A Fluvial Origin for the Neoproterozoic Morar Group, NW Scotland

By Krabbendam, Maarten Prave, Tony; Cheer, David

Abstract: Precambrian sedimentary successions are difficult to date and correlate. In the Scottish Highlands, potential correlations between the thick, undeformed siliciclastic ‘Torridonian’ successions in the foreland of the Caledonian Orogen and the highly deformed and metamorphosed siliciclastic Moine succession within the Caledonian Orogen have long intrigued geologists. New and detailed mapping of the Neoproterozoic Altnaharra Formation (Morar Group, lowest Moine Supergroup) in Sutherland has discovered low-strain zones exhibiting well- preserved sedimentary features. The formation comprises 3-5 km of coarse, thick-bedded psammite with abundant nested trough and planar cross-bedding bedforms, defining metre-scale channels. Palaeocurrent directions are broadly unimodal to the NNE-ENE. We interpret the Altnaharra Formation as high-energy, braided fluvial deposits. The Altnaharra Formation and the unrnetamorphosed, Neoproterozoic Applecross-Aultbea formations (Torridon Group) are similar in terms of lithology, stratigraphical thickness, sedimentology, geochemistry, detrital zircon ages and stratigraphical position on Archaean basement. Depositional age constraints for both successions overlap and are coeval with late Grenvillean orogenic activity. Detrital zircons imply similar source regions from the Grenville Orogen. The Morar and Torridon groups can thus be correlated across the Caledonian Moine Thrust and are best explained as parts of a single, large-scale, orogen-parallel foreland basin to the Grenville Orogen.

The interpretation, correlation and age-control of Precambrian clastic sequences are hampered by a lack of biostratigraphical control. Post-depositional tectonometamorphic events may have obscured or destroyed sedimentological evidence and subsequent plate motions may have transported formerly adjacent source and sink regions over long distances. In addition, some controlling geomorphological factors such as rates of weathering and erosion were different in the Precambrian compared with more modernday processes (e.g. Eriksson et al. 2001). Over recent decades, the application of detrital zircon dating has provided a means to constrain the maximum age of deposition and the provenance of the detritus of such sequences (Froude et al. 1983; Nelson 2001; Cawood et al. 2004) and the database of such dates is growing fast (e.g. Cawood et al. 2007). However, correlations based solely on detrital zircon ages may be equivocal and in this study we use lithostratigraphy, sedimentology, geochemistry and published detrital zircon geochronology to interpret and correlate Neoproterozoic siliciclastic sequences in Northern Scotland.

The metamorphosed Morar Group in the Northern Scottish Highlands occurs east of the Caledonian (Silurian) Moine Thrust and is the structurally lowest part of the Moine Supergroup. It comprises several kilometres of siliciclastic rocks (mainly psammite), has a large (>2000km^sup 2^) outcrop area and is generally regarded as shallow marine in origin (e.g. Glendinning 1988; Holdsworth et al. 1994; Strachan et al. 2002).

West of the Moine Thrust, the unmetamorphosed Torridon Group represents a similarly widespread, thick and monotonous siliciclastic sequence. It is interpreted to be of mostly fluvial origin (Nicholson 1993; Stewart 2002). A number of workers have suggested that the ‘Torridonian’ (the informal stratigraphical parent of the Torridon Group) and the Moine Supergroup may be equivalent (Peach et al. 1907, 1913; Peach & Home 1930; Kennedy 1951; Sutton & Watson 1964; Johnstone et al. 1969; Nicholson 1993; Prave et al. 1999). However, such a correlation has generally not been accepted and the two sequences are formally regarded as distinct (Clough in Peach et al. 1910, p. 46; Gibbons & Harris 1994; Stewart 2002; Trewin 2002; Friend et al. 2003; Cawood et al. 2004). No thorough discussion or review of a potential correlation has been published since that of Kennedy (1951).

Here, we present new sedimentological and geochemical data for the Morar Group in the Northern Highlands of Scotland. These and other published data are used to compare and contrast the Morar and Torridon groups and to discuss possible basin interpretations. It is concluded that they represent a single foreland basin to the GrenviUe Orogen.

Geological setting

Moine Supergroup-Morar Group

The Moine Supergroup occurs east of, and structurally above, the Caledonian Moine Thrust and north of the Great Glen Fault (Fig. 1). After deposition, it was subjected to a number of tectonometamorphic events that have been the subject of some debate (Tanner & Bluck 1999, and references therein). The most common model involves an extensional event at c. 870 Ma, followed by Knoydartian (820-740 Ma) and Caledonian (470460 Ma and 430-400 Ma) orogenic events (Strachan et al. 2002). Sedimentary structures, especially in pelitic and semipelitic lithologies, are generally deformed, obscured or obliterated by greenschist- to amphibolite-facies metamorphism and deformation. A number of ductile thrust faults disrupt the stratigraphy (Johnstone et al. 1969; Barr et al. 1986; Holdsworth et al. 1994). Within the outcrop of the Moine Supergroup are several Lewisianoid basement gneiss inliers with late Archaean protolith ages that are broadly similar to Lewisian gneisses west of the Moine Thrust (Friend et al. 2007).

The Moine Supergroup has been divided into three groups: the Morar, Glenfinnan and Loch EiI (Johnstone et al. 1969; Holdsworth et al. 1994; Soper et al. 1998).

The Morar Group, the lowest and westernmost group (Figs 1 and 2), is dominated by psammite with minor pelitic, semipelitic and pebbly layers. Estimates of stratigraphical thickness in the literature (e.g. Holdsworth et al. 1994) are poorly constrained because of structural complexities. In Morar in the Western Highlands, the Group comprises four formations; these are in ascending order the Basal Pelite, Lower Morar Psammite, Morar Striped and Pelite and Upper Morar Psammite formations (Johnstone et al. 1969; Holdsworth et al. 1994). Locally, the base of the Basal Pelite Formation is marked by a thin, highly deformed basal meta-conglomerate, showing an unconformity with the Lewisianoid basement, and it is now generally accepted that the Morar Group was deposited unconformably upon Lewisianoid basement (Peach et al. 1910; Ramsay 1957; Holdsworth et al. 1994, 2001). Previously, sedimentary structures have been studied in detail only in the Upper Morar Psammite Formation (Glendinning 1988) and include tabular and trough cross- bedding in co-sets up to 0.5 m thick. Coarse-grained to gravelly psammite locally displays cross-beds >0.5 m thick. Most palaeocurrents are unidirectional to the north or NE, but bipolar ‘herring-bone’ cross-stratification and dunes and ripples with mudstone drapes are present locally. Glendinning (1988) interpreted the Upper Morar Psammite as a tidal, shallow-marine deposit but noted that the unit is unusually immature (arkosic) compared with other shallow-marine shelf deposits, and that a fluvial origin of these sediments could not be discounted.

In contrast, in the Northern Highlands north of Glen Oykel (Figs 1 and 2), the Morar Group stratigraphy is rather simple. It is dominated by the psammitic Altnaharra Formation (Fig. 2), which crops out over c. 1500 km2 (Figs 1 and 3). The Altnaharra Formation comprises several kilometres of psammite. A highly deformed pelitic- conglomeratic unit is intermittently present on or slightly above Lewisianoid Gneiss inliers (Mendum 1976; Holdsworth et al. 2001). The relationship between the Altnaharra Formation and the Altnaharra and Glascarnoch Formation above the Achness Thrust (Fig. 3) remains unclear.

The Morar Group is structurally overlain by the semipelitedominated Glenfinnan Group, but the contact is generally marked by the ductile Sgurr Beag Thrust (Fig. 1) and the original relationship between the two groups is unclear. The close association of Glenfinnan Group rocks and basement inliers suggests an original unconformable relationship (Holdsworth et al. 1994; Soper et al. 1998), and the group may represent a distal, lateral equivalent to the Morar Group. However, Morar Group rocks on the Ross of Mull appear to pass stratigraphically upward into Glenfinnan Group rocks; the section is, however, locally highly deformed and the field relationships are not unequivocal (Holdsworth et al. 1987). The Glenfinnan Group preserves few sedimentary structures and its depositional environment is unclear. The stratigraphically overlying psammite-dominated Loch EiI Group

(Roberts et al. 1987) contains locally abundant sedimentary structures including unidirectional and bipolar ‘herring-bone’ cross- bedding and wave ripples and has been interpreted as a shallow marine shelf deposit (Strachan 1986).

