terrane (ter-rane’) A fault-bounded body of rock of regional extent, characterized by a geological history different from that of contiguous terranes. A terrane is generally considered to be a discrete allochthonous fragment of oceanic or continental material added to a craton at an active margin by accretion. A. G. I. Glossary of Geology. 3rd edition, 1987.

Offshore seismic reflection profiles crossing the Caledonian orogenic front (locally, ?Moine thrust) between Shetland and the Scottish Highlands show a singular coherent east-dipping reflection underlain by a highly reflective dipping zone in the middle crust extending down to the Moho discontinuity. This reflector pattern varies spatially with respect to the eastern edge of the Precambrian Lewisian foreland as well as to previously mapped locations of the Moine thrust. Southwest of Shetland, the reflector pattern coincides with the offshore projection of the Moine thrust, but further south, toward the Highlands, the reflector diverges to the west of the thrust and actually underlies autochthonous Lewisian foreland basement. Where this reflector pattern diverges from the Moine thrust, two interpretations are possible: the prominent reflector is part of a basement imbricate thrust duplex within the footwall (or ‘lower plate’) of the Moine thrust that merges with the thrust zone further north and/or it was originally a Proterozoic normal fault which, further north around Shetland, was reactivated or over-printed by the Moine thrust. The latter interpretation supports the importance of pre-existing crustal structure in controlling Caledonian compressional deformation.

The Lower Cambrian Leny Limestone at Callander in Scotland lies within a sequence of palecoloured grits and dark slates (here named the Keltie Water Grit Formation (KWGF)) which has lithological similarities with members of the Highland Border Complex (HBC) (Ordovician) seen elsewhere along the Highland Border. The Keltie Water Grit Formation has a transitional boundary with a grit-slate sequence of undoubted Dalradian parentage; as the ‘Leny Grits’ of previous workers include both the KWGF and part of the Dalradian sequence, this term is now rendered invalid. The entire sequence youngs upwards from the Dalradian to the top of the overlying Keltie Water Grit Formation, shares the same structural sequence and geometry, and has the same facing and vergence direction on the main cleavage. All field and petrographic data are consistent with a minimum age of post-early Cambrian for the Grampian event, the main orogenic event to affect the Dalradian. Examination of critical sections elsewhere across the Highland Border shows that there is an apparently consistent stratigraphical and structural relationship between the Highland Border Complex and the Dalradian which, as suggested by some previous workers, would require the Grampian event to be post-Arenig in age. However, we are faced with a so-far unresolved paradox that there are certain palaeontological and radiometric data which are in conflict with this conclusion, and support the alternative hypothesis that the Highland Border Complex docked with the Dalradian in post-Ordovician times.

The Caledonides of the west of Ireland provide a well-exposed and well-mapped example of an oblique collision zone. The east-northeast trending Deer Park and Achill Beg Fault system is a crustal scale ductile sinistral strike-slip duplex of late Ordovician age, imbricating late Precambrian granulite facies lower crustal rocks, near eclogite facies supracrustal rocks, up to amphibolite facies Dalradian metasedimentary rocks and greenschist facies Cambro-Ordovician rocks. This fault system is correlated with a pre-Devonian component of the Highland Boundary Fault system in southern Scotland. In the Clew Bay area, the high pressure-low temperature facies metamorphic rocks, in tectonic contact with greenschist facies Cambro-Ordovician rocks, are together interpreted as an accretionary prism complex related to northwestward directed subduction. Both of these are allocthonous terrains with respect to the Dalradian terrane to the north (North West Mayo). To the south, the Cambro-Ordovician rocks docked with a probable Dalradian block containing ultramafic intrusives (Deer Park Complex) during the late Ordovician. The Deer Park Complex and South Mayo Trough linked earlier, during the Arenig.

Silurian and Lower-Middle Devonian redbed successions sit unconformably on the metamorphic rocks. Deposition and deformation of these cover rocks was controlled by oblique strike-slip movements on the Leek Fault whose strike swings from west-northwest to north-northeast, following earlier basement trends, as it is traced eastwards from Clew Bay. The Leek Fault System may be correlated with the Leannan Fault of northwest Donegal, a splay of the Great Glen Fault system of central Scotland. East of Clew Bay, this sinistral shear generated local dilation on the more northerly trending bend of the Leek Fault. Lower and Middle Old Red Sandstone redbeds were developed here. The west-northwest trend of the Leek Fault in Clew Bay acted as a compressional bend during these sinistral movements and transpressional southwest directed thrusting developed in Silurian rocks. Post-Middle Old Red Sandstone pre-late Tournaisian dextral displacement on the Leek Fault reversed this pattern with transtension in Clew Bay allowing intrusion of small carbonated peridotite bodies into Silurian rocks and easterly directed thrusting of Middle Old Red Sandstone rocks east of the Bay on the transpressional north-south bend.

