loch Skeen [NT 1630 1668] M. sedgwickii Zone fossils occur just below the base of the Gala sandstones; this is the highest zone recognised in the Moffat Shale Group at this locality. The fauna from Seavy Sike [NT 177 146] may indicate the succeeding R. maxims Subzone of the M. turriculatus Zone, and the subzone fauna is certainly present from Dob's Linn [NT 1958 1583] south-eastwards to Black Grain [NT 2243 1407]. Farther south-east in Back Burn (see below) and across the Ettrick Valley Fault at Entertrona Burn [NT 1872 0800] and Master Grain [NT 2307 1190] the R. maximus Subzone faunas are present, within the basal Gala sandstones.

Monograptus turriculatus Zone, Rastrites maximus Subzone

At Back Burn, approximately 30 m downstream from its confluence with Black Grain [NT 2251 1334], 3.5m of shale includes graptolite horizons in which Monograptus aff. pseudoruncinatus and Diversograptus runcinatus are relatively common, as well as R. maximus. An outcrop of what may be the same shale repeated by faulting can be examined in Back Burn, by the sheepfold [NT 2244 1340], a few metres upstream from the confluence with Black Grain. It contains a similar graptolite fauna, but there are also thin interbedded metabentonites, not seen at the downstream locality.

Monograptus turriculatus Zone, Post-R. maximus Subzone

The upper part of the M. turriculatus Zone has not been firmly identified. A thin shale bed in Glendearg (north) Burn [NT 2129 1023] has yielded Monograptus cf. plumosus, M. proteus, Monoclimacis? and Rastrites sp., which may represent this zone but could indicate a higher one. This shale lies some 100 m above the base of the Gala Group. Graptolite occurrences of similar age are recorded at a number of other localities including Back Burn [NT 2264 1325]. Locally, up to 50 m of Gala sandstones occur above this level.

Monograptus crispus Zone

This zone is defined principally on the occurrence of M. crispus which has not been found within the district, but does occur along strike in the area to the north-north-west. Thin shale beds tentatively assigned to this zone occur some 100 m above the base of the Hawick Group at Master Grain. The latter locality yielded M. exiguus subspp.

Monoclimacis griestoniensis Zone

This zone has been proved at a small section alongside a forestry track [NT 2214 0971] close to the watershed separating the northerly and southerly flowing streams, both named Glendearg Burn. The graptolites occur in black shale and the base of an associated greywacke, and include Mcl. griestoniensis and M. exiguus exiguus.

Monoclimacis crenulata Zone

This is the uppermost zone of the Llandovery Series and although no evidence was found of its presence within the area covered by the map, it was proved in Barr Burn, at a locality approximately 1 km along strike to the south-southwest [NT 2272 0500].

At a locality [NT 2307 0612] in Glendearg (south) Burn, only 70 m beyond the south-east corner of the map, thin shales have yielded the youngest recorded graptolite assemblage from the Hawick Group. Its age is uncertain, as the fauna is not exclusively diagnostic of a particular zone, but is considered to belong either to the Monoclimacis crenulata or Cyrtograptus centrifugus zones, the boundary between which marks the Llandovery–Wenlock Series junction.

Chapter 2 Structure

An understanding of the structure of the Moffatdale area is fundamental both to the production of a geological map and to any further geological investigation. Good exposure occurs only in the lower reaches of the numerous tributaries to the Ettrick, in a strip of ground extending south-westwards from Loch Skeen and, to a lesser extent, along Moffatdale. Elsewhere exposures are very limited. Superficially the stream sections and other, more isolated exposures show a simple north-westwards-dipping succession of turbidite sandstones, the Gala and Hawick groups, with sporadic outcrops of black shales. Since the black shales are largely referable to a single lithostratigraphic unit, the Moffat Shale Group, their repetition may be explained by isoclinal folding and/or strike faulting (Figure 11). Although they recognised the local importance of strike-parallel thrust faults, Peach and Horne, following Lapworth, considered isoclinal folding the more important structural feature and drew their maps and sections accordingly (Figure 12). Since the discovery that sedimentary structures within the rocks indicate their facing direction, folds have been found to be more restricted than was previously thought and thrusts identified as the more important structures over most of the area. The overall 'structural style appears to be that of an imbricate thrust system (Figure 13) in which thrusts rise as listric splays from a more gently inclined sole thrust developed along an underlying ductile horizon, in this case the Moffat Shale Group.

The most important of the thrusts, the Ettrick Valley Fault, separates two tectono-stratigraphic units. North-west of the fault the Gala Group forms a south-eastwards-thinning duplex bounded by the sole thrust within the Moffat Shale Group and a roof thrust developed along some overlying, more ductile, horizon which has now been largely removed by erosion. South-east of the fault the duplex terminates and the imbricate fan becomes partly blind so that folds begin to predominate over thrusts at outcrop level. This change in structural style accompanies a change in lithostratigraphy as the Gala Group passes into the thicker and mainly younger Hawick Group.

The major folds and thrust faults

The main deformation which produced the imbricate structure involved both folding and thrusting. The folds are the earlier structures. They are commonly tight with axial traces trending approximately north-east–south-west. They are inclined or overturned towards the south-east and their axes plunge gently to moderately both to the north-east and, less commonly, to the south-west. The poor alignment of plunge culminations and depressions suggests that individual folds are periclinal and do not persist over long distances. In the Gala and Hawick groups sandstone beds behaved differently to siltstone beds during deformation. The sandstone beds maintain a constant thickness around folds giving them a class IB (Ramsay, 1967), concentric style. Siltstone beds are considerably thinned in the limbs of folds and thickened in the hinge regions producing folds of class III. The alternation of these two classes enables the folds to propagate indefinitely through the succession. The concentric fold style of the sandstone leads to pinching of the anticlinal closures towards the bases of sandstone units and of the synclinal closures towards the tops of units. Fusion of sandstone beds within synclinal closures and partly intrusive downwarping of such closures into the underlying siltstone (Webb, 1983) indicate initiation of folding whilst the sediments were still poorly consolidated.

