Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

Part 3. The legacy of Lyell

Lyell would have approved of plate tectonics. Indeed, by implication, he almost anticipated it. Recognizing the connection between climate and the global distribution of land and sea, as Fleming points out (pp. 164-165 of this volume), Lyell explained climatic change in the geological past in terms of changes in the geographical distribution of land and sea - continental drift in disguise. Furthermore, since the paradigm of plate tectonics is founded upon an understanding of processes active at present, whose rates can be measured, then applying them to the past, it is entirely in accord with Lyell's principle that geological 'phenomena should be explained only in terms of causal agencies that are observably effective, both in kind and degree' (Rudwick p. 4 of this volume). In a new synthesis of the Lower Palaeozoic tectonic evolution of the northern Appalachians and British Caledonides, Cees van Staal and his colleagues draw upon the modem analogue of the southeast Asia region. Intriguingly, they are able to turn the comparisons around to predict that the modem tectonic activity of southeast Asia will eventually end up in structures looking much like those of the Appalachian-Caledonian orogenic belt. This carefully argued work, paying close attention to field observations, is in the true tradition of Lyell and is firmly set within his principles. The legacy of Lyell is equally evident in Andrew Scott's paper in which he has literally followed in Lyell's footsteps across to North America. Scott discusses on-going research into the occurrence of reptiles preserved inside the upright trunks of Upper Carboniferous trees at Joggins, Nova Scotia, first discovered by Lyell and Dawson in 1852. Scott examines Lyell's ideas on the formation of coal and brings them up to date using modem analogues from southeast Asia. However, he sounds a note of caution in the use of modem analogues since it has now been established that Carboniferous coal-forming plants had life habits and growth mechanisms that were radically different from those of modem peats. The recognition by Lyell of charcoal in coal deposits resulted only many years later in an understanding of its origin from ancient forest fires. Scott explains how current research on ancient charcoal is providing new insights into climatic change, atmospheric composition, and processes of erosion and sedimentation. Lyell visited Nova Scotia twice during his visits to North America. The influence of Lyell on Nova Scotian geology and the use that Lyell made of his fieldwork in his publications is examined by John Calder. Calder provides us with a succinct account of the Carboniferous evolution of Nova Scotia and places Lyell's observations and conclusions in context. Clearly Nova Scotia provides an important link between Europe and North America. Lyell was impressed by the geology of Nova Scotia and his enthusiasm was reciprocated by the local geologists. Particularly important was the influence that Lyell had on Sir William Dawson who was to become one of Nova Scotia's most important and influential geologists. Lyell used many of his discoveries in Nova Scotia to illustrate subsequent editions of his volumes including his discovery of tetrapods within sandstone-filled lycophyte trunks at Joggins. Unravelling the stratigraphic record was always a fundamental objective of Lyell. His desire to understand rates of sedimentation, rises and fall of land and sea level changes as seen both in ancient rocks and as active processes in the modem world is evident from his field observations as recorded in his diaries, travel, and geological books and papers. Our current understanding of the stratigraphic record has been revolutionized in the past twenty years with the advent of sequence stratigraphy as shown by Chris Wilson. The paper by Wilson looks at how the sequence stratigraphic model evolved and its utility. Wilson argues that the application of sequence stratigraphy has provided us with a revolutionary new way of interpreting the stratigraphical record. However, Wilson adds caution in that a new global stratigraphy, allowing global correlations based upon eustatic signals, is not yet a reality as their recognition can not yet be detected unequivocally in the record. Clearly again Lyell takes us back to looking at the rock record but as he also maintained, we must also consider the processes involved that were responsible for what we observe. This point is amplified by Chris Talbot in a review of salt tectonics in the Zagros mountains of Iran. Talbot notes the parallels between the controversy in Lyell's day about whether ice could flow uphill and the more recent controversy about whether salt can flow across the surface of the land or, indeed, the sea bed. Talbot takes the argument a stage further in demonstrating how field measurements of the rates of salt flow and observations of the domal shapes of extrusions are used to quantify the dynamics of the process. He hints at the possible analogy between salt extrusion and the extrusion and gravity spreading of metamorphic core complexes. This richly illustrated, authoritative account of salt tectonics in a classic area is yet further evidence of Lyell's influence on the approach to modem research. Lyell was much concemed with the most tangible expressions of current geological activity in the form of volcanic eruptions and earthquakes. He spent a major part of his life in the field studying volcanoes in Italy and elsewhere, as Wilson describes on pp. 24-31 of this volume, and he was well aware of their Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

198 PART 3 destructive power. Mount Etna was of particular significance, so it is an essential part of Lyell's legacy that volcanic activity on Etna should be closely monitored, both to give greater quantitative understanding of the process of eruption and to give predictive power to mitigate the risks from future eruptions. Hazel Rymer is a member of an international, multidisciplinary team of scientists who have set up a huge array of instrumentation to monitor every waking moment of this volcanic giant. Their paper gives a historical account of volcano monitoring and describes the modern techniques being utilized on Etna, which are advancing all the time. This is especially so with satellite measurements, including observations of ground deformation using GPS and SAR, which they confidently expect to revolutionize monitoring methods within the next decade. Bruce Bolt brings a wealth of experience to review the advancement of seismology since Lyell's day. Arguably the most significant advance was the construction of the Milne-Shaw seismograph and its use in a global network of observatories set up at the end of the nineteenth century. This duly advanced to the Worldwide Standardized Seismograph Network in the 1950s which has now evolved into the Global Digital Network of seismograph stations. Bolt traces the history of seismology from the valuable descriptive accounts of earthquakes in Lyell's time to the present day analysis of seismic waveforms which gives detailed information about both the nature of the earthquake itself and the nature of the Earth's interior along the pathways of the seismic waves. Seismology has become one of the most powerfol tools for exploring and imaging the Earth's interior, well beyond anything that Lyell could have dreamed of, and continues to be the main means of monitoring earthquakes, in efforts to mitigate their risks. Earthquake prediction, as Bolt explains, remains elusive despite huge research efforts in recent years. The current view is that it is likely to remain so because of the frictional nature of the earthquake mechanism and the heterogeneity of the Earth's crust. A better strategy for risk mitigation is to concentrate on reducing the vulnerability of the population at risk by means of planning, education and the construction or upgrading of buildings and other structures to appropriate levels of earthquake resistance. Lyell was intensely interested in the antiquity of Man, as Cohen (pp. 83-93) and other authors in Part 1 of this book have made clear. Lyell was also intensely interested in the natural environment since this is the laboratory in which to record and measure geological phenomena active at present which are the basis for understanding the geological past. Lyell was well aware of the power of nature over Man and of human dependence on the Earth for its resources, but there was little regard in Lyell's day for the need to protect the Earth from human activities. Nowadays it is a major issue and Sir John Knill completes this book with a telling essay on Man and the modern environment to make the point that, more than ever, we need to apply Lyell's principles. It is now imperative to understand and quantify the mechanisms of global climate change in the geological past, especially over the last 500 000 years, in order to predict future climate change including the anthropogenic contribution. Equally, geological processes on all scales must be better understood in order to build strategies for the environmental sustainability of the Earth. There is thus a critical role for geologists in modern society. The old textbook shorthand for Lyell's principles that 'the present is the key to the past' has to be turned around into the title of this book: the past is the key to the present - and, indeed, to the future. Derek J. Blundell Andrew C. Scott Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

