1. Introduction
In his paper ‘Tectonic evolution of the Caledonian Orogeny in Scotland: a review based on the timing of magmatism, metamorphism and deformation’, Searle (Reference Searle2022) argues that the deformation and metamorphism of the Scottish Highlands is continuously diachronous from the early Ordovician (Grampian) in the Dalradian of the Grampian Highlands to the mid-Silurian (Scandian) in the Moine of the Northwest Highlands. Necessarily, he disputes substantial offset along the Great Glen Fault (GGF). He compares the Scottish Caledonides with the Himalayas as a diachronous, tapering, thrust wedge from a collisional suture (cf. Indus Suture Zone and Highland Boundary Fault) to a frontal thrust system (cf. Himalayan Boundary Thrust and Moine Thrust Zone).
We take the quite different view that the Grampian and Scandian orogenies were discrete events, not the beginning and end members of a continuous NW-migrating wedge. The Grampian Orogeny involved major deformation and metamorphism across the Grampian and Northwest Highlands to the Moine Thrust Zone, whereas the Scandian Orogeny affected only the Northwest Highlands. The early Ordovician Grampian Orogeny was generated by the collision and obduction of an oceanic arc-ophiolite with and onto the Laurentian continent. The mid-Silurian Scandian Orogeny was caused by continental collision between Laurentia and Baltica. Late Silurian – early Devonian sinistral slip of at least 500 km along the GGF – Walls Boundary Fault brought the Moine of the Northwest Highlands from a distant location to the north in the Scandian collision zone, against the Dalradian, which had not been involved in the Scandian collision. We see no evidence for progressive northwestward propagation of a thrust wedge from the early Ordovician to the mid-Silurian. We see little resemblance between the Highlands and the Himalayas in scale, structure and evolution The Highland Orogen is much narrower and shorter, was probably much lower with less molasse, is less cylindroidal and has a much more complicated history involving Neo-Proterozoic deformation and metamorphism. The Himalayan wedge was driven by the Tibetan thick crustal ram whereas no such ram existed in Scotland.
A major point of difference is that Searle (Reference Searle2022) bases his model on data from Scotland alone; no justification is given for ignoring along-strike data. Any tectonic model is limited by the evidence within the subset of data considered. We argue below that Searle’s (Reference Searle2022) analysis of the Caledonian Orogen in Scotland is not supported by evidence in Scotland and also ignores data from Ireland and Newfoundland.
We are not suggesting that ‘the science is settled’ or that it is supported by a ‘consensus’, which are dangerous terms that should never be used in science. Science is never ‘settled’ but evolves by the accumulation of facts, new ideas and vigorous open discussion and debate. Consensus is irrelevant in science; only truth matters. The Highlands are complex and imperfectly understood; we should be avidly mapping, and collecting and synthesizing a mass of new data, and we welcome Searle’s contribution as a basis for future discussion and debate.
2. Relevant data and main arguments against the Searle model
2.a. The Great Glen Fault
The GGF has a minimum sinistral displacement of c. 170 km because there are no matching points across this exposed length. Kennedy (Reference Kennedy1946) suggested an offset of c. 96 km by matching the Foyers and Strontian Granites; these granites do not match and the displacement must be at least 160 km (Winchester, Reference Winchester1973). To the southwest, it passes through Mull and it may continue as the Cabot Fault in Newfoundland. Strike-slip faults of this length and displacement can only end at triple junctions, such as the San Andreas Fault terminated by the Mendocino and Rivera triple junctions, and may vary in displacement as the triple junctions evolve. The GGF is one of several sinistral faults and it is possible that, to the north, the GGF breaks into several faults but, nevertheless, the gross offset of at least 170 km must continue a long way north. The Walls Boundary Fault (WBF) in Shetland is almost linear, has no matches across it for 70 km and is a likely candidate for the continuation of the GGF. That neither fault can be seen, obviously, in seismic sections does not prove their absence: strike-slip faults, being vertical, are difficult to recognize in seismic section. Also, it would be astonishing if the exposed lengths of the faults just happened to be coincident with their displacements. At Rosemarkie on the Black Isle, a long coastal outcrop of the GGF exhibits a very strong sinistral ductile deformation over a width of c. 100 m, suggesting a very large displacement. The Moine has moved south by at least 170 km relative to the Dalradian; therefore, the Moine and Dalradian of the Highlands were not adjacent to each other during Lower Palaeozoic times, even with a low estimate of 170 km displacement. A single cross-sectional approach across them is inappropriate. The principal reason for suggesting that the GGF–WBF has a very large displacement is the absence of a Scandian imprint in the Dalradian and, therefore, the need for it not to be in the jaws of the Laurentia–Baltica collision that caused the Scandian event (Dewey & Strachan, Reference Dewey and Strachan2003). The bare minimum displacement to restore the Moine into the Scandian collision zone is c. 400 km.