Torridon Group

The Torridon Group occurs west of the Caledonian Moine Thrust and in thrust sheets within the Moine Thrust Zone (Figs 1 and 3). The Torridon Group is generally unmetamorphosed and undeformed, except for gentle tilting. The Torridon Group was mostly deposited upon an exhumed land surface of Archaean Lewisian Gneiss with palaeo-relief up to 600 m (Peach et al. 1907; Stewart 1972). Locally, the group unconformably overlies the Mesoproterozoic Stoer and Sleat Groups, described elsewhere (Rainbird et al. 2001; Stewart 2002; Kinnaird et al. 2006). Including its inferred offshore extent, the Torridon Group currently occupies an area of c. 80 by 200 km (Stewart 2002). However, Torridon Group rocks also occur in the highest thrust sheets in the Moine Thrust Zone (e.g. Ben More Thrust Sheet and Kinlochewe Thrust Sheet; Peach et al. 1907; Butler 1997; Krabbendam & Leslie 2004; see also Fig. 3), so that prior to Caledonian thrusting the Torridon Group must have extended some 50-100 km farther east (Elliott & Johnson 1980; Butler & Coward 1984). The succession is c. 5-6 km thick but the top of the sequence is not exposed (Stewart 2002) because the group is unconformably overlain by Cambro-Ordovician sandstone. The Torridon Group has been divided (base to top) into the Diabaig, Applecross, Aultbea and Cailleach Head formations (Stewart 2002). The Diabaig Formation comprises breccia, conglomerate, siltstone and sandstone. Cobble breccia or conglomerate infill palaeo-valleys and are rich in vein-quartz clasts. The siltstones have been interpreted as lacustrine (Stewart 1988). The Diabaig Formation is absent in the Cape Wrath area in the north, occurs intermittently in Assynt and thickens to c. 200 m on Skye. The Applecross and Aultbea formations, two very similar sandstone formations, form the bulk of the Torridon Group, totalling c. 4-5 km in thickness. The contact with the underlying Diabaig Formation is sharp, locally erosional and may represent a disconformity (Kinnaird et al. 2006). The Applecross Formation consists predominantly of coarse to very coarse red sandstone in beds 0.1-5 m thick. Pebble conglomerate and siltstone-mudstone beds occur locally. The Aultbea Formation comprises mainly fine- to mediumgrained sandstone and minor mudstone. Flat bedding, planar cross-bedding and trough cross-bedding are common in both formations (Nicholson 1993; Stewart 2002). Soft-sediment deformation structures are locally abundant (Selley et al. 1963; Owen 1995) and affect beds up to 5 m thick. Palaeocurrents are broadly eastward, but vary between NE and SE (Williams 1969a; Nicholson 1993; Williams 2001). The pebble fraction of the Applecross Formation consists mostly of vein quartz or gneiss but also contains up to 30% of ‘exotic’ clasts (quartzfuchsite schist, orthoquartzite, metaquartzite, microgranite, rhyolite, chert and red jasper) that cannot be linked to underlying rock units (Peach et al. 1907; Gracie & Stewart 1967; Williams 1969e). The formations are interpreted as alluvial braid plain deposits (Nicholson 1993; Stewart 2002), although Williams (2001) suggested an alluvial ‘mega fan’ environment.

Sedimentology of the Morar Group in the Northern Highlands

Well-preserved sedimentary structures are only rarely present in Moine rocks but are observed in several low-strain zones within the Altnaharra Formation in the Ben Hee area (Cheer 2006) and Glen Cassley (British Geological Survey (BGS), unpublished data) (Fig. 3). The structure of the Ben Hee-Glen Cassley area is dominated by kilometre-scale, west-facing and west-verging folds, alternating with regional-scale ductile thrusts, all developed under greenschist- to lower amphibolite-facies metamorphism, presumed to be of Scandian (Silurian) age (Cheer 2006). The folds (Fig. 3) trend roughly north- south, have shallow plunging axes and are near-cylindrical over many kilometres. The folds have highly sheared gently east-dipping long limbs, some of which are ductile thrusts (e.g. Ben Hope and Achness thrusts; Fig. 3). Between these thrusts are low-strain zones, commonly in the steep to vertical short limbs of the large-scale folds. Such limbs are up to 500 m thick and many kilometres wide across strike (cross-sections in Fig. 3). In these zones, strata have been rotated c. 800 -100[degrees] to subvertical attitudes, but nevertheless show undeformed sedimentary structures (Fig. 4). A modest fabric is locally present in rare semipelite or gritty units (Fig. 4a), but most exposures of psammite show a complete lack of any tectonic fabric. Low-strain zones with well-preserved sedimentary structures were found in two thrust sheets, above and below the Ben Hope Thrust (Fig. 3) commonly on large, glacially polished outcrops.

Constraints on stratigraphical thickness

The stratigraphical thickness of the Altnaharra Formation is well constrained between River Cassley and Cam nam Bo Maola [NC 462 095] (Fig. 3). Here, a 3 km long section exposes subvertical strata that strike NNW-ESE and consistently young to the west; this equates to 3 km of stratigraphical thickness (Fig. 3, cross section B-B’). To the west in the AHt na Faile [NC 432 080], the strata are folded over c. 500 m section distance. West from this, another 3 km long section of steep to moderate-dipping strata stretches west as far as Beinn an Eoin Bheag [NC 375 055], possibly adding another 2-3 km to the total stratigraphical thickness (cross-section B-B’ in Fig. 3). Neither the stratigraphical top nor base of the Altnaharra Formation occurs in this section but it is clear that the formation has a stratigraphical thickness of at least 3 km, and possibly more than 5 km.

Lithology

The dominant lithology of the Altnaharra Formation is a fine to coarse quartzo-feldspathic psammite (grain size varies between 0.5 and 3 mm) with rare layers of pelite and semipelite. The psammites contain 80-90% quartz, 3-8% alkali-feldspar and <4% plagioclase and biotite, with accessory opaques (derived from thin-section study). Gritty beds (Fig. 4a) are common, particularly in the lower parts of the sequence (e.g. east of Cam nam Bo Maola), with clasts up to 30 mm. Pebbles are mainly well-rounded (vein?) quartz with subordinate clasts of feldspar and rarer quartzofeldspathic gneiss and/or granitoid. Semipelite layers become more common (c. 5% of section) at higher levels in the west near Beinn an Eoin Bheag, defining an overall fining upward trend. Overall, the formation is exceptionally uniform and no distinct marker beds have been found.

Sedimentary structures

Observed sedimentary structures include isolated channels, nested channels, planar and trough cross-bedding, planar stratification and abundant soft-sediment deformation structures (Figs 4 and 5). Trough cross-bed sets, typically 0.1 -1 m deep, infill channels up to several metres deep and 3-15 m wide. The sets occur as nested stacked units (co-sets) up to 8 m thick (Fig. 4bd). Gravel-pebble lags occur in the bases of larger channels whereas heavy mineral bands (up to 10 mm thick) are locally preserved along the bases of smaller channels. Planar crossstratification (Fig. 4c) makes up as much as one-third of exposures and occurs as sets and co-sets that are laterally truncated by overlying channels or display migration toward channel thalwegs away from channel margins. Planar crossbedded co-sets range in thickness from 0.1 to 1 m. Both planar and trough cross-bedding locally display a fining upward trend along foresets; coarser grain sizes (in places pebbly) define bottomsets whereas topsets are characterized by finer grain sizes (fine sand to semi pelite). Soft-sediment deformation affected c. 20-30% of the well-preserved outcrops (Fig. 4e and f). Features include dewatering ‘pipes’ 0.2-2.5 m in height, typically confined to single beds, and oversteepened to overturned crossbedding that can affect cross- stratified strata up to 5 m thick; in almost all cases, overturning is towards the east or NE; that is, in the sediment transport direction. Slumping is developed locally and typically on decimetre scales but can incorporate up to 10 m of stratigraphy, involving single beds or groups of beds.