The Cambrian-Silurian tectonic evolution of the northern Appalachians and British Caledonides: history of a complex, west and southwest Pacific-type segment of Iapetus

C. R. VAN STAAL 1, J. F. DEWEY 2 , C. MAC NIOCAILL 2 & W. S. MCKERROW 2 1 Geological Survey of Canada, 601 Booth Street, Ottawa, Ontario, K1A OE8 Canada 2 Department of Earth Sciences, Oxford University, Parks Road, Oxford OX1 3PR, UK

Abstract: This paper presents new ideas on the Early Palaeozoic geography and tectonic history of the Iapetus Ocean involved in the formation of the northern Appalachian-British Caledonide Orogen. Based on an extensive compilation of data along the length of the orogen, particularly using well-preserved relationships in Newfoundland as a template, we show that this orogen may have experienced a very complicated tectonic evolution that resembles parts of the present west and southwest Pacific Ocean in its tectonic complexities. Closure of the west and southwest Pacific Ocean by forward modelling of the oblique collision between Australia and Asia shows that transpressional flattening and non-coaxial strain during terminal collision may impose a deceptively simple linearity and zonation to the resultant orogen and, hence, may produce a linear orogen like the Appalachian-Caledonian Belt. Oceanic elements may preserve along-strike coherency for up to several thousands of kilometres, but excision and strike-slip duplication, as a result of oblique convergence and terminal collisional processes, is expected to obscure elucidation of the intricacies of their accretion and collisional processes. Applying these lessons to the northern Appalachian-Caledonian belt, we rely principally on critical relationships preserved in different parts of the orogen to constrain tectonic models of kinematically-related rock assemblages. The rift-drift transition, and opening of the Iapetus Ocean took place between c. 590-550 Ma. Opening of Iapetus was temporally and spatially related to final closure of the Brazilide Ocean and amalgamation of Gondwanaland. During the Early Ordovician, the Laurentian margin experienced obduction of young, supra-subduction-zone oceanic lithosphere along the length of the northern Appalachian-British Caledonian Belt. Remnants of this lithosphere are best preserved in western Newfoundland and are referred to as the Baie Verte Oceanic Tract. Convergence between Laurentia and the Baie Verte Oceanic Tract was probably dextrally oblique. Slab break-off and a subsequent subduction polarity reversal produced a continental magmatic arc, the Notre Dame Arc, on the edge of the composite Laurentian margin. The Notre Dame Arc was mainly active during the late Tremadoc-Caradoc interval and was flanked by a southeast- or south-facing accretionary complex, the Annieopsquotch Accretionary Tract. Southerly drift of Laurentia to intermediate latitudes of c. 20-25~ was associated with the compressive (Andean) nature of the arc and the accompanying backthrusting of the already- accreted Baie Verte Oceanic Tract further onto the Laurentian foreland. Equivalents of the Notre Dame Arc and its forearc elements in the British Isles have been preserved as independent slices in the Midland Valley and possibly the Northern Belt of the Southern Uplands. During the late Tremadoc (c. 485 Ma), the passive margin on the eastern side of Iapetus also experienced obduction of primitive oceanic arc lithosphere. This arc is referred to as the Penobscot Arc. The eastern passive margin was built upon a Gondwanan fragment (Ganderia) that rifted off Amazonia during the Early Ordovician and probably travelled together with the Avalonian terranes as one microcontinent. The departure of Ganderia and Avalonia from Gondwana opened the Rheic Ocean. Equivalents of the Penobscot Arc may be preserved in New Brunswick and Maine, Leinster in eastern Ireland, and Anglesey in Wales. An arc-polarity reversal along the Ganderian margin after the soft Penobscot collision produced a new arc: the west-facing Popelogan-Victoria Arc, which probably formed a continuous arc system with the Bronson Hill Arc in New England. The Popelogan-Victoria Arc transgressed from a continental to an oceanic substrate from southern to northeastern Newfoundland. Rapid roll-back rifted the Popelogan-Victoria Arc away from Ganderia during the late Arenig (c. 473 Ma) and opened a wide back-arc basin; the Tetagouche-Exploits back-arc basin. The Popelogan-Victoria Arc was accreted sinistrally oblique to the Notre Dame Arc and, by implication, Laurentia during the Late Ordovician. After accretion, the northwestward-dipping subduction zone stepped eastwards into the Tetagouche-Exploits back-arc basin. Equivalents of the Popelogan-Victoria Arc in the British Isles may be preserved as small remnants in the Longford Down Inlier in Ireland. The Longford

VAN STAAL, C. R., DEWEY, J. E, MAC NIOCAILL, C. • MCKERROW, W. S. 1998. The Cambrian-Silurian 199 tectonic evolution of the northern Appalachians and British Caledonides: history of a complex, west and southwest Pacific-type segment of Iapetus. In: BLUNDELL,D. J. & SCOTT, A. C. (eds) Lyell: the Past is the Key to the Present. Geological Society, London, Special Publications, 143, 199-242. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

200 r R. VAN STAAL ET AL.

Down Arc is not preserved in Scotland, although its presence has been inferred there on the tenuous basis of arc detritus. The suture between the Notre Dame Arc and the Popelogan- Victoria-Longford Down Arc system is the Red Indian Line in the Northern Appalachians, but in the British Isles the position is not clear. The fault-bounded Grangegeeth Arc terrane in eastern Ireland, immediately to the south of the Longford Down inlier, may be a displaced piece of the Popelogan-Victoria-Longford Down Arc system. Diachronous closure of the Tetagouche- Exploits basin during the Ashgill to the Wenlock finally caused the collision between Ganderia/Avaloniaand Laurentia, whereas the Lake District Arc is related to an earlier closure of the Tornquist Sea between Baltica and Avalonia. After arrival of Avalonia at the Laurentian margin, continuous, dextral oblique convergence between Gondwana and Laurentia was accommodated by another northwest-dippingsubduction zone, this time in the Rheic Ocean. The Acadian orogeny in both North America and the British Isles occurred in the Early to Mid- Devonian and is probably related to the collision of Gondwana and/or peri-Gondwanan elements (Meguma, Armorica etc.) with the northern continents.