Complete folds are only well developed in the Hawick Group and in a strip of ground south-west of Loch Skeen where relatively thin, massive sandstone is interbedded with thick, more ductile siltstone. Elsewhere the south-east-facing limbs have been largely sheared out and replaced by thrusts. Only a limited amount of shortening is possible by concentric folding because the folds become isoclinal in their pinched cores and lock up. For shortening to continue without flattening the folds must thrust. In the resultant imbricate structure (Figure 14) thrusts are intimately related to folds. Since it is known that concentric buckle folds develop with a particular dominant wavelength which is proportional to the thickness of the bed or beds being folded, it should follow that thrusts, initiated on such folds, will cut up through the beds at distances proportional to the thickness of the thrust slices. This hypothesis is extremely important for predicting the cross-sectional lengths of imbricate slices on geological sections and the amount of movement on the thrusts. Only in the area south-west of Loch Skeen, where siltstone/sandstone contacts act as marker horizons within the Gala Group, can the amount of movement on thrusts be measured directly. Above the Selcoth Burn [NT 159 062], near Craigmichan Scaur, minor imbricate slices involving single sandstone beds or small groups of beds show cross-sectional length to thickness ratios of between 3 and 10. A ratio of 6 has been obtained for similar slices at an exposure alongside the Hawick road [NT 318 164], near Tushielaw some nine kilometres east of the Moffatdale area (Webb, 1983) and this has been used as an average figure in the construction of the geological section shown on the face of the accompanying map.

Gala Group Duplex

Within the outcrop of the Gala Group duplex the predominant dip of bedding is steep to the north-west and beds almost invariably young in that direction. North-westward from the Ettrick Valley Fault uninterrupted sequences of beds, representing single thrust slices, increase in thickness from around 100 m to nearly 500 m thick north-west of Moffatdale. These sequences are separated by strike-parallel reversed faults or high-angled thrusts whose fault planes dip subparallel to, or more steeply than, the predominant dip of bedding in the sequences they separate. Apart from the area south-west of Loch Skeen, folds are poorly developed and closely associated with the thrusts. Fold closures are only sporadically exposed but anticlines invariably lie immediately north-west of thrusts, on their hangingwall side, and synclines lie immediately south-east of thrusts, on their foot-wall side. The axial planes of such fold pairs commonly dip only a few degrees more steeply than the thrust plane so that the south-east-facing common limb of the folds is extended, sheared and largely replaced by the fault (Figure 15). Locally, for example at [NT 201 112], this limb is represented by thrust-bounded slices or horses of strata cut off by rejoining splays (Figure 13) and in many cases the limb is completely absent (Figure 15)b. Ductile lithologies such as the Moffat Group shales, laminated siltstones and silty mudstones within the Gala Group crop out along the thrust planes. Shearing of these lithologies is commonly evident but may be obscure or even absent with only folds present (Figure 15)c.

The base of the Gala Group can be traced across thrust slices at a number of localities but it is rarely completely exposed. Towards the trailing edge of a slice this boundary is commonly complex with subsidiary blind thrusts and rejoining splays producing much minor imbrication which involves both individual sandstone beds and small groups of beds. Such features are well exposed above the Selcoth Burn near Craigmichan Scaur [NT 1613 0622]. Convergence of the thrusts with depth is most clearly indicated at Swine Cleuch [NT 172 134] where five thrusts cutting the Gala Group pass down into the Moffat Shale Group. Where they enter the shales, the thrusts have a total separation of some 350 m but downhill to the south-west the width of the shale outcrop decreases to 50 m with no evidence for the thrusts having re-entered the Gala Group. Parallelism of the thrusts with bedding in the underlying slice indicates their propagation along a ductile horizon within or above the Gala Group sandstones. The decrease in thickness of the thrust slices from north-west of Moffatdale towards the Ettrick Valley could be related to south-eastwards thinning of the Gala sandstones between this higher roof thrust horizon and the underlying sole thrust. In a duplex structure such thinning should be accompanied by a decrease in the inclination of the thrust faults. North-west of Moffatdale the dip of bedding rarely falls below 70° but on the south-eastern side of the dale and in the upper reaches of the Ettrick tributaries dips become more variable and are commonly between 50° and 60°. Nearer the Ettrick River and in Craigmichan Scaur dips below 40° become common. Since the thrusts are sub-parallel to bedding they too must decrease in inclination south-eastwards as predicted by the duplex model.

The acceptance of a duplex model for the structure allows two important predictions to be made.

1. Assuming that the thickness of the Gala Group increases relatively uniformly from the Ettrick to north-west of Moffatdale then the separation of the major thrusts should also increase uniformly. A locally excessive separation is only permissible if it can be shown that folds are present with at least parts of the south-east-facing limbs included within the intervening thrust slice. Separations less than the local norm are possible near the leading and trailing edges of thrust slices where minor imbrication is commonly present.

2. If there is no evidence for a lateral change in the thickness of the Gala Group sandstones then thrust slices should remain fairly constant in thickness along strike. Slices can terminate where a splay joins the thrust from which it arises and the slice may thin approaching the termination.

Since these predictions are nowhere contravened in the better-exposed ground they have been used to position major thrusts in areas of poor exposure. Descriptions of these more conjectural structures are given below. It follows that the resultant pattern of thrusts shown on the accompanying map and section must be oversimplified.

As exposure deteriorates in the middle reaches of the Ettrick tributaries two strike-parallel outcrops of the Moffat Shale Group are exposed. In Brockhope Burn [NT 209 137] and [NT 210 136] these have a separation of 150 m, the local thrust slice thickness. South-westwards, the outcrops diverge until they have a separation of about 600 m in Longhope Burn [NT 179 101] and [NT 184 098]. The duplex model predicts the presence of two, locally three, further thrust slices in the south-west related to splays rising from the lower thrust. The limited field evidence permits this hypothesis although possible splays are only rarely exposed [NT 1896 1083] and [NT 1982 1167].