2.b. Sea level
There is no discussion of the fact that the mid-Ordovician (Darriwilian, ∼465 Ma) to late Silurian (Ludfordian, ∼425 Ma) sediments of the Midland Valley terrane in both Scotland and Ireland are predominantly marine (e.g. Midland Valley (Cameron & Stephenson, Reference Cameron and Stephenson1985; Stone et al. Reference Stone, McMillan, Floyd, Barnes and Phillips2012); Pomeroy and Lissbelaw (Simon, Reference Simon1986); Charlestown (Parkes, Reference Parkes1993); South Mayo and north Connemara (Ryan & Dewey, Reference Ryan, Dewey, Brown and Ryan2011; Graham, Reference Graham2021)) and may form part of the same basin system (Williams & Harper, Reference Williams and Harper1988). In Searle’s (Reference Searle2022) model the Midland Valley terrane occupies the hanging wall in a similar manner to Tibet, but Tibet is 5 km above sea level. Also, the Midland Valley was at sea level in the immediate hanging wall of the supposed long-lived suture.
2.c. The Southern Uplands
The Southern Uplands, from c. 453 (M. clingani) to 424 (M. lundgreni), developed as a clear and perfect example of a subduction–accretion prism above a north-dipping subduction zone (seen in British Institutions Reflection Profiling Syndicate (BIRPS) section, Klemperer, Reference Klemperer1989), indicated by short-lived packages of greywackes above dark oceanic shales that become progressively younger southwards (Dewey et al. Reference Dewey, Dalziel, Reavy and Strachan2012), terminating with a late Silurian (Erian) collision across the Iapetus suture along the Carlisle–Newcastle line, the line separating the Cambro-Ordovician Pacific and Atlantic faunal realms. Therefore, the north-dipping Southern Uplands subduction zone conflicts with Searle’s south-dipping subduction in the same place and time. The early Ordovician Ballantrae ophiolite, with its metamorphic sole, and the scraps of ophiolite along the Highland Boundary Fault are likely parts of the Midland Valley oceanic arc slab that was obducted during the early Ordovician to generate the Grampian Orogeny (Stewart et al. Reference Stewart, Baxter and Ague2017), after which there is no evidence of southward subduction or any obvious place to site a subduction zone, especially a long-lived one with no arc.
2.d. The Caledonides of western Ireland
The Caledonides of western Ireland are critical to the debate, The Dalradian in Connemara was deformed and metamorphosed (Grampian Orogeny) in north-facing fold nappes from 477 to 467 Ma below sea level and an obducting oceanic arc, ophiolite and fore-arc basin (South Mayo Trough), followed by long-wavelength mid-Ordovician folding (the Mayoian), and there was no later Scandian event (e.g. Ryan & Dewey, Reference Ryan, Dewey, Brown and Ryan2011; Chew & Strachan, Reference Chew, Strachan, Chew, Gasser and Corfu2014). Also, the Dalradian sits unconformably on Grenville basement; there is no Moine in Ireland.