Most bed contacts are erosional and vertical trends are difficult to ascertain. However, it is apparent that the channelized, trough cross-bedded units tend to display a decrease in grain size (at least as coarse-tail fining) and scale of co-sets upward from an erosive base (Fig. 4d). Large outcrop surfaces reveal that the planar cross-bed sets display lateral migration directions that are typically at high angles to the scooped-shaped bounding surfaces of the channels. Planar stratification and/or finer-grained facies occupy a stratigraphical position either in the topmost portions of the flared margins of the channels or along the tops of planar cross- bed co-sets.

Channel orientations typically trend approximately east-west and the infilling trough cross-strata indicate that overall sediment transport was generally to the east to NNE (Fig. 4b-d). Only few channels are exposed in 3D; however, planar cross-bedded strata at Cam Mor (Glencassley area) consistently indicate unidirectional palaeocurrents to the east or NE (Fig. 4c).

Sedimentological interpretation

The Altnaharra Formation consists of metamorphosed sandstones and pebbly sandstones exhibiting a wide range of structures formed by bedload traction. The grain-size distribution combined with the decimetre- to metre-scale trough and planar cross-bed sets imply high flow velocities in channels deep enough to permit development of metre-scale bedforms (i.e. dunes). High flow velocities are also indicated by (1) sigmoidal shaped foresets and the asymptotic toes of metre-scale trough cross-bed sets, (2) the presence of flat stratification in coarse to pebbly grain sizes (upper flow regime plane beds), (3) the syndepositional shearing that steepened or overturned metre-scale foresets and (4) the overall coarse grain size of the psammites. The channel-fills commonly display a sequence of sedimentary structures that decrease in scale, and fine upwards, indicating progressive channel abandonment. The arrangement of channelized beds in nested and stacked units several metres thick, which display fining upwards in both grain size and bedform scale, is a characteristic facies of braided fluvial environments (Miall 1985, 1992; Collinson 1996). A fluvial setting is also supported by the unidirectional palaeocurrents displayed by the planar and trough cross-beds, which consistently show east to NNE-directed sediment transport. Occurrences of planar and trough cross-bedding oriented at high angles to the channel margins are interpreted as laterally accreting bars. By contrast, bedforms showing migration parallel to the trough and channel axes are interpreted as downstreammigrating bars (e.g. Smith 1970; Cant & Walker 1978; Miall 1992). These facies are arranged in 20-50 m thick packages in which coarser-grained, channelized and trough cross-bedded units dominate the lower portions, with planar stratified and relatively finer-grained units (including thin semipelitic intervals) characterizing the upper parts. We interpret these decametrescale patterns as recording lateral variation between channelized braided fluves and bars, interfluve areas and intermittent more widespread sheetfloods.

The Altnaharra Formation lacks well-developed vertical grainsize and bedding thickness trends. This absence is typical for pre-land- plant braid plain settings (Schunrm 1968; Cotter 1978). In contrast, metre- to decametre-scale ‘cyclicity’ is what characterizes parasequence development of shoreline and marine shelf settings whether tide, storm or fluvial dominated (e.g. Walker & Plint 1992; Johnson & Baldwin 1996; Reading & Collinson 1996). In summary, the evidence indicates that the Altnaharra Formation records fluvial deposition in a high-energy braided fluvial setting.

Geochemistry

Whole-rock and stream sediment geochemical data have been used to argue for and against a correlation between the Moine and ‘Torridonian’ rocks (Kennedy 1951; Stone et al. 1999; Stewart 2002). However, no modem whole-rock analyses are available for the Altnaharra Formation in the study area. A series of samples from the Altnaharra Formation were collected for whole-rock geochemical analysis as part of this project The samples come from a section from Glen Cassley to Cam nam Bo Maola and represent c. 2 km of the stratigraphical succession (Fig. 3, cross-section B-B’); the data are presented in Table 1 and Figure 6 and discussed in more detail below. The samples plot as arkosic to sub-arkosic, with an overall trend to more subarkosic (mature) compositions higher in the stratigraphy (Fig. 6). The samples indicate a mineralogical immaturity in accordance with the textural immaturity and the suggested fluvial depositional setting. Overall there is remarkably little geochemical variation between the samples, attesting to the lithological monotony of the Altnaharra Formation.

Discussion

Correlating the A ‘Mhoine and Applecross-Aultbea formations

The Torridon Group was deposited in a fluvial environment characterized by braided rivers flowing from the west (Stewart 2002). As the Moine Thrust has an overall WNW-directed transport direction, restoration of the thrust would place the Morar Group ‘downstream’ from the Torridon Group, so that a correlation between the two groups is a distinct possibility. We suggest a correlation between the Applecross-Aultbea Formation (Torridon Group) and the Altnaharra Formation (Morar Group) as represented in the area north of Glen Oykel (Figs 1 and 3).

General position, lithology and sedimentology. The Morar and Torridon groups both unconformably overlie Archaean- Palaeoproterozoic basement of comparable age (Holdsworth et al. 1994; Friend et al. 2002; Stewart 2002; Kinny et al. 2005). Both sequences have a basal conglomeratic facies, together with siltstone- pelite and sandstone-psammite, which occurs intermittently above the unconformity. Both sequences are several kilometres thick (>3 to 5 km) and are typified by monotonous, coarse to very coarse (meta)sandstone with local pebble lags and some finer-grained sandstone and minor muddy-pelitic layers becoming more frequent at higher stratigraphical levels. The two sequences lack marker horizons of different lithologies.

Sedimentary structures in both the Applecross-Aultbea and Altnaharra formations are comparable in style, scale and frequency: metre-thick cross-stratified beds, unidirectional trough cross- bedding and nested channels 1-5 m deep. Soft-sediment deformation structures are common and include metre-scale contorted bedding, oversteepened to overturned cross-beds, small-scale sag-structures involving heavy mineral bands, and these structures are typically confined to single beds (this study; Selley et al. 1963; Selley 1969; Williams 1970, 2001; Nicholson 1993; Owen 1995; Stewart 2002, for the Torridon Group). Both deposits are fluviatile, and were rapidly deposited in a highenergy, braid plain environment (this study; Williams 1969a, 2001; Nicholson 1993; Stewart 2002).

Age of deposition. The youngest U-Pb age on detrital zircons from the Altnaharra Formation, dated at 1032 +- 32 Ma (Friend et al. 2003), is within error of the youngest detrital zircon ages of 1060 +-18 Ma and 1046 +-26 Ma in the Applecross and Aultbea formations, respectively (Rainbird et al. 2001). Rb/Sr ages from mudstone from the Applecross Formation are 994 +-48 Ma and 977 +- 38 Ma, and have been interpreted to date diagenesis (Turnbull et al. 1996). The Glenfinnan and Loch EiI groups are intruded by the c. 870 Ma West Highland Granite Gneiss Suite (Friend et al. 1997; Millar 1999) and this date is generally taken as the minimum age of Moine Supergroup. Thus, deposition of the Torridon Group occurred after c. 1050 Ma and probably around c. 980 Ma, whereas deposition of the Morar Group occurred sometime between c. 1030 and c. 870Ma, so that the age constraints overlap. It is likely that both the Applecross-Aultbea and the A ‘Mhoine Formations were deposited between c. 1000 and 950 Ma.

Detrital zircon ages. Detrital zircon data from the Torridon, Morar and adjacent groups, obtained by Rainbird et al. (2001), Friend et al. (2003) and Cawood et al. (2004), are summarized in Figure 7. The detrital zircon age pattern of the Altnaharra Formation (Friend et al. 2003) shows a sharply defined dominant cluster at c. 1650Ma, minor clusters at c. 1800Ma and c. 1400Ma, and a few analyses between 1400 and 1000 Ma. Additionally, c. 8% of analysed grains were Archaean in age. The detrital zircon age patterns of the Loch Eil and Glenfinnan groups (Friend et al. 2003; Cawood et al. 2004) differ considerably from the Morar Group pattern: most zircons are younger than c. 1500 Ma and there is no clearly defined 1650 Ma cluster.