Farther north-west, in the upper reaches of the tributaries and into Moffatdale, exposure is very poor. A predicted thrust separation approaching 300 m is to some extent substantiated by ground features [NT 188 127], [NT 190 125] and [NT 191 123] but the actual structure will undoubtedly be more complex.

The thrust slice immediately south-east of Dob's Linn has an outcrop width of 1000 m but thins south-westwards being only 400 m wide in Bodesbeck [NT 1563 0944] to [NT 1610 0942]. This is partly due to south-westwards thinning of the Gala Group but folds are also present [NT 2039 1582], [NT 1585 0946] and [NT 1602 0945]. The proposed structure, although fitting the limited field evidence, is clearly conjectural.

Within the next thrust slice, immediately north-west of Dob's Linn, the Gala Group also thins to the south-west and the folding is preserved within the slice. Beds dip predominantly 60°–80° to the north-west but this falls to below 30° near the foot of Tail Burn and on Deacon Snout [NT 187 151]. This is interpreted as a major early fold which failed to thrust and on this basis a sizeable inlier of the Moffat Shale Group is predicted in the unexposed floor of Moffatdale. An alternative hypothesis considers the fold as a ramp structure developed where this slice has been thrust over a horse of sandstones trapped between it and the underlying slice to the south-east. In this case the outcrop of the Moffat Shale Group in the valley floor would be considerably reduced.

Near the south-east end of Loch Skeen outcrops of the Moffat Shale Group [NT 1687 1634] and [NT 1713 1572] define a thrust slice 500 m thick. South-westwards, through White Coomb, the outcrop width of this slice increases to 2500 m at the western edge of the map. Sedimentological evidence indicates south-westwards thickening of the succession but it seems unlikely that this is the sole explanation. In the well-exposed thrust slices to the north-west the Gala Group comprises two sandstone units separated by siltstone. Similar siltstone is present along the north-western flank of White Coomb where the rocks are tightly folded [NT 163 154]. Tentative correlation with horizons of siltstone and thin-bedded sandstone in Carrifran [NT 157 124] and Blackhope [NT 144 106] burns permits the inclusion of a higher sandstone unit in this thrust slice accompanied by repetition of the lower unit partly by thrust folds.

South-eastwards from the Ettrick Valley Fault exposure becomes increasingly limited. The Gala Group duplex persists south-eastwards for about 1 km with much minor imbrication. Thrusts define the bases of the imbricate slices within the Moffat Shale Group at least as far south-east as the thrust bounding the large shale inlier at Entertrona Burn [NT 1867 0803]. Smaller inliers are highly sheared and have not yielded graptolites. Thin black shale interbedded with the Gala sandstones and yielding high M. turriculatus Zone graptolites crops out some 50 m below the tops of imbricate slices at several localities [NT 2129 1023], [NT 2136 1001] and [NT 2264 1325].

Hawick Group

The thrusts defining the top and bottom of the most northerly thrust slice of the Hawick Group were described in the 'stratigraphy' section. These suggest that the sole thrust to the Hawick Group lies some 100 m below a shale horizon within the M. crispus Zone and the roof thrust some 300 m above it. The sole thrust is positioned at the same horizon as the roof thrust to the Gala Group and this requires thrust slices of the Gala Group to lie beneath at least the initial Hawick Group slices. A graptolitic shale bed possibly of the M. crispus Zone crops out in the next Hawick Group thrust slice to the south at Buzzard's Linn [NT 2130 0852] near the axial trace of an isoclinal anticline. It is separated from the Mcl. griestoniensis horizon to the north [NT 2214 0971] by only a little over 200 m of strata suggesting that the higher shale bed could also be present in the first thrust slice. In Master Grain a thin black shale unit crops out at approximately the same level [NT 2311 1180] but its age is uncertain. The return to north-east-facing strata south-east of the anticline at Buzzard's Linn probably marks the outcrop of a thrust passing through cols on Mitchell Hill [NT 218 087] and Bloodhope Head [NT 225 095]. South-east of this thrust, folds become the predominant structural feature and, although faults are locally present in the axial zones of folds, major thrusts are not readily apparent. Thus, between Buzzard's Linn and Strongcleuch [NT 224 075] a number of well-exposed folds with overturned, south-east-facing limbs bring in younger strata to the south-east. No graptolites have been recorded however and it is presumed that these beds should all be assigned to the Mcl. griestoniensis Zone. Even though the stream section along the south-flowing Glendearg tributary to the River Esk is almost unbroken, graptolitic shales are only seen at [NT 2307 0612], 70 m downstream from the map boundary. These shales yield graptolites of the uppermost Llandovery or lowest Wenlock zones and occur at the top of the local succession in the overturned, south-east-facing limb of a major anticline. Minor folds in the axial region of the anticline are exposed between Kid Sike [NT 227 064] and Aspic Sike [NT 225 070]. However, a thrust may be present at Kid Sike since the south-east-facing limb of the major fold is not present in neighbouring Bloodhope Burn [NT 237 072] to the north-east. The Mcl. crenulata Zone, at the top of the Llandovery Series has not been unequivocally identified in Glendearg Burn although it occurs in Barr Burn [NT 2272 0500] 1 km south of the south-east corner of the map.