2.e. Post-Grampian subduction flip
The evidence for subduction flip in the late Ordovician following Grampian arc–continent collision is not fully debated by Searle (Reference Searle2022). North-verging subduction beneath the Laurentian margin during late Ordovician to Silurian times is supported by the nature of the volcano–sedimentary complexes in Ireland, deep seismic data indicating north-dipping fabrics in the middle and lower crust and the Southern Uplands – Longford Down accretionary complex. A suite of Darriwilian calc-alkali granitoids with xenocrystic zircons postdate arc–continent collision in Tyrone (Cooper et al. Reference Cooper, Crowley and Rushton2008). The Charlestown Group contains calc-alkali eruptives associated with an arc founded on continental crust (Herrington et al. Reference Herrington, Hollis, Cooper, Stobbs, Tapster, Rushton, McConnell and Jeffries2018). The Telychian lavas, shallow diorite intrusives and volcaniclastic turbidites of north Connemara rest unconformably upon both the older oceanic arc rocks of the Lough Nafooey Group and exhumed Dalradian migmatites and are related to north-dipping subduction (Williams et al. Reference Williams, O’Connor and Menuge1992; Mohr, Reference Mohr1998). Although there are no known equivalents to these complexes in Scotland, Telychian bentonites occur in both the Midlands Valley Terrane and the Southern Uplands Terrane whose source may have been within 500 km (Batchelor et al. Reference Batchelor, Harper and Anderson2003). These were erupted at a time when there was no granite magmatism in Scotland (Oliver et al. Reference Oliver, Wilde and Wan2008; Miles et al. Reference Miles, Woodcock and Hawkesworth2016). Klemperer (Reference Klemperer1989) reports BIRPS deep seismic reflection data showing prominent north-dipping reflectors running from the mid-crust to the Moho which coincide with the inferred position of the Silurian Iapetus suture running from the Solway Firth to the Shannon Estuary (Soper et al. Reference Soper, England, Snyder and Ryan1992). Stratigraphic and structural evidence from the Ordovician to Silurian deposits of the Southern Uplands (Floyd, Reference Floyd2000), Longford Down massif (Anderson, Reference Anderson2000) and South Connemara (Ryan & Dewey, Reference Ryan and Dewey2004) show all the hallmarks of a south-facing accretionary prism. Pre-turbidite black shales and rare mafic slices indicate that this accretionary complex was founded on oceanic lithosphere (Lambert et al. Reference Lambert, Holland and Leggett1981; Dewey et al. Reference Dewey, Dalziel, Reavy and Strachan2012) and formed at an active Laurentian continental margin (Waldron et al. Reference Waldron, Floyd, Simonetti and Heaman2008).
2.f. Red Indian – Southern Uplands line
The site of subduction reversal was roughly along the Southern Uplands Fault during D. artus times and passes along-strike to South Connemara (Ryan & Dewey, Reference Ryan and Dewey2004) where an oceanic basalt–chert sequence lies beneath a turbidite trench fill, thrusted beneath slices of basalt interpreted as accreted seamounts. Further along-strike, in Newfoundland, this subduction reversal zone is seen as the Red Indian Line (Zagorevski & McNicoll, Reference Zagorevski and McNicoll2011). In Newfoundland, the early Ordovician arc-ophiolite is thrusted westwards over the Fleur-de-Lys metamorphic complex, a narrow-zone equivalent of the Dalradian but deformed slightly later in the Taconic Orogeny. West of the Fleur-de-Lys, the arc-ophiolite slab was obducted across the continental margin onto the continental shelf to form the highest thrust unit of the Humber Allochthon in a foreland basin (Dewey & Casey, Reference Dewey and Casey2021). Silurian arc volcanics and shallow marine sequences rest unconformably across the Taconic zone. The key point is that, during and following the Taconic–Grampian Orogeny, the Orogen was at, near or below sea level (Dewey, Reference Dewey2005). After the Orogeny, southward subduction ceased, northward subduction was established and the Grampian Orogen rose to provide most of the sediment in the turbidite belt south of the Red Indian – Southern Uplands Fault Line. Hence, north of the Red Indian – Southern Uplands Line the Grampian was now in the upper plate enjoying only minor deformation. There was no continuous link to the Scandian of the Moine, which occurred in the Scandian collision zone between Greenland and Norway. It is difficult to understand how a migrating continental thrust wedge could be driven by an area at or below sea level, in contrast to the Himalayas, which are ‘driven’ by the thick Tibetan crust.