Similarly, the detrital zircon age patterns of the Applecross Formation and the Aultbea Formation both show a sharply defined cluster at c. 1650 Ma and a smaller cluster at c. 1800 Ma (Rainbird et al. 2001). Some Archaean grains (25% and 15%, respectively) occur, as well as a small, broad cluster between c. 1200 and 1000 Ma. In contrast, the underlying Stoer Group shows a dominant Late Archaean detrital zircon population (Fig. 7f), with a peak at 2900- 2700 Ma, and the youngest zircon has an age of c. 1740 Ma (Rainbird et al. 2001).

Overall, the detrital zircon age patterns of the Morar and Torridon groups have more in common (including the same dominant peak at c. 1650Ma) with each other than with the sequences with which they are normally associated.

Geochemistry. Stone et al. (1999) noted broad geochemical similarities between the ‘Torridonian’ and the Moine Supergroup, based on regional stream sediment geochemistry (Institute of Geological Sciences 1982). In contrast, Stewart (2002) noted that boron concentrations in stream sediment over the Moine are 2-5 times lower than in the Torridon Group, and discounted a correlation on that basis. However, boron is a highly mobile element and can be depleted by a factor of two or more during medium-grade metamorphism (Moran et al. 1992). Therefore, significant boron depletion can be expected in the Morar metasediments and boron (and other fluid- mobile elements) should not be used to compare and contrast unmetamorphosed and metamorphosed rocks. Virtually all other elements analysed for stream sediment geochemistry in Sutherland show very similar values for the Torridon and Morar groups (Institute of Geological Sciences 1982).

The analysed whole-rock geochemistry of the Altnaharra Formation psammites is compared in Table 1 with analyses from sandstones of the Applecross-Aultbea formations (Stewart & Donnellan 1992; Van de Kamp & Leake 1997) and Sleat Group (Stewart 1991). Generally, the arkosic Sleat Group rocks contain more Al, Fe, Ca and Na, with concomitantly less Si; on the log (Fe^sub 2^O^sub 3^/K^sub 20^)/log (SiO^sub 2^/Al^sub 2^O^sub 3^) plot (Herron 1988) they plot close to the wacke field (Fig. 6); these rocks are clearly less mature than the Morar and Torridon Group rocks. The Altnaharra Formation and Torridon Group rocks are fairly similar, and plot in overlapping fields (Fig. 6). The range of SiO^sub 2^, TiO^sub 2^, Al^sub 2^O^sub 3^, Fe, MgO and K^sub 2^O within the Morar samples overlaps with those from the Torridon, similarly so for most trace elements. Calcium and strontium are both higher in the Altnaharra Formation (Table 1); this would suggest a higher component of calcic over sodic and potassic feldspar in the detritus; alternatively, albitization may have selectively affected the Torridon Group. The Torridon Group sandstones are all arkosic, whereas some Altnaharra Formation rocks are subarkosic. Also, the Chemical Index of Alteration (CIA = Al^sub 2^O^sub 3^/(Al^sub 2^O^sub 3^ + CaO+ Na^sub 2^O+ K^sub 2^O); Nesbitt & Young 1982) is somewhat lower for the Altnaharra Formation. Overall, the small differences between the Altnaharra Formation and the Applecross and Aultbea formation rocks can be well explained by better sorting and slightly higher maturity of the Altnaharra Formation, as this tends to lower the CIA by removing more clay from the sand (Nesbitt et al. 1996). Better sorting is expected if the Morar Group was deposited farther downstream from the Torridon Group. There are no significant differences between the geochemistry of the Applecross, Aultbea and Altnaharra formations, and geochemistry can certainly not be used to discount a correlation (see Stewart 2002). Overall, we conclude that the Applecross-Aultbea formations and the Altnaharra Formation can be correlated, without much lateral variation, across the Moine Thrust. Detritus provenance

Williams (2001) and Stewart (2002) suggested the Lewisian Gneiss Complex as a source area for the Applecross Formation, whereas Van de Kamp & Leake (1997) and Rainbird et al. (2001) suggested the Grenville Orogen as the main source. For the Moine Supergroup as a whole, Friend et al. (2003) and Cawood et al. (2004, 2007) suggested a more general Laurentian provenance. All suggested source areas lie to the west, consistent with the dominant palaeocurrent directions.

Broadly speaking, the Laurentia craton (and the Lewisian Gneiss) is dominated by Late Archaean rocks with subordinate c. 2100-1800Ma Palaeoproterozoic orogenic belts (e.g. Torngat, Trans-Hudson; see Fig. 8). Only some of these belts produced juvenile crust, others mainly reworked Archaean crust (Hoffman 1988), so that Palaeoproterozoic felsic igneous rocks are relatively rare. A large belt of juvenile crust dated between 1800 and 1700Ma (Yavapai and Mazatzal Province and KetilidianMakkovik belt) lies south and SW of the Archaean craton. In Labrador, abundant Mesoproterozoic anorogenic magmatism occurred between 1460 and 1420Ma and between 1350 and 1290 Ma (Nain Plutonic Suite). Most of these latter plutons lie north of the Grenville Front.

The Grenville Orogen in North America comprises several Meso- Palaeoproterozoic terranes that were amalgamated, reworked and exhumed between 1100 and 950Ma. About 50% of the currently exposed rocks of the Eastern Grenville Orogen are of igneous origin (Fig. 8), but only a small proportion are synGrenville granitoids (Gower et al. 1991; Rivers 1997; Gower & Krogh 2002). The bulk of the felsic igneous rocks are older and include Pre-Labradorian 1780- 1710Ma granitic orthogneisses and large volumes of Labradorian (1710- 1600 Ma) calc-alkaline igneous rocks in the northern part of the orogen. The latter include the 600 km long, c. 1650Ma Trans- Labrador batholith. In contrast, the southern part of the Grenville Orogen is dominated by Pinwarian granitoid intrusions (1520-1460Ma) and the Adirondian anorthosite-mafic-granite suite (1200-1130Ma). In eastern Canada, the main Grenville orogenic activity spanned the period between 1080 and 970 Ma, and synto post-orogenic granitic plutonism occurred between c. 1025 and 920 Ma. Ar/Ar cooling ages suggest significant uplift and erosion between 980 and 930 Ma (Haggart et al. 1993).

From the aerial extent of magmatic rocks, it is possible to predict in a qualitative manner what age ranges of detrital minerals would be expected either from the Grenville Orogen or from Laurentia outside the Grenville Belt, bearing in mind that high-level parts of the orogen are missing, having already been unroofed. Two such ‘predictive’ detrital mineral age patterns for the Grenville Orogen and the Laurentian cratonic interior are shown in Figure 7g and h.

The dominant c. 1650Ma cluster of the Applecross, Aultbea and Morar zircons can be confidently linked to the TransLabradorian batholith (see also Rainbird et al. 2001), on the northern side within the Grenville Orogen (Figs 7 and 8). The c. 1200-1000Ma cluster is derived from the Grenville Orogen itself. The c. 1800Ma cluster in the Torridon Group is most probably derived from the Ketilidian-Makkovik belt; it is highly unlikely that these zircons are derived from the ‘Laxfordian’ c. 1850Ma intrusions within the Lewisian Complex, as these intrusions are minor (<2%) in aerial extent compared with Archaean gneisses.

Also noteworthy is the scarcity (and absence in the case of the Applecross Formation) of zircons dated between 1600 and 1250 Ma. Igneous rocks in this age bracket are common in Labrador and Greenland in the foreland of the Grenville Orogen, but are rare in the northern part of the orogen itself. This would suggest that during deposition of the basins, the immediate foreland of the Grenville Orogen was covered and not available as a source area (Figs 7 and 8; see also Cawood et al. 2004, 2007). An exception is the small c. 1450Ma cluster in the Altnaharra Formation. If this cluster is significant it may correlate with the Pinware terrane (see also Cawood et al. 2004), and relate to occasional southward stream capture across a drainage divide in the Grenville Orogen.

The relatively minor, variable component of Archaean age argues against the Lewisian Gneiss or the Laurentian Craton as the main source. Nevertheless, the 8-25% Archaean grains must have come from the cratonic interior, as little or no Archaean material appears to be incorporated into the Grenville Orogen (Figs 7 and 8). In the lowermost Applecross Formation, some Archaean grains may be of Lewisian origin because the high (100-600m) palaeorelief means that Lewisian hills remained exposed while the first few hundred metres of Applecross sandstones were being deposited.