The main cleavage and related structures

The main cleavage is a variably penetrative fracture cleavage of pressure solution type. It is commonly well developed in the finer-grained lithologies of the Gala and Hawick groups but surprisingly absent in the Moffat Shale Group. It has a north-east–south-west trend and is generally congruous with the main folds although not exactly axial-planar to them, commonly striking a few degrees clockwise of the fold axial planes. At the Hawick road section [NT 318 164] the cleavage is superimposed across imbricated beds and preferentially developed on the more gently dipping, north-west-facing limbs of the folds (Webb, 1983). The poor development of the cleavage on the steep south-east-facing limbs suggests that bedding here was already in an orientation close to that of the cleavage and thus the development of a new planar fabric was obviated. Post-lithification tightening of the folds leading to the production of brittle accommodation shears by tangential longitudinal strain may be related to shortening during cleavage formation as may be sporadic, moderately north-west-dipping, brittle thrusts of small displacement which cut south-east-facing fold limbs, for example at [NT 1464 0836].

In the Moffat Shale Group fissility is dominantly bedding-parallel but this is disrupted by numerous, complex, anastomosing shear planes associated with tight, aberrant, steeply plunging and locally downward-facing folds. In the Whithorn area of the Southern Uplands (Barnes, 1989) rotation of early folds into steeply plunging attitudes and the generation of new folds which are commonly steeply plunging and locally downward-facing is related to a superimposed, sinistral, simple shear strain which also reactivated the major thrusts as sinistral wrench faults (Rust, 1965). A major phase of sinistral transpression involving both a north-east–south-west sinistral simple shear strain and a north-west–south-east flattening, or pure shear strain, appears to have affected the Caledonides in the late Silurian to early Devonian (Soper and Hutton, 1984). In the Irish Caledonides, Soper and Hutton (1984) postulate that the main cleavage developed during this transpressive phase and its orientation was controlled by variations in the shear strain component. Such a model is applicable in the Moffatdale area. Under a transpressive regime the major thrusts would be reactivated as sinistral wrench faults and take up most of the simple shear component of the strain. The Moffat Shales along the faults would be subject to intense shear producing the shear planes and aberrant folds. In the intervening sandstone slices flattening would be dominant, tightening the earlier structures, but a small component of sinistral shear would cause the cleavage to develop slightly clockwise of the earlier folds (Figure 16). At Bill Cleuch [NT 1580 0672] in the Gala Group sandstones, a steeply plunging fold may have formed during such deformation. Folds of this type are rare in the Moffatdale area presumably because most of the simple shear strain was taken up by movements on the numerous, pre-existing faults. In the Whithorn area radiometric dates of 418–395 Ma (Rock et al., 1986) from post-cleavage dykes indicate that there, at least, the cleavage must have developed early in the transpressional phase or before it.

Minor, post-cleavage structures

Minor, subvertical shear belts and wrench faults are quite commonly exposed. Although some aberrant examples occur, a north- or north-north-west-trending fault set with sinistral displacements of up to a few metres is dominant, for example at [NT 1478 1467] and [NT 2179 1406] and sporadic north-westward-trending examples with dextral displacement are probably a conjugate set, for example at [NT 2156 1130]. This strain is also manifest as minor, subvertical, open folds, for example at [NT 1444 1065] and [NT 1840 1486] with pronounced axial-planar joint sets. The joint sets are quite common north-west of the Ettrick watershed. They locally replace bedding as the dominant planar structure visible at exposure, for example at [NT 144 136] and are sporadically closed-spaced, resembling a fracture cleavage. Post-cleavage dykes are intruded along some of the wrench faults and, in the Whithorn area, may be cut by them (Barnes, Rock and Gaskarth, 1986).

In the Gala Group at Black Grain [NT 2236 1376] the inverted limb of a first fold, probably a thrust-bounded horse, is refolded about a poorly defined, moderately to steeply dipping axial plane trending north-east–south-west. The later fold closure is poorly exposed and facing evidence on the steeper limb is unclear but it is comparable with minor, post-cleavage folds in the Whithorn area (Barnes, 1989). These have steep, south-eastward dipping axial planes and form a conjugate set with other minor folds whose axial planes have a gentle dip (Figure 17). Correlatives of this second set occur in the Moffat Shale Group at Black Grain [NT 2244 1411] and elsewhere in the Moffatdale area but many are difficult to distinguish from the effects of superficial, gravity-induced creep (for example, Dob's Linn).

Structural and plate-tectonic synthesis

Alternative models have been proposed for the development of the Southern Uplands terrane. An accretionary prism developed above a north-west-dipping subduction zone was proposed by Leggett et al. (1979) whereas Stone et al. (1987) envisaged a sequential back-arc to foreland basin thrust system.

In both models a south-east-propagating thrust belt would have been developed. However, if deformation of these rocks was consequent on the closure of the ocean and the subsequent collision of the Laurentian and Avalonian or Cadomian (Soper and Hutton, 1984) continental plates two phases of deformation should be separable; one pertaining to the accretion of the rocks the other to the collision event. In the Moffatdale area there is no evidence for any cessation of tectonic activity between these two phases, but the structures do indicate a change in the type of strain during what may have been one continuous deformation spanning the whole of the Silurian Period.

The earliest deformation was penecontemporaneous with deposition of the sediments. Strain is manifest as folds, overturned and thrust south-eastwards. It can be resolved in terms of a pure shear strain (flattening) with maximum shortening north-west–south-east and maximum extension subvertical, coupled with a dextral, simple shear strain in which the shear plane dips relatively gently north-westward, with the movement direction up-dip (Figure 17). The later deformation was transpressional. It can be resolved in terms of the same pure shear strain as the early deformation coupled with sinistral simple shear in which the shear plane is subvertical, north-east– south-west-trending and the movement direction is subhorizontal (Figure 17). Previous workers (Eales, 1979; Stringer and Treagus, 1981) have related the whole of this deformation sequence to accretion tectonics invoking oblique subduction to account for the sinistral shear strain. This may be correct, although Soper and Hutton (1984) have proposed that sinistral shear strain reached a peak in this part of the Caledonides in late Silurian to early Devonian time. If the sinistral shear replaced the early south-eastwards extension it may be related to a collision deformation, the restraint on south-eastward extension being caused by contact with the Avalonian plate margin. The generation of much of the main cleavage during the collision event would mean it could be largely equated with the morphologically similar, main cleavage in the English Lake District (the Avalonian island arc) which developed immediately prior to the intrusion of the Skiddaw Granite (Soper and Roberts, 1971) dated at 399 + 14 Ma (Miller, 1961).