A tectonic model for the region is presented here. This model involves a northwestward directed subduction system, 150 to 750 km of Arenig sinistral strike slip movement, and eastwards insertion of the Connemara block with formation of the Ordovician South Mayo Trough as a pull-apart basin. Subsequently, a further 130 to 650 km eastward displacement of rocks took place south of the Deer Park Fault in later Ordovician times. The magnitudes of these estimates are directly proportional to an assumed maximum wavelength of 1500 km for promontories on the original Laurentian margin, and using the current juxtaposition of terranes, a minimum wavelength of 300 km is inferred.

Geological studies along the southern margin of the Midland Valley of Scotland, together with previously published work, suggest a strike-slip model for its evolution during Silurian and Devonian times. The data emphasize the ‘older Caledonian’ influence and persistent, albeit intermittent, transcurrent activity along the Southern Upland Fault which is the boundary between the Midland Valley and Southern Upland terranes. A comparison of the geological characteristics of the Silurian and Devonian rocks in the southern Midland Valley with characteristics from strike-slip basins elsewhere suggests that the Midland Valley rocks were deposited and locally deformed under sinistral strike-slip regimes with two main episodes of transtension — the first acting in Llandovery—Wenlock times and the second in Lower Old Red Sandstone times. A strike-slip model involving transpression may explain why there is a strong unconformity between the Llandovery—Wenlock and the Lower Old Red Sandstone strata in the Pentland Hills and Girvan inliers but only a marked change in sedimentary facies at the equivalent horizon in the inliers of the central Midland Valley. The Lower Old Red Sandstone was probably deposited in a transtensional regime, but the succession is complicated by the widespread Lower Old Red Sandstone calc-alkaline magmatic event. The mid- Devonian deformation is also variable in its effects and the more easterly trend of the en echelon fold axes relative to the Southern Upland Fault trend is indicative of sinistral transpression. Carboniferous and older rocks were affected by smaller dextral movements on the Southern Upland Fault and related faults which started during Namurian times when a change to an overall dextral regime took place.

Prominent aeromagnetic anomalies around Ballantrae in southwest Scotland have long been associated with extensive outcrops of basic/ultrabasic igneous rocks in the Midland Valley terrane. However, recent reprocessing of data from a 1978 detailed aeromagnetic survey has demonstrated excellent correlation between discrete aeromagnetic features and certain magnetic greywacke formations within the early Caradoc Tappins Group. This heterogeneous unit forms the oldest and most northwesterly tectonostratigraphic tract in the Southern Uplands terrane and is characterized by relatively high detrital magnetite(and serpentinite) content and elevated values for Cr and Ni. This suggests original close sedimentary links with an ophiolitic terrane not unlike the presently adjacent Ballantrae Complex (or its extensive along-strike correlatives), a consideration which has important implications for terrane reconstructions in this region.