Many Palaeozoic and older orogens, such as the Caledonide segment (Figs 1-3) have few, if Urals and the Appalachian-Caledonide belt (inset any, comparable equivalents in the southern of Figs 1-3), although narrowing and widening and Appalachians. Some related oceanic element(s) possessing modest, very open oroclines, are may be involved in the formation of the British and remarkably linear or very gently arcuate over Scandinavian Caledonides, but the collisonal thousands of kilometres. Within these orogens, it histories of these two segments are markedly has been normal practice to draw zones and different. Because several tectonic events took subzones characterized by particular tectonic place temporally so closely in each of the segments, elements and styles in semi-linear fashion along these differences have often not been fully great strike-lengths (e.g. Dewey 1969, H. Williams appreciated and there has been a tendency to relate 1978); i.e. we may have tended to 'impose' an these tectonic events to one unifying kinematic along-strike correlative unity and continuity of process. For example, evidence of tectonic loading elements and zones. This is in striking contrast to and orogenesis of the Laurentian margin during the many Mesozoic orogens such as the Tethysides Early to Mid-Ordovician (Taconic Orogeny) can be with their great complexity of internal loops, oro- found from Spitsbergen in the Arctic through the clines and syntaxes. It is, also, in profound contrast British Caledonides (Fig. 3) and Northern to modern convergent plate boundary systems such Appalachians (Figs 1 and 2) into the Southern as the west and southwest Pacific (Fig. 4) where Appalachians in the United States, with the notable very great geometric and kinematic complexity is exception of East Greenland. The subsequent kine- accompanied by rapid temporal and spatial tectonic matic correlation along the length of the change. From the typical array of tectonic Appalachian-Caledonide Orogen led to a recent indicators of plate boundary zones such as rifted controversy whether the Taconic Orogeny of the margins, ophiolites, blueschists and calc-alkaline northern Appalachians resulted from collision volcanism, we are, nevertheless, confident that between Laurentia and the South American margin these older, now intracontinental, orogens were of Gondwana (Dalziel et al. 1996) or from an arc- formed by plate tectonic processes (Bird & Dewey continent collision. The former model is incom- 1970; Hamilton 1970). Hence, we must ask the patible with the geological relationships preserved question: is substantial along-strike continuity a in the northern Appalachians and the British reality or is it a consequence of a linearity imposed Caledonides because it is difficult to see how the by terminal continental collision following an Gondwanan margin could have bypassed the earlier long and complicated continental margin independent systems of oceanic arcs in the Iapetus and oceanic tectonic history? Ocean that were converging and/or colliding with This problem is particularly pertinent to the the Laurentian margin during the Ordovician (e.g. Appalachian-Caledonide Orogen, because faunal Mac Niocaill et al. 1997). Moreover, the faunal provinciality, palaeomagnetic data and structural distinctions between Laurentia and Gondwana are studies indicate that it consists of along-strike not consistent with any close convergence before segments that experienced different tectonic the Late Ordovician (Cocks & Fortey 1990) histories. This is largely the result of the contem- In this paper, we present new ideas on the tec- poraneous convergence of several continents and tonic evolution of the northern Appalachian- microcontinents, Baltica, Avalonia and Carolinia, British Caledonide Belt and their implications for with Laurentia. The various oceanic elements the history of Iapetus. The rapidly-growing geo- preserved in the Northern Appalachian-British logical database for this segment of the Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 201

Appalachian-Caledonide Orogen, particularly the the Llandeilo series is reduced to a stage large number of recent, high quality, U-Pb age (Llandeilian) of the Llanvirn series. The expanded dates and isotope tracer and provenance studies Llanvirn now ranges from 470 to 458 Ma. The combined with detailed field studies, has indicated Tremadoc (the base of the Ordovician) starts at c. that its tectonic history was much more complex 495 Ma and ends at c. 480Ma (Landing et al. than previously thought and resembles parts of the 1997). The Tremadoc and Arenig together define modern west and southwest Pacific Ocean. Despite the Early Ordovician (495-470 Ma). The com- identification of the internal complexities, we will position and tectonic setting of volcanic rocks is argue that the northern Appalachian-British indicated by their commonly-used abbreviations. Caledonide belt largely represents a tectonic/ MORB stands for mid ocean ridge basalt; the prefix kinematic entity, i.e. all parts of this segment of the N, T or E refers to normal, transitional or enriched, orogen experienced more or less the same major respectively. OFB and OIB refer to ocean floor and tectonic-kinematic events, although some tectonic ocean island basalts. IAT and CAB refer to island elements appear to be confined to restricted parts of arc tholeiite and calc-alkaline basalt. the orogen. For example, the Mid- to Late Ordovician arc volcanics in the English Lake District, Wales and south-east Ireland have no The west and southwest Pacific and a obvious equivalents further west in the northern future Australia-Asia sinistral oblique Appalachians. collision Tectonic reconstructions are hindered by struc- tural excision, particularly during large-scale We have constructed an extremely simplified orogen-parallel strike-slip motion. For example, version (Fig. 5a) of the tectonic map of the west and some Irish and Scottish accretionary terranes may southwest Pacific (Fig. 4) and allowed the present have been moved for large distances along the relative motions (De Mets et al. 1990) among the eastern margin of Laurentia. Elsewhere, post- Asian, Pacific and Indian/Australian Plates and Ordovician cover sequences may mask important smaller plates to run to 'completion', that is the Early Palaeozoic structures. These problems can be collision of Australia with the Asia continent along overcome only through an overview of the whole a broad zone of sinistral oblique convergence (Fig. orogen, so that segments with missing or additional 5b). The model assumes a 55 million year con- components can be recognized. We use the con- stancy of relative plate motions. We recognize, of straints imposed on different parts of the orogen to course, that the Pacific has been a long-lived ocean model the overall tectonic processes responsible for since the late Proterozoic and may not fully close in some regions where local evidence is lacking or the short-lived way that Iapetus opened and closed poorly preserved. The focus of this paper is on in c. 160 Ma. This exercize was carried out mainly tectonic processes active during the Cambrian and to test whether the superficial simple zonation and Ordovician, and their implications for under- linearity of the Appalachian-Caledonide Orogen standing the destruction of the Iapetus Ocean. We could be a result of terminal collisional processes. will reaffirm earlier notions that the Ordovician In making this prognostic reconstruction, we have orogenesis in the northern Applachians and British had to make a series of 'iterative tectonic Caledonides (Taconic and Grampian orogenies, decisions', especially in the siting of transform respectively) cannot be related exclusively to faults and how volcanic arcs are transported and interaction of one single volcanic/magmatic arc rotated. Others might have made different tectonic with the Laurentian margin during the Ordovician. decisions but the final collided result would be Palaeomagnetic and geological data (H. Williams et unlikely to be fundamentally different from the one al. 1988; Colman-Sadd et al. 1992a; van der Pluijm shown in Fig. 5b in its general form of sliced linear et al. 1995; Mac Niocaill et al. 1997), orogenesis component terranes that are exotic with respect to along the Avalonian margin (van Staal 1987, 1994: the cratons and to each other. We have run the Colman-Sadd et al. 1992b) and significant model forward four times, each time making inconsistencies between timing of obduction of different iterative decisions: the results differ ophiolites, Mid-Ordovician Taconic deformation slightly in details but not in essentials. There are and metamorphism, and age of classic Taconic arc four basic conclusions to be drawn from this magmatism (e.g. Tucker & Robinson 1990, exercise. First, plate boundary zones of immense Karabinos et al. 1993; Pinet & Tremblay 1995) complexity may be swept up and transpressionally make such a model untenable. flattened, sheared and rotated into semi-linear In this paper, we follow Fortey et al. (1995) in orogenic zones resulting in a pseudo-simplicity that the coincidence of the base of the Caradoc series conceals an earlier complicated history and with the base of the N. gracilis zone (c. 458 Ma; geometry. Secondly, contemporaneous volcanic Tucker & McKerrow 1995 and references therein); arcs with quite different original palaeogeographic Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