The orientation of the minor conjugate wrench faults is compatible with Caledonian compression (Figure 17). They were probably initiated at an early date and reactivated at various times during the continuous deformation. The conjugate fold sets with gently and steeply dipping, north-east–south-west-trending axial planes imply a northwestwards-plunging compression direction (Figure 17). They may have resulted from gravitational collapse of the tectonically thickened sequence as Caledonian compression waned. Comparable folds with subhorizontal axial planes are present in the Lake District where they immediately postdate the main cleavage (Soper and Roberts, 1971).

Deep structure

The deep structure of the Southern Uplands is poorly understood and problematic. The present structural interpretation of the exposed Lower Palaeozoic rocks indicates that they probably extend to only a little over 2 km below surface. The accretionary hypothesis would predict that they are underlain by thrust slices of oceanic cherts and basalts, the upper layers of the Iapetus oceanic crust, but this is not supported by geophysical evidence; magnetic evidence is equivocal and the LISPB seismic profile (Bamford et al., 1977) predicted up to 15 km of Lower Palaeozoic sedimentary rocks above a thin continental basement. More recent seismic work (Hall et al., 1983) supports the structural evidence for a thin (less than 5 km) sedimentary cover but again indicates a granitic basement which appears to be in continuity with that beneath the Midland Valley to the north. If the accretionary hypothesis is correct then the apparent absence of the Iapetus oceanic crust could be because the accretionary complex was either thrust south-eastwards over the Avalonian continental foreland (Leggett et el., 1983) or backthrust, north-westwards over the Laurentian foreland (Hall et al., 1983). Alternatively, it has been suggested that the accretionary hypothesis is incorrect and that a supra-continental, back-arc setting is more plausible (Stone et al., 1987).

Chapter 4 Minor intrusions

The igneous dykes and sills of the area were not examined in detail. Dolerite is most common but lamprophyre and felsite occur sporadically, the latter generally as discontinuous lenses within the Moffat Shale Group. The dolerite dykes have been predominantly intruded along the north-west-trending wrench faults and shear belts and are from 0.2 to 5.0 m thick and generally discontinuous at outcrop. The larger dykes commonly include a central zone of brecciated country rock, for example at [NT 201 128]. The dolerite commonly weathers to a creamy white clay, for example at [NT 186 108]. Exposure is insufficient to clearly demonstrate any relationship between intrusive activity and tectonics but the discontinuity and degree of alteration of the dolerite intrusions suggests that the majority are Caledonian in age.

References

BAMFORD, D, NUNN, K, PRODEHL, C, and JACOB, B. 1977. Upper crustal structure of northern Britain. Journal of the Geological Society of London, Vol. 133, 481–488.

BARNES, R P. 1989. Geology of the Whithorn district. Memoir of the British Geological Survey, Sheet 2, Scotland.

BARNES, R P.,  ROCK, N M S, and GASKARTH, J W. 1986. The Caledonian dyke-swarms in Southern Scotland: new field and petrological data for the Wig-town Peninsula, Galloway. Geological Journal, Vol. 21, 101–125.

BERRY, W B N. 1962. Graptolite occurrence and ecology. Journal of Palaeontology, Vol. 36, 285–293.

COCKS, L R M. 1985. The Ordovician–Silurian boundary. Episodes, Vol. 8, 98–100.

COCKS, L R M.  and RICKARDS, R B (editors). 1988. A global analysis of the Ordovician - Silurian boundary. Bulletin of the British Museum (Natural History): Geology, Vol. 43, 1–394.

CRAIG, G Y, and WALTON, E K. 1959. Sequence and structure in the Silurian rocks of Kirkcudbrightshire. Geological Magazine, Vol. 96, 209–220.

DEWEY, J F, and SHACKLETON, R M. 1984. A model for the evolution of the Grampian tract in the early Caledonides and Appalachians. Nature, London, Vol. 312, 115–121.

EALES, M H. 1979. Structure of the Southern Uplands of Scotland. 269–273 in The Caledonides of the British Isles-reviewed. HARRIS, A L, HOLLAND, C H, and LEAKE, B E (editors). Special Publication of the Geological Society of London, No. 8.

ELLES, G L, and WOOD, E M R. 1904. A monograph of British graptolites. Part 4. Monograph of the Palaeontographical Society, Vol. 58 for 1904.

ELLES, G L, and WOOD, E M R. 1906. A monograph of British graptolites. Part 5, 181–216. Monograph of the Palaeontographical Society, Vol. 60 for 1906.

ELLES, G L, and WOOD, E M R. 1914. A monograph of British graptolites. Part 10, 487–526. Monograph of the Palaeontographical Society, Vol. 67 for 1913.

FLOYD, J D. 1982. Stratigraphy of a flysch succession: the Ordovician of W. Nithsdale, SW Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 73, 1–9.

FYFE, T B, and WEIR, J A. 1976. The Ettrick Valley Thrust and the upper limit of the Moffat Shales in Craigmichan Scaurs (Dumfries & Galloway Region: Annandale & Eskdale District). Scottish Journal of Geology, Vol. 12, 93–102.

HALL, J, POWELL, D W, WARNER, M R, Z H M, ADESANYA, O, and BLUCK, B J. 1983. Seismological evidence for shallow crystalline basement in the Southern Uplands of Scotland. Nature, London, Vol. 305, 418–420.

HARKNESS, R. 1851. On the Silurian rocks of Dumfriesshire and Kirkcudbrightshire. Quarterly Journal of the Geological Society of London, Vol. 7, 46–65.