2.g. Progression of granite ages
There is no northward progression of older to younger granite ages across the Highlands just as there is no northward progression of metamorphic ages. The Grampian older granites are late syntectonic (D3), roughly the same age as the Newer Gabbro Suite (470–480 Ma), and are S-type (low temperature, ilmenite-reduced, commonly with cordierite, garnet and sediment schlieren) lower slab crustal melts. A suite of late-Ordovician post-tectonic decompression S-type granites (Oliver et al. Reference Oliver, Wilde and Wan2008) are likely upper slab. Later silicic plutons of the Grampian and Northwest Highlands are mainly I-type (high-temperature, oxidized-magnetite), commonly with associated co-mingling mafic-intermediate suites (Oliver et al. Reference Oliver, Wilde and Wan2008). Archibald et al. (Reference Archibald, Murphy, Fowler, Strachan and Hidebrand2022) specifically rule out A-type magmatism. The Argyll (421 Ma) and Cairngorm (408 Ma) suites are post-tectonic to the Dalradian and are upper slab likely derived by slab break-off (Atherton & Ghani, Reference Atherton and Ghani2002; Miles et al. Reference Miles, Woodcock and Hawkesworth2016). The 430 Ma Newer Granites of the Northwest Highlands are I-type alkali plutons co-genetic with mafic intrusives, becoming unsaturated along the Moine Thrust Zone, and likely the result of crustal melting above a mafic underplate in the Baltica–Laurentia collision zone. Archibald et al. (Reference Archibald, Lauren, Macquarrie, Murphy, Strachan, McFarlane, Button, Larson and Dunlop2021) have shown that there is a fundamental change in granite chemistry across the GGF from high Ba, Sr, K and light rare earth to the northwest and low values to the southeast.
2.h. Gap in granite magmatism
There is an apparent gap in the granite magmatism in the Grampian Highlands terrane (Oliver et al. Reference Oliver, Wilde and Wan2008) between the pre-449 Ma (Katian) S-type granites and the post-432 Ma (Sheinwoodian) I-type granites, which is correctly shown in figure 9 of Searle (Reference Searle2022). His figure 4, however, shows granite magmatism ‘spanning’ this interval. Some plutons are not plotted in the correct position. The Strontian granite is plotted at ∼437 (age of the granite 418 ± 1 Ma (Paterson et al. Reference Paterson, Rogers, Stephens and Hinton1993) and of the granodiorite 425 ± 3 Ma (Rogers & Dunning, Reference Rogers and Dunning1991)) although it lies in the Northern Highland terrane, Strath Spey at ∼442 Ma (age 449 ± 6 Ma; Oliver et al. Reference Oliver, Wilde and Wan2008), Inzie at ∼445 Ma (age 453 ± 8 Ma; Oliver et al. Reference Oliver, Wilde and Wan2008) and Moy at ∼448 Ma (age 452 ± 3 Ma; Oliver et al. Reference Oliver, Wilde and Wan2008). Miles et al. (Reference Miles, Woodcock and Hawkesworth2016) argue, using area-weighted age spectra, for an apparent absence of granite plutonism in Scotland between ∼455 Ma and ∼425 Ma. A similar gap is apparent in the Irish data for the along-strike equivalents to the Grampian Highlands, Midland Valley and Southern Upland terranes (Cooper et al. Reference Cooper, Crowley and Rushton2008; Chew & Stillman, Reference Chew and Stillman2009; Kirkland et al. Reference Kirkland, Alsop, Daly, Whitehouse, Lam and Clark2013; Murphy et al. Reference Murphy, Nance, Gabler, Martell and Archibald2019; Archibald et al. Reference Archibald, Lauren, Macquarrie, Murphy, Strachan, McFarlane, Button, Larson and Dunlop2021) where no granites are recorded between 462.5 Ma and 437.0 Ma. This pattern of magmatism was a result of more complex processes than just radiogenic heating during continental convergence.