The Orlock Bridge Fault separates the Ordovician and Silurian turbidite sequences within the Southern Uplands thrust belt. A large biostratigraphical break and the 1 km wide sinistral Slieve Glah Shear Zone associated with the fault in northern Ireland led to previous interpretation as a major regional structure, possibly a terrane boundary. In Scotland, however, the stratigraphical break is much less and an association with inliers of the Moffat Shale Group suggests that the fault is essentially similar to the other tract-bounding faults which originated as syn-D1 thrusts within the imbricate stack. Localized sinistral deformation apparent along the trace of the Orlock Bridge Fault in southwestern Scotland, associated with post-1 reactivation, is comparable to that seen at Slieve Glah. Further east, a broad zone (up to 8 km) of sinistral ductile deformation, the Moniaive Shear Zone, is recognized adjacent to the Orlock Bridge Fault over a strike length of about 100 km. However, this zone differs from the Slieve Glah Shear Zone in its width and its location relative to the fault, suggesting that it is not simply related to the fault but represents a more regional deformation. Sinistral reactivation of the Orlock Bridge Fault was possibly initiated in the Wenlock during the peak of sinistral shear at the thrust front, although it may have developed over a long time contemporaneously with, but locally post-dating, the Moniaive Shear Zone. The latter deforms porphyroblasts with the thermal aureole of the c. 392 Ma Cairnsmore of Fleet granite pluton, which was emplaced into and largely post-dates the shear zone, but is deformed by the Orlock Bridge Fault. Major dip-slip reactivation of the fault post-dates the Moniaive Shear Zone and regional metamorphism and probably occurred in the Carboniferous or Permian. There is some evidence for a deep crustal feature coincident with the Orlock Bridge Fault, possibly the boundary between different crustal blocks in the collage of terrane fragments accreted during the final closure of Iapetus, which may explain the unusual extent of the reactivation of the Orlock Bridge Fault within the allochthonous Southern Uplands thrust stack. However, the situation of the fault within the Southern Uplands terrane and, in Scotland, the biostratigraphical evidence of no major stratigraphical break across the fault and the lack of any clear relationship between the Orlock Bridge Fault and the Moniaive Shear Zone indicate that the fault should not be regarded as a terrane boundary.

Magnetotelluric data from 16 soundings have provided a northwest-southeast traverse across the Northern and Central belts of the Southern Uplands. The 34 km profile extends from the Ordovician Portpatrick Formation (northwest of the Thornhill Basin) across the Silurian Gala Group onto the Permian Lochmaben Basin in the southeast. The survey traverses, and has particular relevance to, the Ordovician/Silurian boundary (Orlock Bridge Fault), the associated Moniaive Shear Zone, and the Moffat Valley Lineament. The data obtained have been successfully modelled (inverted) to provide a crustal-scale resistivity cross-section. The two Upper Palaeozoic basins are resolved as relatively conducting features. Basin depths are about 200 m (Thornhill) and 1100 m (Lochmaben). Elsewhere, through the upper crustal interval, resistivities over 500 ohm m are associated with the Lower Palaeozoic greywackes. The Silurian formations appear more resistive (and slightly thicker) than their Ordovician counterparts. Depth to basement (base Lower Palaeozoic) is not a well-resolved characteristic but resistivity gradients observed below 5 km suggest a depth of this order. Concealed beneath the Thornhill Basin lie two narrow, vertical, conductive zones. The more substantial feature lies towards the northwest margin of the basin, and is probably associated with the Orlock Bridge Fault. The second zone lies towards the southeast margin of the basin. The conductivity minima, in both cases, are located in the depth range 2 to 3 km; they are separate and distinct from the overlying basin sediments. If the two anomalies, some 5 to 6 km apart, can be considered an expression of the Moniaive Shear Zone, then the major conductive constituents are concentrated at the margins. Two dipping, resistive features in the upper kilometre are imaged in the vicinity of the Moffat Valley Lineament.

Detrital clinopyroxenes occur abundantly in several greywacke formations in the Southern Uplands, both as single grains and within andesite lithoclasts. Compositionally, these pyroxenes fall into two main groups. Those from the oldest unit, the early Caradoc Galdenoch Formation, were derived from a simple tholeiitic volcanic arc. In contrast, the pyroxenes in the younger Portpatrick (late Caradoc) and Glenlee (Ashgill) Formations and the Gala Group (early Llandovery), show initial strongly calc-alkaline affinities, followed by a more complex calc-alkaline to transitional (calc-alkaline to tholeiitic) volcanic arc signature. The latter may be due to a mixture of calc-alkaline and tholeiitic components. A presently exposed source for the andesites has not yet been established, although the Ballantrae Complex is a potential candidate for a tholeiitic component.