KIRK, N H. 1972. More thoughts on the automobility of the graptolites. Quarterly Journal of the Geological Society of London, Vol. 128, 127 –133.

LAPWORTH, C. 1878. The Moffat Series. Quarterly Journal of the Geological Society of London, Vol. 34, 240–346.

LAPWORTH, C. 1879–1880. On the geological distribution of the Rhabdophora. (9 parts). Annals and Magazine of Natural History, Series 5, Vols. 3–6.

LEGGETT, J K, MCKERROW, W S, and EALES, M H. 1979. The Southern Uplands of Scotland: a lower Palaeozoic accretionary prism. Journal of the Geological Society of London, Vol. 136, 755–770.

MARK, J. 1880. On the pre-Devonian rocks of Bohemia. Quarterly Journal of the Geological Society of London, Vol. 35, 591– 619.

MCKERROW, W S, LEGGETT, J K, and EALES, M H. 1977. An imbricate thrust model for the Southern Uplands of Scotland. Nature, London, Vol. 267, 237–239.

MERRIMAN, R J, and ROBERTS, B. 1990. Metabentonites in the Moffat Shale Group, Southern Uplands of Scotland: geochemical evidence of ensialic marginal basin volcanism. Geological Magazine, 127, 259–271.

MILLER, J A. 1961. The potassium-argon ages of the Skiddaw and Eskdale Granites. Geophysical Journal, Vol. 6, 391–393.

MOORE, J. 1849. On some fossiliferous beds in the Silurian rocks of Wigtonshire and Ayrshire. Quarterly Journal of the Geological Society of London, Vol. 5, 7–17.

MURPHY, F D, and HUTTON, D H W. 1986. Is the Southern Uplands of Scotland really an accretionary prism? Geology, Vol. 14, 354–357.

NICOL, J. 1848. On the geology of the Silurian rocks in the valley of the Tweed. Quarterly Journal of the Geological Society of London, Vol. 4, 195–209.

PEACH, B N, and HORNE, J. 1899. The Silurian rocks of Britain, Vol. 1, Scotland. Memoir of the Geological Survey of the United Kingdom.

RAMSAY, J G. 1967. Folding and fracturing of rocks. (New York: McGraw-Hill).

RICKARDS, R B. 1976. The sequence of the Silurian graptolite zones in the British Isles. Geological Journal, Vol. 11, 153–188.

RIVA, J. 1976. Climacograptus bicornis bicornis (Hall), its ancestor and likely descendants. 589–619 in The Ordovician System. BASSETT, M G (editor). (University of Wales Press.)

ROCK, N M S, GASKARTH, J W, and RUNDLE, C C. 1986. Late Caledonian dyke swarms of southern Scotland: a regional zone of primitive K-rich lamprophyres and associated vents. Journal of Geology, Vol. 94, 505–522.

RUST, B R. 1965. The stratigraphy and structure of the Whithorn area of Wigtownshire, Scotland. Scottish Journal of Geology, Vol. 1, 101–133.

SIVETER, D J, INGHAM, J K, RICKARDS, R B, and ARNOLD, B. 1980. Highest Ordovician trilobites and graptolites from County Cavan, Ireland. Journal of Earth Sciences of the Royal Dublin Society, Vol. 2, 193–207.

SOPER, N J, and HUTTON, D H W. 1984. Late Caledonian sinistral displacements in Britain: implications for a three-plate collision model. Tectonics, Vol. 3, 781–794.

SOPER, N J. and ROBERTS, D E. 1971. Age of cleavage in the Skiddaw Slates in relation to the Skiddaw aureole. Geological Magazine, Vol. 108, 293–302.

STONE, P, FLOYD, J D, BARNES, R P, and LINTERN, B C. 1987. A sequential back-arc and foreland basin thrust duplex model for the Southern Uplands of Scotland. Journal of the Geological Society, Vol. 144, 753–764.

STRACHAN, I. 1960. The Ordovician and Silurian graptolite zones in Britain. XXI Session of the International Geological Congress, Part VII, 109–113.

STRINGER, P, and TREAGUS, J E. 1980. Non-axial planar cleavage in the Hawick Rocks of the Galloway area, Southern Uplands, Scotland. Journal of Structural Geology, Vol. 2, 317–331.

STRINGER, P, and TREAGUS, J E. 1981. Asymmetrical folding in the Hawick Rocks of the Galloway area, Southern Uplands. Scottish Journal of Geology, Vol. 17, 129–148.

TOGHILL, P. 1968. The graptolite assemblages and zones of the Birkhill Shales (Lower Silurian) at Dobb's Linn. Palaeontology, Vol. 11, 654–668.

TOGHILL, P. 1970a. Highest Ordovician (Hartfell Shales) graptolite faunas from the Moffat area, south Scotland. Bulletin of the British Museum (Natural History): Geology, Vol. 19, 1–26.

TOGHILL, P. 1970b. A fauna from the Hendre Shales (Llandeilo) of the Mydrim area, Carmarthenshire. Proceedings of the Geological Society of London, No. 1663, 121–129.

TOGHILL, P. 1970c. The southeast limit of the Moffat Shales in the Upper Ettrick Valley region, Selkirkshire. Scottish Journal of Geology, Vol. 6, 233–242.

TOGHILL, P. and STRACHAN, I. 1970. The graptolite fauna of Greiston quarry, near Innerleithen, Peebleshire. Palaeontology, Vol. 13, 511 - 521.

WALKER, R G, and Morn, E. 1973. Turbidite facies and facies associations. 119–157 in Turbidites and deep water sedimentation. Society of Economic Palaeontologists and Mineralogists Pacific Section, Short course.

WALTON, E K. 1955. Silurian greywackes in Peeblesshire. Proceedings of the Royal Society of Edinburgh, B, Vol. 65, 327–357.