1. Introduction
In his paper ‘Tectonic evolution of the Caledonian Orogeny in Scotland: a review based on the timing of magmatism, metamorphism and deformation’, Searle (Reference Searle2022) argues that the deformation and metamorphism of the Scottish Highlands is continuously diachronous from the early Ordovician (Grampian) in the Dalradian of the Grampian Highlands to the mid-Silurian (Scandian) in the Moine of the Northwest Highlands. Necessarily, he disputes substantial offset along the Great Glen Fault (GGF). He compares the Scottish Caledonides with the Himalayas as a diachronous, tapering, thrust wedge from a collisional suture (cf. Indus Suture Zone and Highland Boundary Fault) to a frontal thrust system (cf. Himalayan Boundary Thrust and Moine Thrust Zone).
We take the quite different view that the Grampian and Scandian orogenies were discrete events, not the beginning and end members of a continuous NW-migrating wedge. The Grampian Orogeny involved major deformation and metamorphism across the Grampian and Northwest Highlands to the Moine Thrust Zone, whereas the Scandian Orogeny affected only the Northwest Highlands. The early Ordovician Grampian Orogeny was generated by the collision and obduction of an oceanic arc-ophiolite with and onto the Laurentian continent. The mid-Silurian Scandian Orogeny was caused by continental collision between Laurentia and Baltica. Late Silurian – early Devonian sinistral slip of at least 500 km along the GGF – Walls Boundary Fault brought the Moine of the Northwest Highlands from a distant location to the north in the Scandian collision zone, against the Dalradian, which had not been involved in the Scandian collision. We see no evidence for progressive northwestward propagation of a thrust wedge from the early Ordovician to the mid-Silurian. We see little resemblance between the Highlands and the Himalayas in scale, structure and evolution The Highland Orogen is much narrower and shorter, was probably much lower with less molasse, is less cylindroidal and has a much more complicated history involving Neo-Proterozoic deformation and metamorphism. The Himalayan wedge was driven by the Tibetan thick crustal ram whereas no such ram existed in Scotland.
A major point of difference is that Searle (Reference Searle2022) bases his model on data from Scotland alone; no justification is given for ignoring along-strike data. Any tectonic model is limited by the evidence within the subset of data considered. We argue below that Searle’s (Reference Searle2022) analysis of the Caledonian Orogen in Scotland is not supported by evidence in Scotland and also ignores data from Ireland and Newfoundland.
We are not suggesting that ‘the science is settled’ or that it is supported by a ‘consensus’, which are dangerous terms that should never be used in science. Science is never ‘settled’ but evolves by the accumulation of facts, new ideas and vigorous open discussion and debate. Consensus is irrelevant in science; only truth matters. The Highlands are complex and imperfectly understood; we should be avidly mapping, and collecting and synthesizing a mass of new data, and we welcome Searle’s contribution as a basis for future discussion and debate.
2. Relevant data and main arguments against the Searle model
2.a. The Great Glen Fault
The GGF has a minimum sinistral displacement of c. 170 km because there are no matching points across this exposed length. Kennedy (Reference Kennedy1946) suggested an offset of c. 96 km by matching the Foyers and Strontian Granites; these granites do not match and the displacement must be at least 160 km (Winchester, Reference Winchester1973). To the southwest, it passes through Mull and it may continue as the Cabot Fault in Newfoundland. Strike-slip faults of this length and displacement can only end at triple junctions, such as the San Andreas Fault terminated by the Mendocino and Rivera triple junctions, and may vary in displacement as the triple junctions evolve. The GGF is one of several sinistral faults and it is possible that, to the north, the GGF breaks into several faults but, nevertheless, the gross offset of at least 170 km must continue a long way north. The Walls Boundary Fault (WBF) in Shetland is almost linear, has no matches across it for 70 km and is a likely candidate for the continuation of the GGF. That neither fault can be seen, obviously, in seismic sections does not prove their absence: strike-slip faults, being vertical, are difficult to recognize in seismic section. Also, it would be astonishing if the exposed lengths of the faults just happened to be coincident with their displacements. At Rosemarkie on the Black Isle, a long coastal outcrop of the GGF exhibits a very strong sinistral ductile deformation over a width of c. 100 m, suggesting a very large displacement. The Moine has moved south by at least 170 km relative to the Dalradian; therefore, the Moine and Dalradian of the Highlands were not adjacent to each other during Lower Palaeozoic times, even with a low estimate of 170 km displacement. A single cross-sectional approach across them is inappropriate. The principal reason for suggesting that the GGF–WBF has a very large displacement is the absence of a Scandian imprint in the Dalradian and, therefore, the need for it not to be in the jaws of the Laurentia–Baltica collision that caused the Scandian event (Dewey & Strachan, Reference Dewey and Strachan2003). The bare minimum displacement to restore the Moine into the Scandian collision zone is c. 400 km.