Overall, the Grenville Orogen is likely to have provided the bulk of the detritus of the Torridon and Morar Group, with a small input from the Makkovikian-Ketilidian and the Archaean Laurentian foreland. A general Laurentian source outside the Grenville Orogen, let alone a Lewisian Gneiss source, is not compatible with the detrital zircon age patterns (see also Cawood et al. 2004, 2007).

Basin interpretation

If the Morar and Torridon groups can be correlated and were deposited in the same basin, what was its setting? Previously, the Torridon and Morar groups have been interpreted as separate rift basins (Stewart 1982; Soper et al. 1998; Williams 2001; Stewart 2002, and references therein) but this interpretation is problematic.

Rift sedimentation and subsidence is primarily controlled by episodic faulting and basin subsidence. This results in alternating periods of quiescence and progradation of coarse clastic sediment into finer-grained and commonly lacustrine basinal settings. The net result is a stratigraphical framework replete with lateral and vertical facies changes; for example, the Tertiary extensional basins in the Death Valley region, USA (Wright & Troxel 1999), the Suez Rift (Jackson et al. 2006) and the Jurassic basins of the North Sea (Underhill 1998). In addition, volcanic, evaporitic and lacustrine deposits are common in rift basins. The Torridon and Morar groups exhibit none of these features. In fact, few, if any rift basins (particularly half-grabens) are characterized by >5 km vertically and >200km horizontally similar siliciclastic sediments (see also Nicholson 1993; Prave 1999; Cawood et al. 2004).

The detrital zircons show a distal, rather than proximal source. Continental rift basins typically have a proximal source, with commonly a large age difference between the youngest age of detritus and the onset of sedimentation (e.g. Stoer Group, Fig. 7f; Rainbird et al. 2001).

The Minch Fault (Fig. 1) has been invoked as a large-scale basin- bounding fault to the suggested Torridon rift basin (Williams 19696, 2001; Stewart 2002). Williams (19696, 2001) argued that the Torridon Group consisted of a series of alluvial megafans with their apices near the Minch Fault. Nicholson (1993), however, showed that the palaeocurrents do not support such fans. Moreover, the pebble content and the detrital zircon data do not match a detrital source in the Outer Hebrides (composed mainly of Archaean rocks). Also, there is no evidence of syndepositional fault activity; nowhere along the basal Torridon unconformity, well exposed over several hundred kilometres, is there evidence for syn-Applecross-Formation extensional faults.

The abundance of soft-sediment deformation in the Torridon Group has also been used to argue for frequent seismic activity and hence rifting. However, convolute bedding can be generated without seismicity by bed liquidization during rapid deposition of water- saturated sand in combination with a high water table (Selley et al. 1963; Selley 1969; Williams 1970, 2001; Nicholson 1993; Owen 1995). The lack of terrestrial vegetation during the Neoproterozoic would have exacerbated such conditions (e.g. Eriksson et al. 2001).

Nicholson (1993) and Cawood et al. (2004) suggested an intracratonic basin setting for the Torridon Group and Moine Supergroup, respectively. Intracratonic basins, however, are typically long-lived and slowly subsiding, are sensitive to environmental change and hence contain significant vertical facies changes, the Neoproterozoic to Palaeozoic Taoudeni Basin (West Africa) being a good example (Bertrand-Sarfati et al. 1991). A major problem, therefore, is to provide sufficient accommodation space for rapid deposition of a 5 km of laterally and vertically uniform siliciclastic succession.

Foreland basin setting. In contrast, there is a growing body of work that suggests that the Torridon Group was deposited as a non- marine molasse, in a foreland basin setting (Rainbird et al. 2001; Kinnaird et al. 2006). This model explains the distal provenance of the detrital zircons analysed from the Torridon Group. Deposition in a trunk river system in an axial, orogenparallel foreland basin setting best explains the features observed in both the Applecross- Aultbea and Altnaharra formations. The envisaged basin would be analogous to the modem-day Ganges basin, in that the preserved part of the basin would have been deposited in a braided river system flowing in front of, and generally parallel to, the orogen. The position of the GrenviUe Orogen to the south (present-day orientation), and the east- to NNE-directed palaeocurrents fit such a palaeogeography. An orogen-parallel foreland-basin setting is further supported by the following factors. (1) Age. Accepting that the age of deposition of the Applecross-Aultbea and Altnaharra formations is broadly equivalent, then the constraint for their deposition at between c. 1000 and 950 Ma overlaps with the last stages of the GrenviUe Orogeny. The intrusion of late orogenic granites, decompression metamorphism and metamorphic cooling in the GrenviUe Orogen all occurred between 1025 and 950 Ma (e.g. Gower et al. 1991; Haggart et al. 1993; Cox et al. 2002; Gower & Krogh 2002); such processes are generally accompanied by overall unroofing of the orogen, resulting in the formation of an approximately coeval foreland basin.

(2) Sedimentology. The Applecross-Aultbea and Altnaharra Formations comprise a c. 5 km thick sequence of alluvial-fluvial siliciclastic rocks deposited in a wide braid plain system. The basin was characterized by large, relatively deep rivers, high peak runoff and rapid deposition. Rapid deposition of a thick sequence requires rapid, sustained subsidence. These are features typical for molasse-type foreland basin (e.g. Pfiffner 1986). Foreland basins typically have subsidence rates 3-10 times faster than most rift basins, and can achieve 2-3 km of subsidence in less than 10Ma (e.g. Homewood et al. 1986); this provides a good explanation for the deposition of a great thickness of highenergy clastic sediments over a wide area.

(3) Provenance. The detrital zircon age patterns suggest that the Grenville Orogen was the main source of detritus; this detritus comprises both synorogenic and pre-orogenic material uplifted in the orogen (e.g. the c. 1650Ma cluster). Such a combination of synorogenic and pre-orogenic material is common in foreland basins, as shown by Hercynian and Alpine detrital micas in the North Alpine foreland basin (Von Eynatten & Wijbrans 2003). Orogen-parallel foreland basins have a forebulge, so that part of the drainage and hence a minor component of the detritus originate from the cratonic interior. This is consistent with the variable amount of c. 1800 Ma and Archaean grains present in the successions; this detritus most probably originated from the area north of the Grenville Orogen (e.g. Ketilidian and cratonic parts of Laurentia).

Many foreland basins show an evolution from deep-water clastic sedimentation (‘flysch’) during early orogenesis, followed by shallow marine and finally non-marine (‘molasse’) sedimentation (e.g. Pfiffher 1986; Miall 1995). The earliest sediments are commonly caught up in foreland-propagating thrust systems and are uplifted and eroded, thus having a low preservation potential. The younger and shallower ‘molasse’ sediment onlap far onto the foreland and parts of this ‘molasse’ system may thus escape subsequent thrusting, uplift and erosion. It is this part of the foreland basin system that is preserved in the Morar and Torridon groups. The 1080- 1050 Ma Flinton Group in Eastern Ontario, Canada, may represent earlier, more varied and partially marine foreland basin rocks caught up in the Grenville Orogen (Moore & Thompson 1980), but similar rocks appear not to be present in Scotland.

Displacement on the Moine Thrust

The Moine Thrust separates the Torridon and Morar groups, and the displacement along this major structure must be taken into account for their correlation or otherwise. The total displacement of the Moine Thrust Zone as a whole is generally assumed to be >100 km (Strachan et al. 2002). However, Torridon Group rocks occur in the highest thrust sheets, so that the Torridon basin must have extended considerably farther east with respect to the foreland.

Consequently, it is only the Moine Thrust itself and its associated mylonites that truly separate the Torridon and Morar groups. The total displacement taken up by these structures is difficult to constrain. It is more than 20 km, as evidenced by the down-faulted block of Moine Mylonites at Faraid Head (Peach et al. 1907) and must be sufficient to have emplaced medium-grade metamorphic rocks over unmetamorphosed rocks. A reasonable estimate is probably c. 100 km.