202 c.R. VAN STAAL ET AL. position and orientation may give an illusion of c. 100 Ma to close at present rates of relative plate along-strike continuity. Thirdly, fragments of a motion. Given the final collisional result (Fig. 5b), once continuous tectonic element, like the Ontong- we are faced with an intractable facet of inverse Java/Caroline-Truk Plateau may be scattered science, i.e. we can forward model with 'impunity' throughout thousands of kilometres of the final but backward modelling is a near impossibility collisional system. Fourthly, the interval from the given the multiple pathways that may lead to a final present to the terminal Australia-Asia collision is result. Perhaps the greatest real continuity exists only c. 45 Ma, about the same interval as between along the edges of an orogenic belt where the the first interaction of oceanic elements with the results of collision between the old rifted margins Laurentian and Avalonian margins in the Early and early oceanic elements (e.g. arcs) have been Ordovician and Late Silurian terminal closure of preserved. The further one goes into the interior of Iapetus. The western Pacific would take only an orogen with progressively more oceanic Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 203 elements, the more one faces the destructive com- single plate boundary depending upon the way in plexity of plate tectonics. In orogenic interiors, two which it crosses rotational latitudes. Secondly, the particular problems raise further difficulties, enormous variety of triple junctions (McKenzie & namely significant 'terraning' and bulk rotation of Morgan 1969) and their evolution can yield fast- whole arcs and/or other tectonic elements. changing tectonic environments that allow the rapid Quite apart from the flattening and smearing superposition of quite different strain fields and imposed by oblique terminal collision, the earlier consequent polyphase deformation and rapid tectonic collages are beset by a number of sub- igneous, metamorphic and stratigraphic changes stantial complexities and subtleties that are inherent (Dewey 1975). Third, in a mosaic consisting of in plate kinematics but which are unlikely to be three or more plates, slip vectors across plate preserved in the final collisional product. When we boundary zones must change progressively so that analyse old orogens, we try to apply the broad rates and directions of plate motion change and tenets and geological corollaries of plate tectonics may eventually cause a change in plate boundary but the difficulties are many. First, very rapid type (Dewey 1975). One especially important spatial changes in tectonics may occur along a example is the nucleation of intraoceanic arcs on

Fig. 1. Simplified geological map (a) of Newfoundland after Colman-Sadd et al. (1990) with the various tectonic elements discussed in this paper. The legend (b) represents a summary of the structural relationships between the various tectonic elements. This legend also applies to Figs 2 and 3, where the Newfoundland-defined elements are replaced by their equivalents in New England and the British Isles. respectively. The white areas in the Central Mobile Belt represent Silurian or younger volcanic or sedimentary rocks. An = Annieopsquotch ophiolite; BC = Betts Cove Ophiolite Complex; BE = Baie d'Espoir Group; BH = Baggs Hill Granite; BHC = Blue Hills of Couteau Ophiolite Complex; BOIC = Bay of Island Ophiolite Complex: BN = Bay du Nord Group: BP = Burlington Pluton; BU = Buchans Group; CC = Cotrell's Cove: CM = Carmanville Melange: CP = Chanceport Group; CPC = Coy Pond Complex; CS = Cold Spring Pond Formation: CRF = Cape Ray Fault; DA = Davidsville Group: DBL = Dog Bay Line; E = Exploits Group; GL = Grand Lake Complex: GRUB = Gander River Ultrabasic Belt: HMT = Hungry Mountain Thrust; LA = Lake Ambrose Volcanic Belt: LB = Lushs Bight Group: LGLF = Little Grand Lake Fault: LP = Little Port Complex; LR = Long Range Ophiolite Complex: Ma=Margaree Complex: MB -- Mings Bight: MH = Moreton's Harbour Group; Pa = Partridgeberry Hill Granite: PP -- Pipestone Pond Ophiolite Complex: RA = Robert's Arm; RIL = Red Indian Line; SA = St Anthony Ophiolite Complex: SC = Sleepy Cove Group: SU = Summerford Group & Dunnage Mdlange; TU = Tulks Hill Volcanic Belt: Tw = Twillingate Trondhjemite; VM = Victoria Mine Volcanic Belt; WB = Wild Bight Group. Inset shows the Appalachian-Caledonian Belt with the approximate distribution of the various (micro)continents involved in this orogen. TL is Tornquist Line and represent the suture between Avalonia and Baltica. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

204 c.R. VAN STAAL ET AL. oceanic transform/fracture zones (Casey & Dewey collisional history. We do our best, but solutions are 1984). Fourth, the relative motion of oceanic arcs exceedingly unlikely to be unique. However, in may be wholly unrelated to motion among the spite of these caveats in attempting pre-Mesozoic larger bounding plates. For example, in the western palaeotectonic syntheses, areas such as the west Mediterranean, (Dewey et al. 1989) arcs roll and and southwest Pacific (Fig. 4) give us deep insights spread radially, with substantial longitudinal into the kinematics and histories of older orogenic extension, driven by subduction roll-back (Dewey belts and into the wide variety of tectonic environ- 1980) such that they invade other oceanic tracts ments and their kinematic relationships that permit leaving young back arc basins in their wake to die portions of those older belts to be explained only when they collide with other arcs and rifted rationally in terms of plate boundary zone margins. Thus, such arcs tend to fill and obliterate tectonics. older oceans and generate young rear-arc oceanic We now, briefly, outline ten features of the lithosphere and develop complex looping and tectonics of the southwest Pacific that, we believe, oroclinal shapes. Fifth, major plate reorganization may be useful as analogues in explaining various may result from large-scale continental collision features of. and providing a backdrop to. the Early whereby new plate boundaries are formed and Palaeozoic evolution of the northern Appalachians major changes occur in the direction and rates of and the British Caledonides. First, passive, rifted relative plate motion. Sixth, we point to the great margins, that form the edges of orogens and are geometric and kinematic complexity of present destroyed commonly by collision with arcs, may be west and southwest Pacific tectonics (Fig. 4) that of several types and ages in relation to the progres- are hardly decipherable today let alone in their sive closure history of oceans. The northern edge of likely condition 45 Ma hence. Our principal con- Australia is a Jurassic rifted margin in collision clusion is that the inherent complexities of plate with the Timor/Banda/New Guinea arc complex, tectonics and the strains imposed by terminal beginning in the Miocene, diachronous from east to collision make difficult the unique elucidation of west, and still progressing westwards. This might the tectonic history of orogenic belts, such as the be termed a grand-scale collision, leading to Appalachian/Caledonian Orogen, that have resulted subduction polarity flip (Flores/Wetar Thrust and from the progressive closure of large oceans. New Guinea/North Solomons Trench) and the This is not an exclusive subscription to pessi- gross destruction of a rifted margin on a massive mism for understanding pre-Mesozoic tectonics. scale, similar to the Early Ordovician destruction of The original rifted margins of a closed oceanic tract the Laurentian margin (Dewey & Shackleton 1984: may well preserve an along-strike continuity Dewey & Ryan 1990). In contrast, the Tasman Sea relating to the timing and geometry of rifting: and the Coral Sea opened to form the eastern rifted oceanic elements that collide early with these rifted margin of Australia during the Late Cretaceous to margins may provide continuity for up to several early Tertiary: its destruction during the next 45 Ma thousand kilometres. Rather, it is a warning that we will probably be by collision with the Lord Howe must not expect to unravel, uniquely, the tectonic Rise and Vanuatu arc, respectively (Fig. 5b). In history of an orogen, especially its pre-terminal even greater contrast are the rifted margins of the