2.b. Sea level
There is no discussion of the fact that the mid-Ordovician (Darriwilian, ∼465 Ma) to late Silurian (Ludfordian, ∼425 Ma) sediments of the Midland Valley terrane in both Scotland and Ireland are predominantly marine (e.g. Midland Valley (Cameron & Stephenson, Reference Cameron and Stephenson1985; Stone et al. Reference Stone, McMillan, Floyd, Barnes and Phillips2012); Pomeroy and Lissbelaw (Simon, Reference Simon1986); Charlestown (Parkes, Reference Parkes1993); South Mayo and north Connemara (Ryan & Dewey, Reference Ryan, Dewey, Brown and Ryan2011; Graham, Reference Graham2021)) and may form part of the same basin system (Williams & Harper, Reference Williams and Harper1988). In Searle’s (Reference Searle2022) model the Midland Valley terrane occupies the hanging wall in a similar manner to Tibet, but Tibet is 5 km above sea level. Also, the Midland Valley was at sea level in the immediate hanging wall of the supposed long-lived suture.
2.c. The Southern Uplands
The Southern Uplands, from c. 453 (M. clingani) to 424 (M. lundgreni), developed as a clear and perfect example of a subduction–accretion prism above a north-dipping subduction zone (seen in British Institutions Reflection Profiling Syndicate (BIRPS) section, Klemperer, Reference Klemperer1989), indicated by short-lived packages of greywackes above dark oceanic shales that become progressively younger southwards (Dewey et al. Reference Dewey, Dalziel, Reavy and Strachan2012), terminating with a late Silurian (Erian) collision across the Iapetus suture along the Carlisle–Newcastle line, the line separating the Cambro-Ordovician Pacific and Atlantic faunal realms. Therefore, the north-dipping Southern Uplands subduction zone conflicts with Searle’s south-dipping subduction in the same place and time. The early Ordovician Ballantrae ophiolite, with its metamorphic sole, and the scraps of ophiolite along the Highland Boundary Fault are likely parts of the Midland Valley oceanic arc slab that was obducted during the early Ordovician to generate the Grampian Orogeny (Stewart et al. Reference Stewart, Baxter and Ague2017), after which there is no evidence of southward subduction or any obvious place to site a subduction zone, especially a long-lived one with no arc.
2.d. The Caledonides of western Ireland
The Caledonides of western Ireland are critical to the debate, The Dalradian in Connemara was deformed and metamorphosed (Grampian Orogeny) in north-facing fold nappes from 477 to 467 Ma below sea level and an obducting oceanic arc, ophiolite and fore-arc basin (South Mayo Trough), followed by long-wavelength mid-Ordovician folding (the Mayoian), and there was no later Scandian event (e.g. Ryan & Dewey, Reference Ryan, Dewey, Brown and Ryan2011; Chew & Strachan, Reference Chew, Strachan, Chew, Gasser and Corfu2014). Also, the Dalradian sits unconformably on Grenville basement; there is no Moine in Ireland.