The broadly eastward palaeocurrents in the sediments are approximately co-axial to the WNW-directed thrust transport direction and there is no evidence for major (>100 km) strikeslip movement along the Moine Thrust or its trace. Therefore the simplest original relationship between the Torridon and Morar Group is that the latter was deposited some 100-200 km downstream from the former. Such a distance is in fact very small for braided river systems in sedimentary basins, which can easily measure >1000 km along their axis of flow, as seen in both ancient and modern examples (e.g. Smith & Rogers 1999; Bridge 2003).

Regional Implications

We have shown that the Applecross-Aultbea formations and the Altnaharra Formation are correlative parts of the same sequence, simply repeated by the Moine Thrust. This invalidates the formal distinction between the Torridon Group and the Morar Group and hence the Moine Supergroup (see Holdsworth et al. 1994; Trewin 2002), and implies that the Proterozoic stratigraphical framework in Scotland needs to be revised. One solution is to include the Torridon Group into the Moine Supergroup, but abandon the term Morar Group. Alternatively, the term ‘Moine Supergroup’ could be abandoned, and the Altnaharra Formation included in the Torridon Group, as the latter is better exposed.

Correlations farther south in the Skye, Morar and Knoydart areas are also likely, but their details require further study, partially because the stratigraphy of both the ‘Torridonian’ and the Morar Group in these areas is more diverse (Ramsay & Spring 1962; Sutton & Watson 1964; Holdsworth et al. 1994; Stewart 2002). Sutton & Watson (1964) proposed the correlation Sleat Group = lower Morar Group and Torridon Group = upper Morar Group (Fig. 2); this proposal deserves renewed attention. Furthermore, it is unclear whether the Altnaharra Formation correlates southward with the Upper or the Lower Morar Psammite Formation in Morar, as the intervening ground has never been mapped in detail. It is prudent to await the outcomes of further studies before erecting a revised stratigraphy in Scotland, while noting that the current framework is unsatisfactory.

In addition, the ‘Hebridean Terrane’ and the ‘Northern Highlands Terrane’ (Bluck et al. 1992) share much of their pre- and post- Caledonian evolution and should be regarded as parautochthonous, and not as exotic to each other (Bluck et al. 1997; Oliver 2002). The Moine Thrust is better regarded as the Caledonian orogenic front, rather than a significant terrane boundary.

Conclusions

The Altnaharra Formation (Morar Group) in the Northern Highlands is characterized by c. 5 km of uniform psammite, devoid of marker horizons, and was deposited in a high-energy fluvial environment characterized by braided rivers flowing from the west. The Altnaharra Formation and the Applecross-Aultbea formations (Torridon Group) are similar in terms of their age of deposition, sedimentology, stratigraphical position, geochemistry, detrital zircon age pattern, age constraints and overall sediment transport direction. The detrital zircon distributions in both groups show that they share a similar, distal source, namely parts of the Grenville Orogen, the final stages of which overlap the age of deposition. It is therefore concluded that the Applecross-Aultbea and the Altnaharra formations are direct correlatives and formed part of an axial trunk fluvial system flowing in front of the Grenville Orogen, forming an orogen-parallel foreland basin. This reinterpretation implies that the currently accepted Proterozoic stratigraphical framework for the Scottish Highlands is in need of revision. The two groups should be regarded as parautochthonous.

BGS-UCAC PhD funding (Project 2K02E020) for D.C. is gratefully acknowledged. M. Smith, G. Leslie, P. Stone, J. Mendum, K. Good- enough and C. Thomas are thanked for comments and discussions. C. Friend, A. Harris and two anonymous reviewers are thanked for their detailed and incisive reviews. This article is published with the permission of the Executive Director of the British Geological Survey.

References

BARR, D., HOLDSWORTH, R.E. & ROBERTS, A.M. 1986. Caledonian ductile thrusting in a Precambrian metamorphic complex: the Moine of NW Scotland. Geological Society of America Bulletin, 97, 754-764.

BERTRAND-SARFATI, I., MOUSSINE-POUCHKINE, A., AFFATON, P., TROMPETTE, R. & BELLION, Y. 1991. Cover sequences of the West African craton. In: DALLMEYER, R.D. & LECORCHE, J.P. (eds) The West African Orogens and Circum-Atlantic Correlatives. Springer, Berlin, 65-84.

BLUCK, B.J., GIBBONS, W. & INGHAM, J.K. 1992. Terranes. In: COPE, J.C.W., INGHAM, J.K. & RAWSON, P.F. (eds) Atlas of Palaeogeography and Lithofacies. Geological Society, London, Memoirs, 13, 3-4.

BLUCK, B.J., DEMPSTER, T.J. & ROGERS, G. 1997. Allochthonous metamorphic blocks on the Hebridean passive margin, Scotland. Journal of the Geological Society, London, 154, 921-924.

BRIDGE, J.S. 2003. Rivers and Floodplains: Forms. Processes, and Sedimentary Record. Blackwell, Oxford.

BRITISH GEOLOGICAL SURVEY 1989a. Great Glen, Sheet 57[degrees]- 06[degrees], 1:250000 Series, Solid Geology. British Geological Survey, Keyworth, Nottingham.

BRITISH GEOLOGICAL SURVEY 1989b. Sutherland, Sheet 58[degrees]- 06[degrees], 1:250000 Series, Solid Geology. British Geological Survey, Keyworth, Nottingham.

BUTLER, R.W.H. 1997. Late Proterozoic rift faults and basement- cover relationships within the Ben More thrust sheet, NW Scotland. Journal of the Geological Society, London, 154, 761-764.

BUTLER, R.W.H. & COWARD, M.P. 1984. Geological constraints, structural evolution and deep geology of the NW Scottish Caledonides. Tectonics, 3, 347-365.

CANT, D.J. & WALKER, R.G. 1978. Fluvial processes and facies sequences in the sandy braided South Saskatchewan River, Canada. Sedimentology, 25, 625-648. CAWOOD, P.A., NEMCHIN, A.A., STRACHAN, R., KINNY, P. & LOEWY, S. 2004. Laurentian provenance and an intracratonic tectonic setting for the Moine Supergroup, Scotland, constrained by detrital zircons from the Loch Eil and Glen Urquhart successions. Journal of the Geological Society, London, 161, 861- 874.

CAWOOD, P.A., NEMCHIN, A.A., STRACHAN, R., PRAVE, A. & KRABBENDAM, M. 2007. Sedimentary basin and detrital zircon record along East Laurentia and Baltica during assembly and breakup of Rodinia. Journal of the Geological Society, London, 164, 257-275.

CHEER, D.A. 2006. The sedimentary, structural and igneous geology of the Moine Thrust Zone and Moine Nappe in the Ben Hee area, NW Scotland. PhD thesis, University of St. Andrews.

COLLINSON, J.D. 1996. Alluvial sediments. In: READING, H.G. (ed.) Sedimentary Environments: Processes, Facies and Stratigraphy. Blackwell Science, Oxford, 37-82.

COTTER, E. 1978. The evolution of sedimentary style, with special reference to the central Appalachian Paleozoic. In: MIALL, A.D. (ed.) Fluvial Sedimentology. Canadian Society of Petroleum Geology, Memoirs, 5, 361-383.

COX, R.A., INDARES, A. & DUNNING, G.R. 2002. Temperature-time paths in the high-P Manicouagan imbricate zone, eastern Grenville Province; evidence for two metamorphic events. Precambrian Research, 117, 225-250.

ELLIOTT, D. & JOHNSON, M.R.W. 1980. Structural evolution in the northern part of the Moine thrust belt, NW Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, 71, 69-96.

ERIKSSON, P.G., MARTINS-NETO, M.A. & NELSON, D.R. ET AL. 2001. An introduction to Precambrian basins; their characteristics and genesis. Sedimentary Geology, 141-142, 1-35.

FRIEND, C.R.L., KINNY, P.D., ROGERS, G., STRACHAN, R.A. & PATTERSON, B.A. 1997. U-Pb zircon geochronological evidence for Neoproterozoic events in the Glenfinnan Group (Moine Supergroup): the formation of the Ardgour granite gneiss, north-west Scotland. Contributions to Mineralogy and Petrology, 128, 101-113.

FRIEND, C.R.L., KINNY, P.D. & STRACHAN, R.A. 2002. Basement inliers in Moine Supergroup of NW Scotland: What do they represent? In: Tectonic Studies Group, Highlands Workshop, University of St Andrews, 5.