WARREN, P T. 1964. The stratigraphy and structure of the Silurian (Wenlock) rocks southeast of Hawick, Roxburghshire, Scotland. Quarterly Journal of the Geological Society of London, Vol. 120, 193–222.

WEBB, B. 1983. Imbricate structure in the Ettrick area, Southern Uplands. Scottish Journal of Geology, Vol. 19, 387–400.

WEIR, J A. 1968. Structural history of the Silurian rocks of the coast west of Gatehouse, Kirkudbrightshire. Scottish Journal of Geology, Vol. 4, 31–52.

WILLIAMS, A, and others. 1972. A correlation of Ordovician rocks in the British Isles. Special Report of the Geological Society of London, No. 3.

WILLIAMS, S H. 1982a. Upper Ordovician graptolites from the top Lower Hartfell Shale Formation (D. clingani and P. linearis zones) near Moffat, southern Scotland. Transactions of the Royal Society of Edinburgh, Vol. 72, 229–255.

WILLIAMS, S H. 1982b. The late Ordovician graptolite fauna of the Anceps Bands at Dob's Linn, southern Scotland. Geologica et Palaeontologica, Vol. 16, 29–56.

WILLIAMS, S H. 1983. The Ordovician–Silurian boundary graptolite fauna of Dob's Linn, southern Scotland. Palaeontology, Vol. 26, 605–639.

WILLIAMS, S H. and RICKARDS, R B. 1984. Palaeoecology of graptolitic black shales in Aspects of the Ordovician System. BRUTON, P L (editor). (Oslo: Universitetsforlaget.)

WILSON, R B. 1977. A history of the Geological Survey in Scotland. (Edinburgh: Her Majesty's Stationery Office for Institute of Geological Sciences.)

Glossary

Figures, plates, tables and maps

Figures

(Figure 1) Location and regional setting of the area.

(Figure 2) Typical graptolites from the Moffat Shale Group (1)–The Glenkiln Shale and the lower part of the Hartfell Shale (Climacograptus wilsoni Zone). The figures are arranged approximately in ascending stratigraphical sequence from bottom left to top right. All the figures are about twice natural size. 1. Didymograptus superstes Lapworth 2. Dicellograptus sextans (Hall) 3. Orthograptus whitfieldi (Hall) 4. Hallograptus mucronatus (Hall) 5. Nemagraptus gracilis (Hall) 6. Dicranograptus rectus Hopkinson 7.Climacograptus bicornis (Hall) 8. Climacograptus wilsoni Lapworth 9. Dicellograptus patulosus Lapworth 10. Orthograptus intermedius Elles and Wood 11. Dicranograptus nicholsoni Hopkinson 12. Glossograptus hincksii (Hopkinson)

(Figure 3) Typical graptolites from the Moffat Shale Group (2)–The Hartfell Shale (above the Climacograptus wilsoni Zone). All the figures are about twice natural size. 1. Climacograptus caudatus Lapworth 2. Orthograptus basilicus Elles and Wood 3. Dicranograptus ramosus (Hall) 4. Corynoides calicularis Nicholson 5. Dicellograptus morrisi Hopkinson 6. Dicranograptus clingani Carruthers 7. Neurograptus margaritatus (Lapworth) 8. Dicellograptus elegans (Carruthers) 9. Dicellograptus pumilus Lapworth 10. Leptograptus flaccidus (Hall) 11. Plurograptus linearis (Carruthers) 12. Climacograptus tubuliferus Lapworth 13. Dicellograptus complanatus Lapworth 14. Orthograptus abbreviatus Elles and Wood 15. Dicellograptus anceps (Nicholson) 16. Climacograptus supernus Elles and Wood 17. Dicellograptus complexus Davies 18. Paraorthograptus paciflcus (Ruedemann)

(Figure 4) Typical graptolites from the Moffat Shale Group (3)—The Lower Birkhill Shale. All the figures are about twice natural size except (Figure 4) which is x1.3. 1. Glyptograptus persculptus Elles and Wood 2. Climacograptus normalis Lapworth 3. Parakidograptus acuminatus (Nicholson) 4. Cystograptus vesiculosus (Nicholson) (x1.3) 5. Atavograptus atavor (Jones) 6. Rhaphidograptus extenuator (Lapworth) 7 Dimorphograptus swanstoni Lapworth 8. Pribylograptus sandersoni (Lapworth) 9. Monograptus revolutus Kurck 10. Coronograptus typhus typhus (Lapworth) 11. Climacograptus rectangularis (McCoy) 12. Coronograptus gregarius (Lapworth) 13. Monograptus triangulatus triangulatus (Harkness) 14. Monograptus triangulatus fimbriatur (Nicholson) 15. Diplograptus magnus H. Lapworth

(Figure 6)). All the figures are about twice natural size. 1 Pribylograptus leptotheca (Lapworth) 2 Monograptus argenteus (Nicholson) 3 Petalograptus ovatoelongalus (Kurck) 4 Rhaphidograptus toernquisti (Elles and Wood) 5 Glyptograptus tamariscus tamariscus (Nicholson) 6 Monograptus lobiferus (McCoy) 7 Rastrites longispinus Perner 8 Monograptus limatulus T�rnquist 9 Monograptus convolutus (Hisinger) 10 Monograptus clingani (Carruthers) 11 Cephalograptus cornea (Geinitz) 12 Monograptus decipiens Tornquist 13 Clinoclimacograptus retroversus Bulman and Rickards 14 Pristiograptus regularis (Tornquist) 15 Monograptus sedgwickii (Forelock)." data-name="images/P992267.jpg">(Figure 5) Typical graptolites from the Moffat Shale Group (4)–The Upper Birkhill Shale (except for the Rastrites maximus Subzone, for which see (Figure 6)). All the figures are about twice natural size. 1 Pribylograptus leptotheca (Lapworth) 2 Monograptus argenteus (Nicholson) 3. Petalograptus ovatoelongalus (Kurck) 4. Rhaphidograptus toernquisti (Elles and Wood) 5. Glyptograptus tamariscus tamariscus (Nicholson) 6. Monograptus lobiferus (McCoy) 7. Rastrites longispinus Perner 8. Monograptus limatulus Törnquist 9 Monograptus convolutus (Hisinger) 10. Monograptus clingani (Carruthers) 11. Cephalograptus cornea (Geinitz) 12. Monograptus decipiens Tornquist 13. Clinoclimacograptus retroversus Bulman and Rickards 14. Pristiograptus regularis (Tornquist) 15. Monograptus sedgwickii (Forelock)