2.e. Post-Grampian subduction flip
The evidence for subduction flip in the late Ordovician following Grampian arc–continent collision is not fully debated by Searle (Reference Searle2022). North-verging subduction beneath the Laurentian margin during late Ordovician to Silurian times is supported by the nature of the volcano–sedimentary complexes in Ireland, deep seismic data indicating north-dipping fabrics in the middle and lower crust and the Southern Uplands – Longford Down accretionary complex. A suite of Darriwilian calc-alkali granitoids with xenocrystic zircons postdate arc–continent collision in Tyrone (Cooper et al. Reference Cooper, Crowley and Rushton2008). The Charlestown Group contains calc-alkali eruptives associated with an arc founded on continental crust (Herrington et al. Reference Herrington, Hollis, Cooper, Stobbs, Tapster, Rushton, McConnell and Jeffries2018). The Telychian lavas, shallow diorite intrusives and volcaniclastic turbidites of north Connemara rest unconformably upon both the older oceanic arc rocks of the Lough Nafooey Group and exhumed Dalradian migmatites and are related to north-dipping subduction (Williams et al. Reference Williams, O’Connor and Menuge1992; Mohr, Reference Mohr1998). Although there are no known equivalents to these complexes in Scotland, Telychian bentonites occur in both the Midlands Valley Terrane and the Southern Uplands Terrane whose source may have been within 500 km (Batchelor et al. Reference Batchelor, Harper and Anderson2003). These were erupted at a time when there was no granite magmatism in Scotland (Oliver et al. Reference Oliver, Wilde and Wan2008; Miles et al. Reference Miles, Woodcock and Hawkesworth2016). Klemperer (Reference Klemperer1989) reports BIRPS deep seismic reflection data showing prominent north-dipping reflectors running from the mid-crust to the Moho which coincide with the inferred position of the Silurian Iapetus suture running from the Solway Firth to the Shannon Estuary (Soper et al. Reference Soper, England, Snyder and Ryan1992). Stratigraphic and structural evidence from the Ordovician to Silurian deposits of the Southern Uplands (Floyd, Reference Floyd2000), Longford Down massif (Anderson, Reference Anderson2000) and South Connemara (Ryan & Dewey, Reference Ryan and Dewey2004) show all the hallmarks of a south-facing accretionary prism. Pre-turbidite black shales and rare mafic slices indicate that this accretionary complex was founded on oceanic lithosphere (Lambert et al. Reference Lambert, Holland and Leggett1981; Dewey et al. Reference Dewey, Dalziel, Reavy and Strachan2012) and formed at an active Laurentian continental margin (Waldron et al. Reference Waldron, Floyd, Simonetti and Heaman2008).
2.f. Red Indian – Southern Uplands line
The site of subduction reversal was roughly along the Southern Uplands Fault during D. artus times and passes along-strike to South Connemara (Ryan & Dewey, Reference Ryan and Dewey2004) where an oceanic basalt–chert sequence lies beneath a turbidite trench fill, thrusted beneath slices of basalt interpreted as accreted seamounts. Further along-strike, in Newfoundland, this subduction reversal zone is seen as the Red Indian Line (Zagorevski & McNicoll, Reference Zagorevski and McNicoll2011). In Newfoundland, the early Ordovician arc-ophiolite is thrusted westwards over the Fleur-de-Lys metamorphic complex, a narrow-zone equivalent of the Dalradian but deformed slightly later in the Taconic Orogeny. West of the Fleur-de-Lys, the arc-ophiolite slab was obducted across the continental margin onto the continental shelf to form the highest thrust unit of the Humber Allochthon in a foreland basin (Dewey & Casey, Reference Dewey and Casey2021). Silurian arc volcanics and shallow marine sequences rest unconformably across the Taconic zone. The key point is that, during and following the Taconic–Grampian Orogeny, the Orogen was at, near or below sea level (Dewey, Reference Dewey2005). After the Orogeny, southward subduction ceased, northward subduction was established and the Grampian Orogen rose to provide most of the sediment in the turbidite belt south of the Red Indian – Southern Uplands Fault Line. Hence, north of the Red Indian – Southern Uplands Line the Grampian was now in the upper plate enjoying only minor deformation. There was no continuous link to the Scandian of the Moine, which occurred in the Scandian collision zone between Greenland and Norway. It is difficult to understand how a migrating continental thrust wedge could be driven by an area at or below sea level, in contrast to the Himalayas, which are ‘driven’ by the thick Tibetan crust.