FRIEND, C.R.L., STRACHAN, R.A., KINNY, P.D. & WATT, G.R. 2003. Provenance of the Moine Supergroup of NW Scotland; evidence from geochronology of detrital and inherited zircons from (meta)sedimentary rocks, granites and migmatites. Journal of the Geological Society, London, 160, 247-257.

FROUDE, D.O., IRELAND, T.R., KINNY, P.D., WILLIAMS, I.S., COMPSTON, W., WILLIAMS, I.R. & MYERS, J.S. 1983. Ion microprobe identification of 4,100-4,200-Myr-old terrestrial zircons. Nature, 304, 616-618.

GIBBONS, W. & HARRIS, A.L. 1994. A Revised Correlation of Precambrian Rocks in the British Isles. Geological Society, London, Special Reports, 22.

GLENDINNING, N.R.W. 1988. Sedimentary structures and sequences within a late Proterozoic tidal shelf deposit; the upper Morar Psammite Formation of northwestern Scotland. In: WINCHESTER, J.A. (ed.) Later Proterozoic Stratigraphy of the Northern Atlantic Regions. Blackie, Glasgow, 17-31.

GOWER, C.F. & KROGH, T.E. 2002. A U-Pb geochronological review of the Proterozoic history of the eastern Grenville Province. Canadian Journal of Earth Sciences, 39, 795-829.

GOWER, CF., HEAMAN, L.M., LOVERIDGE, W.D., SCHAERER, U. & TUCKER, R.D. 1991. Grenvillian magmatism in the eastern Grenville Province, Canada. Precambrian Research, 51, 315-336.

GRACIE, A.J. & STEWART, A.D. 1967. Torridonian sediments at Enard Bay, Rossshire. Scottish Journal of Geology, 3, 181-194.

HAGGART, M.J., JAMIESON, R.A., REYNOLDS, P.H., KROGH, T.E., BEAUMONT, C. & CULSHAW, N.G. 1993. Last gasp of the Grenville Orogeny; thermochronology of the Grenville Front tectonic zone near Killarney, Ontario. Journal of Geology, 101, 575-589.

HERRON, M.M. 1988. Geochemical classification of terrigenous sands and shales from core or log data. Journal of Sedimentary Petrology, 58, 820-829.

HOFFMAN, P.F. 1988. United Plates of America, the birth of a Craton: Early Proterozoic assembly and growth of Laurentia. Annual Review of Earth and Planetary Sciences, 16, 543-603.

HOLDSWORTH, R.E., HARRIS, A.L. & ROBERTS, A.M. 1987. The stratigraphy, structure and regional significance of the Moine Rocks of Mull, Argyllshire, W. Scotland. Geological Journal, 22, 83-107.

HOLDSWORTH, R.E., STRACHAN, R.A. & HARRIS, A.L. 1994. The Moine Supergroup. In: GIBBONS, W. & HARRIS, A.L. (eds) A Revised Correlation of Precambrian Rocks in the British Isles. Geological Society, London, Special Reports, 22, 23-32.

HOLDSWORTH, R.E., STRACHAN, R. & ALSOP, G.I. 2001. Solid geology of the Tongue district: Memoir for 1:50000 Geological Sheet 114 E (Scotland). British Geological Survey, Keyworth, Nottingham.

HOMEWOOD, P., ALLEN, P.A. & WILLIAMS, G.D. 1986. Dynamics of the Molasse Basin of western Switzerland. In: ALLEN, P.A. & HOMEWOOD, P. (eds) Foreland Basins. International Association of Sedimentologists, Special Publications, 8, 199-217.

INSTITUTE OF GEOLOGICAL SCIENCES 1982. Regional Geochemical Atlas: Sutherland. Institute of Geological Sciences, London.

JACKSON, C.A.L., GAWTHORPE, R.L., LEPPARD, CW. & SHARP, I.R. 2006. Riftinitiation development of normal fault blocks; insights from the Hammam Faraun fault block, Suez Rift, Egypt. Journal of the Geological Society, London, 163, 165-183.

JOHNSON, H.D. & BALDWIN, C.T. 1996. Shallow siliciclastic seas. In: READING, H.G. (ed.) Sedimentary Environments: Processes, Facies, Environments. Blackwell Science, Oxford, 232-280.

JOHNSTONE, G.S., SMITH, D.I. & HARRIS, A.L. 1969. The Moinian Assemblage of Scotland. In: KAY, M. (ed.) North Atlantic Geology and Continental Drift: a Symposium. American Association of Petroleum Geologists, Memoirs, 12, 159-180.

KENNEDY, W.Q. 1951. Sedimentary differentiation as a factor in the Moine-Torridonian correlation. Geological Magazine, 88, 257- 266.

KINNAIRD, T.C, PRAVE, A.R., KIRKLAND, CL., HORSTWOOD, M., PARRISH, R.R. & BATCHELOR, R.A. 2006. The late Mesoproterozoic- early Neoproterozoic evolution of northwest Scotland: the Torridonian revisited. Journal of the Geological Society, London, 164, 541-551.

KINNY, P.D., FRIEND, C.R.L. & LOVE, G.J. 2005. Proposal for a terrane-based nomenclature for the Lewisian Gneiss Complex of NW Scotland. Journal of the Geological Society, London, 162, 175-186.

KRABBENDAM, M. & LESLIE, A.G. 2004. Lateral ramps and thrust terminations: an example from the Moine Thrust Zone, NW Scotland. Journal of the Geological Society, London, 161, 551-554.

MENDUM, J.R. 1976. A strain study of the Strathan Conglomerate, North Scotland. Scottish Journal of Geology, 12, 135-146.

MIALL, A.D. 1985. Architectural-element analysis: a new method of facies analysis applied to fluvial deposits. Earth-Science Reviews, 22, 261-308.

MIALL, A.D. 1992. Alluvial deposits. In: WALKER, R.G. & JAMES, N.P. (eds) Facies Models, Response to Sea Level Change. GeoText I. Geological Association of Canada, St. John’s, Newfoundland, 119- 142.

MIALL, A.D. 1995. Collision-related foreland basins. In: BUSBY, C.J. & INGERSOILL, R.V. (eds) Tectonics of Sedimentary Basins. Blackwell, Oxford, 393-424.

MILLAR, I.L. 1999. Neoproterozoic extensional basic magmatism associated with the West Highland granite gneiss in the Moine Supergroup of NW Scotland. Journal of the Geological Society, London, 156, 1153-1162.

MOORE, J.M.J. & THOMPSON, P.H. 1980. The Flinton Group; a late Precambrian metasedimentary succession in the GrenviUe Province of eastern Ontario. Canadian Journal of Earth Sciences, 17, 1685-1707.

MORAN, A.E., SISSON, V.B. & LEEMAN, W.P. 1992. Boron depletion during progressive metamorphism; implications for subduction processes. Earth and Planetary Science Letters, 111, 331-349.

NELSON, D.R. 2001. An assessment of the determination of depositional ages for Precambrian clastic sedimentary rocks by U-Pb dating of detrital zircons. Sedimentary Geology, 141-142, 37-60.

NESBITT, H.W. & YOUNG, G.M. 1982. Early Proterozoic climates and plate motions inferred from major element chemistry of lutites. Nature, 299, 715-717.

NESBITT, H.W., YOUNG, G.M., MCLENNAN, S.M. & KEAYS, R.R. 1996. Effects of chemical weathering and sorting on the petrogenesis of siliciclastic sediments, with implications for provenance studies. Journal of Geology, 104, 525-542.

NICHOLSON, P.G. 1993. A basin reappraisal of the Proterozoic Torridon Group, northwest Scotland. In: FROSTICK, L.E. & STEEL, R.J. (eds) Tectonic Controls and Signatures in Sedimentary Successions. International Association of Sedimentologists, Special Publications, 20, 183-202.

OLIVER, G.J.H. 2002. Chronology and terrane assembly, new and old controversies. In: TREWIN, N.H. (ed.) The Geology of Scotland. Geological Society, London, 201-211.

OWEN, G. 1995. Soft-sediment deformation in upper Proterozoic Torridonian sandstones (Applecross Formation) at Torridon, northwest Scotland. Journal of Sedimentary Research, A65, 495-504.