In the northern part of the Southern Uplands, restricted volumes of basic igneous rocks occur at or near the base of the Ordovician sedimentary strata. These rocks have previously been interpreted as ocean-floor tholeiites representative of the subducted Iapetus oceanic plate, preserved as tectonic slivers in a fore-arc accretionary prism. The alternative, back-arc basin model proposed for the Southern Uplands on sedimentological evidence raises questions over the origin of these rocks. New geochemical data and previously published data clearly indicate that the volcanic material does not have a simple single source. The oldest (Arenig) volcanic rocks from the Moffat Shale Group associated with the Leadhills Fault include alkaline within-plate basalts and tholeiitic lavas which possibly display geochemical characteristics of midocean ridge basalts. In the northernmost occurrence, alkaline and tholeiitic basalts contained within the Caradoc Marchburn Formation are both of within-plate ocean island affinity. To the south, in the Gabsnout Burn area, the Moffat Shale Group contains lenticular bodies of dolerite and basalt which have characteristics of island-arc to transitional basalts. This complex association of basaltic volcanic rocks is, at the present time, difficult to reconcile with either a simple fore-arc or back-arc setting for the Southern Uplands. However, the increasing arc-related chemical influence on basic rock geochemistry towards the southeast may tentatively be used in support of a southern arc-terrane, and as a result, a back-arc situation for the Southern Uplands basin. An alternative is that these volcanic rocks may represent the local basement to the basin and include remnants of an arc precursor to the Southern Uplands basin.

Regional lithogeochemical data from the Southern Uplands have been used to characterize a distinctive stratigraphy across this region. A suite of 840 point-counted and petrographically classified greywacke samples were used to establish chemical fingerprints for a series of greywacke-dominated lithostratigraphical units. These fingerprints were then used to evaluate a further 1455 greywacke samples collected throughout the Southern Uplands and Longford Down and enabled a series of strike-parallel geochemical tracts to be defined.

Four principal geochemical groups are recognized, relating to cratonic- and volcanic-derived greywacke provenances and both carbonate-rich and hydrothermally altered greywackes. Volcanic-derived units display higher Ti, Fe, Mg, Ca, Na, Mn, Cr, Ga, Ni, Sr, V and Zn values than their cratonic counterparts, which, with the exception of the carbonate-rich Hawick Group, display higher Si, K, La, Nb, Rb, Th and Zr. Volcaniclastic greywackes display REE patterns dissimilar to typical post-Archean upper crust, but similar to their andesitic components, whereas cratonic groups have REE patterns close to that of upper crust. Systematic strike-parallel geochemical variation in the Southern Uplands is controlled by petrographical differences which directly reflect provenance, with individual lithostratigraphical units derived from variable mixtures of ophiolitic, calc-alkaline, acid-igneous, low-grade metamorphic and carbonate-rich detritus.

The Southern Uplands greywacke succession (Scotland) accumulated at the Laurentian margin of the Iapetus Ocean. It was sequentially incorporated into an imbricate, accretionary thrust complex until closure of the ocean. Thereafter the thrust belt propagated across the suture zone as a foreland thrust belt directed towards the hinterland of Avalonia. A foreland basin migrating ahead of the thrust belt was the depositional site for the southernmost Southern Uplands units and the Windermere Supergroup (English Lake District). A Nd-isotope study has shown that juvenile ophiolitic detritus was introduced into the oldest, mid-Ordovician, Southern Uplands greywackes before two distinct provenance areas evolved: one supplying juvenile andesitic detritus in addition to a quartzo-feldspathic component, the other Proterozoic and exclusively quartzo-feldspathic. Bimodal composition continued into the early Silurian but was overlapped from late in the Ordovician by greywackes with intermediate Nd-isotope composition. This was not a simple mixing effect since the andesitic component is not represented and the necessary juvenile component comes from granodioritic and felsitic lithologies. Intermediate eNd values are then a consistent feature through the Silurian both in the younger strata of the Southern Uplands and in the earliest foreland basin turbidites of the Windermere Supergroup. The transition suggests cessation of volcanicity and erosion of deeper levels of the provenance terrane(s), possibly linked to the evolution of the basin system from active margin, accretion-related, to a foreland setting. To the north of the Southern Uplands terrane, beyond the Southern Upland Fault, a Caradoc to Wenlock turbidite sequence occupies inliers within the Midland Valley. The older greywackes contain abundant juvenile ophiolite and plutonic detritus in addition to a quartzofeldspathic metamorphic component; there are similarities with the most northerly part of the Southern Uplands. From the late Ordovician, εNd values systematically decline so that early Llandovery Midland Valley greywackes are exclusively quartzo-feldspathic, derived from an ancient source indistinguishable in isotopic terms from that periodically supplying the Southern Uplands. In general the Llandovery Midland Valley provenance was significantly more mature than that contemporaneously supplying the Southern Uplands. Thereafter, the Midland Valley latest Llandovery and early Wenlock greywackes contain a higher proportion of a juvenile component, and by the early Wenlock, greywackes from the Midland Valley, Southern Uplands and Lake District terranes are similar in terms of εNd. A common provenance seems likely and suggests that by the mid-Silurian all three terrenes were in close proximity.