Fig. 2. Simplified geological map of maritime Canada and New England with the continuation of the Newfoundland tectonic elements displayed in Fig. 1. Geology largely based on tectonic lithofacies map of H. Williams (1978) and plate 2 of Rankin et al. (1990) with tectonic additions by the authors. An = Annidale Belt: AW = Ascot Weedon Volcanic Belt; At = Attean Pluton: B = Brookville Terrane; BHA= Bronson Hill Arc: BMC = Boil Mountain Ophiolite Complex; BdO = Bras d'Or Terrane; BRM = Belledune River Melange: C = Caledonia Terrane; CL = Chain Lakes Massif; CM= Caucomgomoc Melange: F = Fournier Group: FT = Fredericton Trough; M = Mira Terrane; HM = Hurricane Mountain Mrlange: KC = Kingston Complex: Mu = Munsungun basalts: Ma = Massabesic Gneiss; N = Nashoba Terrane; P = Pelham Dome; Po = Popelogan Inlier: RHB = Rowe-Hawley Belt; Ro = Rockabema Diorite; S = Shelburn Falls Dome: SQOB=Southern Quebec Ophiolite Belt;T = Tetagouchc Group; U= Upsalquitch Gabbro; W = Winterville Basalts; WL = Weeksburo Lunksoos Inlier. (1) Continuation of Baie Verte Line; (2) Approximate trace of Red Indian Line (the second suture immediately to the south of the Winterville and Munsungun Basalts has been added to emphasize our interpretation that these basalts are remnants of seamounts. Analogous to the Summerford Group in Newfoundland these basalts are thought to have accreted to the Popelogan Arc before accretion to Laurentia); (3) Inferred continuation of Dog Bay Line and/or associated major structures (this suture is poorly defined because the fault system associated with collision between Ganderia and Laurentia in Maritime Canada and New England, during the Early Silurian, probably involved more than one collisional/ accretionary event due to the piecemeal arrival of small continental blocks with Ganderian basement; hence other important sutures probably occur southeast of line 3). Inset same as Fig. 1, Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

LAURENTIAN SHIELD

CALEE

I: i Meguma &Cadomia ,,,.v,,,

G 0

Ol:~ganio (:i3.18t \

3

Frededcton Trough

J

/ ~ ~ ~ ~ ~ ~ ~ .,=..,.,. ,.,~ ~=l=~ ,''.~' -- • ,,.= =~' "w'==" ~

f, ,,-'- f / ' 1 !!i MEGUMA

200 km I J Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 205

Fig. 3. Simplified geological map of the British Isles with the continuation of the Newfoundland tectonic elements displayed in Fig. 1. Map patterns are given by the legend in Fig. lb. B = Bray Group: BB = Balbriggan; Be = Bellewstown: BL = Ballantrae Ophiolite: Bo = Borrowdale: C = Cahore Group: CB = Central Belts: CBC = Clew Bay Complex; Cd = Coedana Complex: CT = Cullenstown Formation: GGF = Great Glen Fault; GR = Grangegeeth Terrane; HBC = Highland Border Complex: HBF = Highland BoundaD' Fault: LL -- Leadhills Line: M = Moine: Ma = Manx Group: MSF = Menai Strait Fault: MSG = Monian Super Group: MT = Moine Thrust: MVT = Midland Valley Terrane; MW -- Mweelrea Ignimbrites; NBMF = Northern Belt Median Fault: NSF = Navan-Silvermines Fault; OBF = Orlock Bridge Fault: R -- Rhobell Volcanic Complex: RDG = Ribband and Dungannon Groups: Ro = Rosslare Complex: SA = Slieve Aughty: SB = Southern Belt: SC = South Connemara Terrane: SF= Slane Fault: SCG = South Connemara Group: Sk = Skiddaw Group; SMT = South Mayo Trough: SUF = Southern Uplands Fault: Ta = Tattinlieve; Tr = Treffgarne: Ty = Tyrone Ophiolite Complex: WFZ = Wicklow Fault Zone. Inset same as Fig. 1. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

206 C. R. VAN STAAL ET AL.

Fig. 4. Simplified tectonic map of the southwest Pacific. FR = Finisterre Ranges. NBT = New Britain Trench. NFF = North Fiji Fracture Zone. SST = South Solomon Trench & TT = Tobriand Trench. Data. principally, from the Geological World Atlas (1976). Hall (1996, 19971. McCaffrey (1996t and the Tectonic Map of the World (1985), Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 207

Fig. 5. Interpretative tectonic map of the future orogen produced by the Australia-Asia collision. (a) illustrates the present day configuration. (b) illustrates the plate configurations projected 45 Ma into the future.

small back-arc basins of the Okinawa Trough, the truncation by a strike-slip (transform) fault (Sorong Japan Sea, the South Sulu Sea and the Celebes Sea, Fault terminates the double Celebes Arc). Arc young rifted margins behind arcs whose ages span polarity may change along strike length across the closing ages of the big oceans. transform or transpressional zones as in the Secondly, the collisions between rifted margins Philippines. Also substantial lengths of arcs may and arcs that lead eventually to subduction polarity have no magmatism for long periods of time (e.g, flips commonly generate short-lived Barrovian non-volcanic segments of the Andes; Dewey & metamorphic events in the substantial nappe sheets Lamb 1992) and individual arc volcanic centres formed beneath major obducted ophiolite com- may be separated by hundreds of kilometres as in plexes and accretionary prisms. There is a strong the Izu-Bonin-Marianas Arc. Thus, the absence of resemblance between the New Guinea ophiolite arc volcanic or plutonic rocks in a particular part of obduction and the related metamorphism of the a fossil arc system does not necessarily imply the Bismarck Range, with the Grampian Barrovian absence or cessation of subduction. metamorphism of the Scottish Highlands and the Fourth, remnant arcs may become isolated from obduction of the Shetlands Ophiolite (Dewey & the active arc by rapidly-opening back-arc oceanic Shackleton 1984). An important deduction from tracts but still showing magmatic ages very close to these early metamorphic complexes, adjacent to those of the active arc. When remnant arcs and early rifted margins, is that they formed and were related, only slightly younger, active arcs are, sub- also unroofed very rapidly, if diachronously, over sequently, packed together in collisional systems, great distances along the orogen. they may be regarded, too easily and incorrectly, as Thirdly, cognate arcs can vary greatly in length, separate, unrelated arcs with their own subduction polarity and mode of termination. Typically, they systems. have lengths of more than 1000km and may Fifth, face-to-face arc collision is occurring in terminate in triple junctions (Shikoku. Honshu. two areas of the west and southwest Pacific. In the Izu-Bonin arcs), by slip vector changes in rate and Molucca Sea, the east-facing northeast Sulawesi direction (south end of Yap/Palau Arc), or by arc is colliding with the west-facing Halmahera arc Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