PEACH, B.N. & HORNE, J. 1930. Chapters on the Geology of Scotland. Oxford University Press, Oxford.

PEACH, B.N., HORNE, J., GUNN, W., CLOUGH, CT., HINXMAN, L.W. & TEALL, JJ.H. 1907. The geological structure of the North-West Highlands of Scotland. Memoirs of the Geological Survey of Great Britain.

PEACH, B.N., HORNE, J., WOODWARD, H.B., CLOUGH, C.T., HARKER, A. & WEBB, C.B. 1910. The geology of Glenelg, Lochalsh and South East part of Skye-Explanation of one-inch map 71. Memoir of the Geological Survey-Scotland.

PEACH, B.N., HORNE, J., HINXMAN, L.W., CRAMPTON, M.B., ANDERSON, E.M. & CARRUTHERS, R.G. 1913. The geology of Central Ross-shire- Explanation of Sheet 82. Memoir of the Geological Survey-Scotland.

PFIFFNER, O.A. 1986. Evolution of the North Alpine foreland basin in the Central Alps. In: ALLEN, P.A. & HOMEWOOD, P. (eds) Foreland Basins. International Association of Sedimentologists, Special Publications, 8, 219-228. PRAVE, A.R. 1999. The Neoproterozoic Dalradian Supergroup of Scotland: an alternative hypothesis. Geological Magazine, 136, 609-617.

RAINBIRD, R.H., HAMILTON, M.A. & YOUNG, G.M. 2001. Detrital zircon geochronology and provenance of the Torridonian, NW Scotland. Journal of the Geological Society, London, 158, 15-27.

RAMSAY, J.G. 1957. Moine-Lewisian relations at Glenelg, Inverness- shire. Quarterly Journal of the Geological Society of London, 113, 487-523.

RAMSAY, J.G. & SPRING, J. 1962. Moine stratigraphy in the Western Highlands of Scotland. Proceedings of the Geologists’ Association, 73, 295-326.

READING, H.G. & COLLINSON, J.D. 1996. Clastic coasts. In: READING, H.G. (ed.) Sedimentary Environments: Processes, Facies and Stratigraphy. Blackwell Science, Oxford, 154-231.

RIVERS, T. 1997. Lithotectonic elements of the Grenville Province; review and tectonic implications. Precambrian Research, 86, 117-154.

ROBERTS, A.M., STRACHAN, R.A., HARRIS, A.L., BARR, D. & HOLDSWORTH, R.E. 1987. The Sgurr Beag nappe: a reassessment of the stratigraphy and structure of the northern Highland Moine. Geological Society of America Bulletin, 98, 497-506.

SCHUMM, S.A. 1968. Speculations concerning paleohydrologic controls of terrestrial sedimentation. Geological Society of America Bulletin, 79, 1573-1588.

SELLEY, R.C. 1969. Torridonian alluvium and quicksands. Scottish Journal of Geology, 5, 328-346.

SELLEY, R.C., SHEARMAN, D.H., SUTTON, J. & WATSON, J.V. 1963. Some underwater disturbances in the Torridonian of Skye and Raasay. Geological Magazine, 100, 224-243.

SMITH, N.D. 1970. The braided stream depositional environment: comparison of the Platte River with some Silurian clastic rocks, north central Appalachians. Geological Society of America Bulletin, 81, 2993-3014.

SMITH, N.D. & ROGERS, J.D. (EDS) 1999. Fluvial Sedimentology IV. International Association of Sedimentologists, Special Publications, 28.

SOPER, N.J., HARRIS, A.L. & STRACHAN, R.A. 1998. Tectonostratigraphy of the Moine Supergroup; a synthesis. Journal of the Geological Society, London, 155, 13-24.

STEWART, A.D. 1972. Pre-Cambrian landscapes in northwest Scotland. Geological Journal, 8, 111-124.

STEWART, A.D. 1982. Late Proterozoic rifting in NW Scotland: the genesis of the ‘Torridonian’. Journal of the Geological Society, London, 139, 413-420.

STEWART, A.D. 1988. The Sleat and Torridon Groups. In: Winchester, J.A. (ed.) Later Proterozoic Stratigraphy of the Northern Atlantic Regions. Blackie, Glasgow, 104-112.

STEWART, A.D. 1991. Geochemistry, provenance and palaeoclimate of the Sleat and Torridon Groups in Skye. Scottish Journal of Geology, 27, 81-95.

STEWART, A.D. (ED.) 2002. The Later Proterozoic Torridonian Rocks of Scotland: their Sedimentology, Geochemistry and Origin. Geological Society, London, Memoirs, 24.

STEWART, A.D. & DONNELLAN, N.C.B. 1992. Geochemistry and provenance of red sandstones in the Upper Proterozoic Torridon Group of Scotland. Scottish Journal of Geology, 28, 143-153.

STONE, P., PLANT, J.A., MENDUM, J.R. & GREEN, P.M. 1999. A regional geochemical assessment of some terrane relationships in the British Caledonides. Scottish Journal of Geology, 35, 145-156.

STRACHAN, R.A. 1986. Shallow-marine sedimentation in the Proterozoic Moine Succession, Northern Scotland. Precambrian Research, 32, 17-33.

STRACHAN, R., SMITH, M., HARRIS, A.L. & FETTES, D.J. 2002. The Northern Highland and Grampian terranes. In: TREWIN, N.H. (ed.) The Geology of Scotland. Geological Society, London, 81-147.

SUTTON, J. & WATSON, J. 1964. Some aspects of Torridonian stratigraphy in Skye. Proceedings of the Geologists’ Association, 75, 251-289.

TANNER, P.W.G. & BLUCK, B.J. 1999. Current controversies in the Caledonides. Journal of the Geological Society, London, 156, 1137- 1141.

TREWIN, N.H. (ED.) 2002. The Geology of Scotland, Geological Society, London.

TURNBULL, M.J.M., WHITEHOUSE, M.J. & MOORBATH, S. 1996. New isotopic age determinations for the Torridonian, NW Scotland. Journal of the Geological Society, London, 153, 955-964.

UNDERHILL, J.R. 1998. Jurassic. In: GLENNIE, K.W. (ed.) Petroleum Geology of the North Sea: Basic Concepts and Recent Advances. Blackwell, Oxford, 245-293.

VAN DE KAMP, P.C. & LEAKE, B.E. 1997. Mineralogy, geochemistry, provenance and sodium metasomatism of Torridonian rift basin clastic rocks, NW Scotland. Scottish Journal of Geology, 33, 105-124.

VON EYNATTEN, H. & WIJBRANS, J.R. 2003. Precise tracing of exhumation and provenance using ^sup 40^Ar/^sup 39^Ar geochronology of detrital white mica; the example of the Central Alps. In: MCCANN, T. & SAINTOT, A. (eds) Tracing Tectonic Deformation Using the Sediment Record. Geological Society, London, Special Publications, 208, 289-305.

WALKER, R.G. & PLINT, A.G. 1992. Wave- and storm-dominated shallow marine systems. In: WALKER, R.G. & JAMES, N.P. (eds) Facies Models, Response to Sea Level Change. GeoText 1. Geological Association of Canada, St. John’s, Newfoundland, 219-238.

WILLIAMS, G.E. 1969a. Characteristics and origin of a Precambrian pediment. Journal of Geology, 77, 183-207.

WILLIAMS, G.E. 1969b. Petrography and origin of pebbles from Torridonian strata (late Precambrian), northwest Scotland. In: KAY, M. (ed.) North Atlantic Geology and Continental Drift: a Symposium. American Association of Petroleum Geologists, Memoirs, 12, 609-629.

WILLIAMS, G.E. 1970. Origin of disturbed bedding in Torridon Group sandstones. Scottish Journal of Geology, 6, 409-411.

WILLIAMS, G.E. 2001. Neoproterozoic (Torridonian) alluvial fan succession, northwest Scotland, and its tectonic setting and provenance. Geological Magazine, 138, 161-184.

WINCHESTER, J.A. 1988. Later Proterozoic environments and tectonic evolution in the northern Atlantic lands. In: WINCHESTER, J.A. (ed.) Later Proterozo