(Figure 6) Typical graptolites from the Gala and Hawick groups. The figures are numbered approximately in ascending stratigraphical sequence; their local recorded stratigraphical ranges are shown in Table 3. All are about twice natural size, except numbers 2, 7, 8 and 11 which are x4. 1. Rastrites maximus Carruthers 2. Diversograptus runcinatus (Lapworth) 3. Monograptus proteus (Barrande) 4.Monograptus turriculatus (Barrande) 5. Monoclimacis? galaensis (Lapworth) 6. Monograptus marri Perner 7. Monograptus exiguus exiguus (Nicholson) 8. Monograptus discus Törnquist 9. Monograptus crispus Lapworth 10. Monograptus priodon (Bronn) 11. Monoclimacis griestoniensis (Nicol) 12. Monoclimacis vomerinus vomerinus (Nicholson).

(Figure 7) Lithological and sedimentological subdivisions of a turbidite deposit.

(Figure 8) Turbidite lithologies found in the Moffatdale area and their relationship to turbidite facies.

(Figure 9) Turbidite facies and facies associations of the Moffatdale area related to a simple submarine fan model, for details of facies type see (Figure 8). Stipple represents dominant sand, horizontal ornament represents dominant silt.

(Figure 10) The deposition of the Gala and Hawick groups: a) Early M. turriculatus Zone (R. maximus Subzone). Main distributary channel of large fan deposits A/B facies of Gala Group. b) Late M. turriculatus Zone. Main channel migrates south-eastwards depositing A/B facies of Gala Group. A suprafan lobe extends south-east (B/C facies) towards another, parallel fan system, depositing the B/C facies of the Hawick Group. c) Mcl. griesioniensis Zone. Gala Group channel abandoned. Hawick Group channel develops and suprafan lobes (B/C facies) transgress north-west across the Gala Group sequence.

(Figure 11) Diagram illustrating how, in a repetitive sequence, folds and faults can only be identified from facing evidence.

(Figure 12) Geological sections in the upper Ettrick area from Peach and Horne (1899) Top: Generalised section from Black Knowe Head across Ettrick Water to Phawhope Kips. Simplified from Peach and Horne 1899.fig 17. Bottom: Generalised section from Potburn to Ettrick Pen Simplified from Peach and Home 1899. fig. 18 Key: 1. Glenkiln Chert and Shale 2. Lower Hartfell Shale 3. Upper Hartfell Shale 4. Lower Birkhill Shale 5. Upper Birkhill Shale 6. Gala and Hawick group.

(Figure 13) Terminology for thrust systems. See text for details.

(Figure 14) A simple imbricate thrust model after Webb (1983). a) Buckle folds (1) are tightened and rotated (2) until they lock up (3). Further shortening is only possible by thrusting (4). b) Section to illustrate the relationship between fold wavelength and bed thickness. c) Imbricate structure developed from (b) t = thickness of imbricating unit; Wd–dominant wavelength of folds; ld–dominant cross-sectional length of imbricate slice; l = length.

(Figure 15) Sketches of thrust faults in the upper Ettrick area. Numbers give dip and strike azimuth clockwise of direction of dip (after Webb, 1983). a) Cossarshill Scar [NT 2263 1333]. b) Bell Cleuch [NT 2191 1121]. c) Ettrick Water [NT 2258 1269]

(Figure 16) The relationship between folds, faults and cleavage under different strain regimes. a) Flattening with maximum shortening perpendicular to folds and faults produces parallel cleavage. b) Homogeneous, sinistral, simple shear parallel to folds and faults produces crosscutting cleavage. c) Combined flattening and simple shear (transpression). The simple shear is inhomogeneous, mainly concentrated along the faults, the small component affecting rocks between faults causes the cleavage to be slightly oblique to the folds. A model of this type explains relationships in the Moffatdale area.

(Figure 17) A summary chart of deformation in the district. Horizontal arrows show the development of structural features through time. Broken lines indicate weak development and solid lines indicate strong or main development of any feature.

Plates

(Plate 1) Thick-bedded greywacke sandstone with massive amalgamated units. Gala Group. The local amalgamation probably marks migratory braided channels. Mid-fan section of submarine fan.

(Plate 2) A minor anticline (F₁) in the Gala Group. The greywacke sandstones have ruptured at the angular hinge and minor imbrication is visible to the left. Banding in the interturbidite siltstones, in the core of the fold, is continuous round the closure.

(Front cover)

(Rear cover)

Tables

(Table 1) Zonal ranges of selected graptolites from the Glenkiln Shale and the Hartfell Shale, based on information drawn from Lapworth (1878, 1879), Elles and Wood (1918) and Williams (1982a,b) and on recent work.

(Table 2) Zonal ranges of selected graptolites from the Birkhill Shale, based largely on information drawn from Toghill (1968), and on recent work.

(Table 3) Zonal ranges of selected graptolites from the Gala and Hawick groups, based mainly on recent work in the Moffat and adjacent areas. The ranges are provisional.

Maps

(Map 1) Relief map of Moffatdale (inside front cover).

(Map 2) Geological map of Moffatdale (inside rear cover).