208 c.R. VAN STAAL ET AL. with an intervening, growing collision zone. with six lines of cross-section. Prior to collision (1) Collision-induced uplift raised the tiny islands of the oceanic arc (Lough Nafooey Arc) with its fore- Mayu and Tifore above sea-level. Across the arc, supra-subduction-zone ophiolite and accre- Solomon Sea, the New Britain and Trobriand tionary prism is about to collide with the Laurentian Trenches, with opposed polarities, join to the east to rifted margin (Dalradian sediments with a Grenville form the south-facing San Cristobal Trench and, to basement) at c. 475 Ma in the Mid-Arenig. Progres- the west, to form the southward-vergent collision sively, from stages 2 to 5 in Fig. 6, the arc and zone between the Finistere Range and the main ophiolite assemblage and subjacent Dalradian mobile core of the Bismarck Range. The Trobriand nappe pile are driven across the Laurentian margin. Trench and associated fore-arc is being subducted The collisional tightening of the suture zone leads diachronously beneath the Finistere-South to subduction polarity flip (stages 2 to 6), Solomon arc; the double subduction zone beneath retrocharriage as seen in nappe root rotation in the eastern New Guinea can be seen in the seismicity Clifden steep belt, and southward thrusting, as (Pegler et al. 1995). along the Mannin Thrust (stage 5). However, for Sixth, a single arc, for example the Banda Arc, northward subduction to be established, the older may pass through an 180 ~ orocline, the same arc southward-subducting slab must break off to allow facing in opposite directions. If subsequently, the the asthenospheric gap through which the north- joining loop is obscured or tectonically removed, ward-subducting slab can be inserted (stages 3 to the opposing-facing segments could be regarded as 6). Just as subduction zones are likely to overlap, so two independent arcs. the break-off 'tear' in the southward subducting Seventh, arc-rifted margin collision, followed by slab is likely to propagate beneath the active arc polarity reversals, may be diachronous as, for collision zone (3 to 4). Thus, for a short, dia- example, between northern Australia and the Banda chronous, syncollisional, period, asthenospheric arc allowing propagation overlap that generates a mantle wells through the break-off tear beneath the zone with synchronous double subduction. deforming nappe-pile (4) and allows the syn- Collision, as in New Guinea during the Early kinematic intrusion of calc-alkaline basaltic Miocene, is commonly accompanied or predated by magma. ophiolite obduction, that pushed and carried a Eighth, the partitioning of oblique plate con- subjacent nappe pile across the old rifted margin to vergence into orthogonal thrust- and plate- form a short-lived Barrovian metamorphic nappe boundary-zone-parallel strike-slip components complex, the Bismarck Range. The subduction (Fitch 1972: Dewey 1980; Molnar 1992) may result polarity reversal allows the superposition of a post- in intra-arc ophiolite generation and obduction. For obduction magmatic arc to form on the deformed example, in New Guinea, southward ophiolite nappe complex. However, the line of subduction obduction in the Miocene was followed by a sub- polarity reversal (i.e. the new trench) is likely to duction polarity flip with the subsequent choking of substantially overlap with the colliding arc- the North Solomon Trench by the Ontong-Java continental margin zone as in Taiwan. This might Plateau. This led to partitioning of the oblique allow calc-alkaline mafic plutons to inject a ENE-convergence into orthogonal and arc-parallel deforming nappe pile, a relationship seen in the slip components; the arc-parallel strike-slip com- Caledonides of Western Ireland in Connemara. In ponent of the sinistral convergence was partly Connemara, substantial calc-alkaline mafic plutons accommodated by the opening of the oceanic, pull- were injected into shear zones within a northward apart Manus Basin. Thus, a new potential (towards and onto Laurentia) vergent Barrovian 'ophiolite" is being created in the Manus Basin nappe pile (Wellings 1998), driven by the collision within a post-collisional arc complex. Other small of a north-facing oceanic arc with the Laurentian ocean basins behind (South China Basin, South margin and the northward obduction of its Sulu Sea, Celebes Sea, Woodlark Basin and Coral ophiolitic fore-arc and accreting prism (Dewey & Sea) or within (Parece Vela Basin, North Fiji Shackleton 1984). Clearly it is difficult to relate, Plateau) arcs yield another potential source for directly and simply, the calc-alkaline plutonism to 'young' ophiolites that may be obducted. The southward subduction during this event because the critical criterion appears to be that obduction must deforming nappe pile was part of the subducting occur shortly after generation to allow the continental footwall. There may, however, be a detachment of a thin (<15 km) slice of oceanic better solution (Fig. 6) than propagating and crust and mantle above the 900~ isotherm (Dewey overlapping subduction zones to explain such & Shackleton 1984). Oceanic lithosphere older kinematic calc-alkaline magmatism. Figure 6 than c. 20 Ma is likely to be subducted rather than illustrates, schematically, map (a), sequential cross- obducted and, therefore, most of the back and intra- sectional (b), and block diagram (c) views of a arc basins mentioned above, with the exception of diachronous arc-continental margin collision zone the Manus Basin, are unlikely to form future Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 209

A B 1 2 3 4 5 6 ~r~ S N

I ~46o

'bceak-oft' gad sv sut>~uct,r,g s~ab , ~ ) l Non-Colhded J Colliding- fliplDtng Collided- ('lDre-tectonic calc-| (syn-teclonic calc- fhpped L .... ,.t >f alkaline Suite) / alkaline suite) :(posl-teclonic calc- / a~kahne suite

i - -t 41 C i

r/// r------"--'~~ 1

i 1 1

Fig. 6. Schematic tectonic model of diachronous, dextral collision between Laurentian margin and Bale Verte oceanic tract and the arc-polarity reversal. Note that subsequent strike-slip dissection of this collision/arc complex may juxtapose segments where the reversal in subduction had taken place at different times. (a) Map view. SV = slip vector of relative plate motion; SO --- component of slip vector orthogonal to plate boundary zone: SS = component of slip vector parallel with plate boundary zone. Lines of section 1-6 are those depicted in (b) and (c). (b) Sectional view illustrating the evolution of any one section through time. (c) Block diagram view illustrating that sectional views also vary spatially; i.e. all sections are not at the same stage of their temporal evolution (b) at the same time.

ophiolite complexes. Ninth, an important feature of Rifting of Rodinia, the opening of lapetus orthogonal and strike-parallel partitioning of and Gondwanan elements in the northern oblique plate convergence is that orogen or arc- Appalachians/British Caledonides normal thrusting is not a guide to relative plate motion, for which the partitioned arc-parallel Most recent palaeogeographic reconstructions strike-slip motion is a critical component. generally assume that the proto-Andean margin Tenth, massive relief changes are related with of Amazonia formed the conjugate margin to rapid lateral changes in tectonic environments. Laurentia during Late Neoproterozoic times For example, the extension and uplift of the (c. 750 Ma); a period when Laurentia formed the D'Entrecasteaux metamorphic core complex core of the Rodinia supercontinent (e.g. Hoffman results from the westward propagation of the 1991). The breakup of Rodinia supposedly took Woodlark Basin spreading centre. The resulting place in several stages (Dalziel 1992), which is geomorphic gradients yield rapid lateral facies consistent with the longevity of rift-related changes and slopes that drive instability and mass magmatism and formation of several large flow. When such already complicated geology is Neoproterozoic rift basins in Laurentia (e.g. Moine intensely strained during terminal collision, it and Dalradian). Relevant to this paper, however, is seems unlikely that unequivocal reconstructions the final break-up, which includes the separation of could be made. Laurentia from Amazonia at some stage between Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