2.g. Progression of granite ages
There is no northward progression of older to younger granite ages across the Highlands just as there is no northward progression of metamorphic ages. The Grampian older granites are late syntectonic (D3), roughly the same age as the Newer Gabbro Suite (470–480 Ma), and are S-type (low temperature, ilmenite-reduced, commonly with cordierite, garnet and sediment schlieren) lower slab crustal melts. A suite of late-Ordovician post-tectonic decompression S-type granites (Oliver et al. Reference Oliver, Wilde and Wan2008) are likely upper slab. Later silicic plutons of the Grampian and Northwest Highlands are mainly I-type (high-temperature, oxidized-magnetite), commonly with associated co-mingling mafic-intermediate suites (Oliver et al. Reference Oliver, Wilde and Wan2008). Archibald et al. (Reference Archibald, Murphy, Fowler, Strachan and Hidebrand2022) specifically rule out A-type magmatism. The Argyll (421 Ma) and Cairngorm (408 Ma) suites are post-tectonic to the Dalradian and are upper slab likely derived by slab break-off (Atherton & Ghani, Reference Atherton and Ghani2002; Miles et al. Reference Miles, Woodcock and Hawkesworth2016). The 430 Ma Newer Granites of the Northwest Highlands are I-type alkali plutons co-genetic with mafic intrusives, becoming unsaturated along the Moine Thrust Zone, and likely the result of crustal melting above a mafic underplate in the Baltica–Laurentia collision zone. Archibald et al. (Reference Archibald, Lauren, Macquarrie, Murphy, Strachan, McFarlane, Button, Larson and Dunlop2021) have shown that there is a fundamental change in granite chemistry across the GGF from high Ba, Sr, K and light rare earth to the northwest and low values to the southeast.
2.h. Gap in granite magmatism
There is an apparent gap in the granite magmatism in the Grampian Highlands terrane (Oliver et al. Reference Oliver, Wilde and Wan2008) between the pre-449 Ma (Katian) S-type granites and the post-432 Ma (Sheinwoodian) I-type granites, which is correctly shown in figure 9 of Searle (Reference Searle2022). His figure 4, however, shows granite magmatism ‘spanning’ this interval. Some plutons are not plotted in the correct position. The Strontian granite is plotted at ∼437 (age of the granite 418 ± 1 Ma (Paterson et al. Reference Paterson, Rogers, Stephens and Hinton1993) and of the granodiorite 425 ± 3 Ma (Rogers & Dunning, Reference Rogers and Dunning1991)) although it lies in the Northern Highland terrane, Strath Spey at ∼442 Ma (age 449 ± 6 Ma; Oliver et al. Reference Oliver, Wilde and Wan2008), Inzie at ∼445 Ma (age 453 ± 8 Ma; Oliver et al. Reference Oliver, Wilde and Wan2008) and Moy at ∼448 Ma (age 452 ± 3 Ma; Oliver et al. Reference Oliver, Wilde and Wan2008). Miles et al. (Reference Miles, Woodcock and Hawkesworth2016) argue, using area-weighted age spectra, for an apparent absence of granite plutonism in Scotland between ∼455 Ma and ∼425 Ma. A similar gap is apparent in the Irish data for the along-strike equivalents to the Grampian Highlands, Midland Valley and Southern Upland terranes (Cooper et al. Reference Cooper, Crowley and Rushton2008; Chew & Stillman, Reference Chew and Stillman2009; Kirkland et al. Reference Kirkland, Alsop, Daly, Whitehouse, Lam and Clark2013; Murphy et al. Reference Murphy, Nance, Gabler, Martell and Archibald2019; Archibald et al. Reference Archibald, Lauren, Macquarrie, Murphy, Strachan, McFarlane, Button, Larson and Dunlop2021) where no granites are recorded between 462.5 Ma and 437.0 Ma. This pattern of magmatism was a result of more complex processes than just radiogenic heating during continental convergence.
Acknowledgements
We are indepted to all those geologists who have taught us so much over so many years.
Conflict of interest
The authors declare that there is no conflict of interest.