In the Irish Caledonides, volcanism has been significant in terrane identification and in reconstructions of the Appalachian/Caledonian orogen. Crucial to these reconstructions is the recognition of ocean margins using obducted ocean floor relics (ophiolites) and supra-subduction zone volcanic assemblages. The volcanic rocks provide much evidence for the affinity of a terrane, however, by analogy with present day examples, the ocean floor sediments may provide the best way of tracing both ocean-floor magmatic activity, and continental source areas. This investigation shows that the Irish Lower Palaeozoic volcanogenic terranes can be discriminated in terms of their shale geochemistry, which also gives information on their provenance and environment of deposition. South Mayo shales are dominated by volcaniclastic material derived both from both an arc and from an ophiolitic source. The Northern and Central belts of the Central Terrane show very similar lithogeochemistries, apparently derived in part from intermediate to silicic volcanic complexes. The Ordovician-Silurian inliers that straddle the Suture Zone, here termed the Southern Domain, show a chemistry close to that of the Leinster Terrane, which, coupled with a greater degree of sea-floor weathering, suggests a terrane with sediment of both volcanic and continental provenance being deposited in deeper water further from land. Across the suture the Leinster Terrane shows a mature chemistry which clearly suggests a continental provenance, together with a volcanogenic input from supra-subduction volcanism. This maturity is probably due to slower rates of sedimentation with longer residence times for volcanic detritus, plus the existence of a deeply weathered continental basement.

The Iapetus Suture (Solway) line coincides with a magnetic low, which lies between magnetic highs over southwestern Scotland and the Lake District-Isle of Man region. Although topography on deep magnetic basement can account for these long wavelength geophysical variations, an explanation which involves lateral basement magnetization contrasts is preferred on the basis of (a) correlations between inferred magnetization boundaries and major structures delineated from other evidence, and (b) the apparent westward continuation of the Solway low through Ireland and Newfoundland across areas with very different subsidence histories but similar position with respect to the collision of Laurentia and Avalonia. In the preferred model, relatively magnetic continental crust beneath the Southern Uplands and Lake District terranes is separated by a zone of less magnetic crust interpreted as sedimentary rock of Avalonian affinity carried to deeper structural levels within the Iapetus Suture Zone. The magnetic unit beneath the Southern Uplands is bounded to the south by the northward-dipping Iapetus Suture and to the north by a structure which may have been reactivated in late Caledonian times to produce the Moniaive Shear Zone in the overlying rocks; this unit may represent the ‘missing’ arc terrane inferred from provenance studies. Alternatively, the two magnetic basement domains may have originally been part of the same terrane, with that portion beneath the Southern Uplands rifting from the Avalonian continent during its northwards drift and being subsequently trapped in the hanging wall of the Iapetus Suture. The southern margin of the Lake District domain appears as a discontinuity in the magnetic anomaly pattern, with long wavelength anomalies to the south having a southeast ‘Tornquist’ trend.

Late Caradoc graptolite assemblages across the Iapetus Ocean in Wales and Scotland became progressively more disparate despite the narrowing of the ocean. We compare faunal distributions in continuous sections from opposite sides of Iapetus, at Whitland in South Wales and Hartfell Score in the Southern Uplands of Scotland. A comprehensive graptolite range-chart is given for each. The graptolite assemblages from the clingani Biozone are subdivided into a lower Ensigraptus caudatus Subzone and upper Dicellograptus morrisi Subzone at both localities, though the faunas differ in detail. Higher in the sequence, the distinctive Scottish linearis Biozone fauna is not recognizable at Whitland, its presumed equivalent being a fauna dominated by variable morphotypes of the genus Normalograptus. This suggests that environmental gradients (depth and/or temperature) were changing more rapidly than the geography. Significantly, the subsequent inception of limestone deposition at Whitland was approximately contemporaneous with widespread replacement of black mudstones by oxic, bioturbated sediments elsewhere in the Welsh Basin and in Scotland. This was possibly a response to an early phase of cooling prior to the end-Ordovician glaciation. In Wales the claimed hiatus at the Caradoc-Ashgill boundary may rather reflect biofacies variation.