210 C.R. VAN STAAL ET AL.

590 and 550 Ma. This separation together with the Caledonides that postdate the Grenville orogenic departure of Baltica led to the opening of the events are probably related to intracontinental Iapetus Ocean (Fig. 7). extension (Soper & Harris 1997). The first post- The opening of the Iapetus Ocean and the origin Grenville contractional event to affect the British of the Laurentian margin are best constrained by Caledonides was in the Early Ordovician (Dewey the onset of thermal subsidence-driven sediment- & Shackleton 1984). However, separation of ation of the continental margins (Bond et al, 1984). contractional from extensional structures is compli- Calibration of the rift-drift transition (e.g. Williams cated because older structures are commonly & Hiscott 1987) against the new Cambrian time reactivated during younger events. For example, we scale (Bowring et al. 1993; Tucker & McKerrow believe that the present Moine Thrust roughly 1995) would indicate that this took place between coincides with the original position of the Grenville 590 and 550 Ma in the northern Appalachians, front in the British Isles, separating a foreland with which is consistent with the youngest known U-Pb Lewisian basement from a basement involved in zircon ages of synrift magmatism along the Grenville orogenesis as witnessed by Grenville-age Laurentian margin (554 +4/-2 Ma, Tibbit Hill eclogites in the Glenelg inlier (Sanders et al. 1984). volcanic rocks in Quebec, Kumarapeli et al. 1989; Accordingly, the Grenville front became an 555 +3/-5 Ma Lady Slipper pluton, Cawood et aI. extensional fault during formation of the Moine 1996; 550 Ma+3/-2 Ma Skinner Cove volcanics in basins and was reactivated as a major structural Newfoundland, McCausland et al. 1997). The 590 front during Caledonian contractional events. Ma U-Pb zircon age of the Ben Vuirich granite in The voluminous, pre-600 Ma part of the Scotland (Rogers et al. 1989) and the deformation Dalradian Supergroup may not extend into it cuts have been reinterpreted as extension-related Newfoundland. Alternatively, it occurs only as a magmatism and deformation, respectively (Ryan & small cryptic unit within the Fleur de Lys Dewey 1991; Soper 1994). When this model is Supergroup (cf. Hibbard 1988) or, more likely, combined with the clear stratigraphic and structural much of the Fleur de Lys Supergroup is the strati- continuity between the southern Highland Group of graphic equivalent of the Late Neoproterozoic- the Dalradian Supergroup and the Cambrian Leny Cambrian southern Highland Group. Thus the Limestone (Harris 1969) we conclude that the Moine/Dalradian extensional basins narrowed rift-drift transition in Scotland also occurred during significantly somewhere between the British and the very latest pre-Cambrian. Hence, the final Canadian segments of the Laurentian margin and, breakout of Laurentia took place in a relatively hence, the presence or absence of Dalradian rocks short period (<40 Ma). Given these age constraints in a displaced piece of the Laurentian margin (such on the opening of the Iapetus Ocean, we suggest as Connemara) constitutes a tectonic tracer (cf. that the c. 576 Ma and 540 Ma ages of, respectively, Hutton 1987). The diminution in space and time, of an eclogite block in a m61ange of the Ballantrae the long-lived, wide and diffuse extensional basin Complex (Hamilton et al. 1984) and garnet complex of the Moine and Dalradian in the British amphibolite in the Highland Border Complex Caledonides (referred to herein as the Dalradian represent cooling ages of deep lithospheric rocks margin) to the short-lived, very narrow and sharply- exhumed during low-angle extensional faulting in a defined rifted margin of the northern Appalachians manner analogous to that observed along the with its very abrupt facies change from Cambrcr- Galicia margin (Boillot et al. 1988), rather than Ordovician platform carbonate to continental rise representing a Late Neoproterozoic/earliest strata immediately above Grenville crystalline Cambrian, syn-rifting obduction event (Dempster basement (referred to herein as the Grenville & Bluck 1991). All pre-Cambrian tectono-meta- margin) from Newfoundland to New York appears morphic events in the Scottish and Irish to have been of substantial tectonic importance in

Fig. 7. (a) Late Neoproterozoic paleogeography as proposed in this paper. Laurentia and Baltica are positioned after the compilation of palaeomagnetic data of Torsvik et al. ( 1996): Siberia after the compilation of Smethurst et ell. (1998); Gondwana is positioned after the data compilation of Meert & Van der Voo ~1997). No palaeomagnetic constraints are available for the position of Amazonia and our positioning here is based on geological grounds (see text). Av = Avalonia: Ga = Gander. (b) Middle to Late Cambrian paleogeography.Pc is the Precordilleran Laurentian fragment incorporated into South American part of Gondwana during the Late Ordovician (see Dalziel et al. 1994. 1996 and references therein). Note that opening of lapetus is related to the closure of the Brazilide Ocean. (c-f) Ordovician to Silurian palaeogeographic/tectonicevolution of the Iapetus Ocean. Laurentia is positioned throughout on the basis of compilation of palaeomagnetic data of Mac Niocaill & Smethurst (1994), Siberia after the palaeomagnetic compilation of Smethurst et ell. (1998). Avalonia. Baltica and Gondwana after Torsvik et al. (1996). Note that the connection between the various subduction zones is poorly known and speculative. ARM: Armorica. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 211 Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

212 c.R. VAN STAAL ET AL. controlling the structural style of the Laurentian from the late Neoproterozoic Avalonian arc ter- margin during Ordovician orogenesis, a point ranes and their Palaeozoic cover (Avalonia) that further discussed in the section on the tectonic make up the eastern part of the Avalon Zone in the model for the Baie Verte Oceanic Tract and Notre northern Appalachians. These terranes (Avalon Dame Arc (see below). Peninsula, Mira, Caledonia and Boston) experi- The time constraints on the rift-drift transition of enced, at least in part, a different geological history the Laurentian margin and, hence, the opening of during the Late Neoproterozoic and Cambrian the Iapetus Ocean have important implications for (Barr & White 1996). Ganderia basement has been palaeogeographic reconstructions. A key issue is correlated with the Rosslare and Coedana whether Amazonia assembled with the African Complexes (Fig. 3) in Ireland and Wales (van Staal cratons before or after the opening of the Iapetus et al. 1996a). An Amazonian provenance for Ocean. If this assembly took place before the Ganderia is consistent with the high Ordovician opening of the Iapetus Ocean, it implies the exist- palaeomagnetic latitudes (Liss et al. 1993), Early ence of another supercontinent in the late Ordovician cold-water, peri-Gondwanan faunas (S. Neoproterozoic (Dalziel 1992). However, such a Williams et al. 1995) and the chemical composition configuration is inconsistent with the present of its Lower Ordovician black shales (Fyffe & palaeomagnetic database because it predicts Pickerill 1993). Ganderia contains a late Neopro- palaeolatitudes of 75-85~ for the western margin terozoic to Early Cambrian (c. 600-530 Ma) conti- of Amazonia if Amazonia formed part of the nental arc and associated metamorphic rocks Gondwanan assembly at 550 Ma, whereas constructed on Mesoproterozoic and older crust Laurentia is situated at intermediate latitudes (Fig. typical of Amazonia that, overall, are markedly 7a), i.e. they cannot be adjacent at this time. These different from the dominantly juvenile Avalonian restraints are lifted if Amazonia accreted to the arc rocks (generally older than 600 Ma). The time African cratons after 550 Ma, an option that we of juxtaposition of Ganderia against Avalonia is prefer (Fig. 7b) and which was also recently poorly constrained. It may have occurred at any proposed by Hoffman (1996), who argues for a time between the Late Neoproterozoic (forming the closure of the Brazilide Ocean between 550 and continuous Avalonian superterrane or arc of 545 Ma at the latitude of the Kalahari Craton (Fig. Gibbons 1990: Horak et al. 1996) and the Devonian 7a). Hence, our interpretation implies the existence (Holdsworth 1994). The presence of the Acado- of two major landmasses at the end of the Baltic fauna in all terranes of the Canadian Avalon Neoproterozoic (Fig. 7a), Gondwana without Zone and an Arenig overstep sequence between the Amazonia and the remnants of Rodinia: namely Monian Terrane and Welsh Avalonia in the British Laurentia-Baltica-Amazonia-Siberia which separ- Isles suggest a linkage by at least the Late ated from each other between 600 and 550 Ma (see Cambrian, a linkage that remained in place during also Pelechaty 1996). most of the Palaeozoic such that Ganderia and The development of the eastern Gondwanan Avalonia behaved as one microcontinent during the margin of the Iapetus Ocean in the northern Ordovician and Silurian, albeit with significant Appalachians and British Caledonides is poorly internal modification by strike-slip motion. Hence, understood because there is no direct sedimentary for the sake of simplicity, we refer to this composite link between the Avalon and the immediately microcontinent as Avalon when discussing Palaeo- adjacent Gander Zones of Williams (1978). zoic events shared by both elements. However, isotope (Nd, Pb and O) and U-Pb zircon Closure of the Brazilide Ocean after 550 Ma also studies indicate that the Cambrian to Early provides a simple mechanism to explain the young Ordovician, continentally-derived, quartz-rich ages of the Late Neoproterozoic-Early Cambrian clastic sediments of the Gander Zone in the magmatic arc on Ganderia and its amalgamation northern Appalachians formed part of a with Avalonia. The arc and accompanying low- Gondwanan passive margin with an Amazonian pressure metamorphism simply formed as a result basement (van Staal et al. 1996a). Such a basement of subduction of the Brazilide Ocean beneath is probably exposed in the Brookville-Bras d'Or- Amazonia and were shut-off when the ocean was Hermitage Flexure terranes of the Canadian Avalon finally closed in the Early Cambrian rather than a Zone (Barr & White 1996), some tectonic inliers in termination as a result of transform activity as the Exploits Subzone in Newfoundland (S. O'Brien proposed by Murphy & Nance (1989). Therefore, et al. 1996), New Brunswick and New England opening of Iapetus is related to the closure of the (Figs 1 and 2); e.g. Late Neoproterozoic/Early Brazilide Ocean, a hypothesis recently also Cambrian Upsalquitch Gabbro (van Staal et al. proposed by Grunow et al. (1996). We speculate 1996a) and Massabesic Gneiss (Aleinikoff et al. that the Brazilide convergence system extended to 1995). Combined, this basement and its Gander northern Africa and may have been responsible for cover are referred to as Ganderia to distinguish it the Early Cambrian magmatic ages (Egal et al. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 213

1996) in Armorica, Iberia and Bohemia (Cadomia, Caledonides, its geology imposes important con- inset of Figs 1-3). These terranes may have formed straints upon tectonic modelling of the whole a promontory extending north from North Africa or northern Appalachian-British Caledonide Belt. a microcontinent peripheral to it (Van der Voo The Dunnage Zone of Newfoundland has been 1988). subdivided into two major subzones by H. Williams The timing of the Early to Mid-Ordovician et al. (1988), mainly on the basis of their marked separation of Avalonia and Ganderia from West Lower to Middle Ordovician stratigraphic and Africa and Amazonia, respectively, is not well faunal differences, namely the Notre Dame Sub- constrained by palaeontology (Cocks & Fortey zone with the Arenig Toquima-Table Head 1982) or by palaeomagnetism (Trench et al. 1992). (Laurentian), low-latitude fauna to the northwest Cladistic analyses have shown that, until the and the Exploits Subzone with an Arenig high- Arenig, shelf benthic faunas of Avalonia clustered latitude, peri-Gondwanan fauna to the southeast with those from Gondwana (Fortey & Cocks 1992), (e.g. see also S. H. Williams et al. 1992, 1995). The but, by the Late Llanvirn, separation was wide latter includes elements of the 'Celtic fauna' of enough (>1000 kin) to reduce faunal interchange Neuman (1984). The Celtic elements may never with Gondwana and allow interchange with have formed part of one coherent ecological group Laurentia and Baltica. Palaeomagnetic (Trench et because they lack internal cohesion. Of the 41 al. 1992) and Sm/Nd sediment provenance genera in the 'Celtic fauna' of Neuman (1984), 34 (Thorogood 1990) studies suggest that separation are endemic and occur in only one or two localities, had taken place earlier in the Arenig. The latter while the remaining seven genera include some that study shows that there was a drastic change from a occur also in Baltica and/or Laurentia. We agree, large continental provenance to a juvenile source in however, that most elements of the ~Celtic fauna" the Arenig to Ashgill Avalonian sediments. Sub- have no affinities with the near-equatorial, sidence analysis (Prigmore et al. 1997) suggests Laurentian Toquima-Table Head fauna (Harper et that it may correspond with a Late Cambrian to al. 1996) and in general are indicative of high(er) early Tremadoc or to an Arenig/Lianvirn interval. latitude, cold water conditions. The latter interval is favoured because the basal The tectonic boundary between the Notre Dame Arenig Stiperstones Quartzite in Shropshire, a and Exploits Subzones is the Red Indian Line (Fig. coarse shallow water sandstone, is probably 1), which is regarded commonly as the main correlative with the Armorican Quartzite of Iapetus suture zone in the Canadian Appalachians. Brittany (Noblet & Lefort 1990). An Arenig age of, The notion of one main suture zone in a complex probably rift-related, magmatism in west Avalonia ocean like Iapetus, which contained several represented by a within-plate rhyolite in the different basins and arcs (van Staal 1994) is, of Caledonian Highlands of southern New Brunswick course, unrealistic and further highlighted by the (U-Pb zircon age of 479 +__ 8 Ma, Barr et al. 1994) subdivision of the two major Dunnage subzones is consistent with this. Anorogenic bimodal into smaller subdivisions (H. Williams 1995). magmatism in Ganderian terranes in southern Nevertheless, the recognition of the Red Indian Newfoundland (Hermitage Flexure) and Nova Line as a fundamental tectonic boundary is very Scotia (Bras d'Or terrane) yielded Late Cambrian important in understanding the tectonic evolution to Tremadoc U-Pb zircon ages between 499 and of the Canadian Appalachians. The Red Indian 493 Ma (Dunning & O'Brien 1989; Dunning et al. Line had a complex and protracted history of 1990) and may be related also to the onset of movements ranging from thrusting to strike slip rifting. (Nelson 1981; Thurlow et al. 1992: Lafrance & Williams 1993; Lin et al. 1994) and should, therefore, not be regarded as a single, continuous Tectonic elements and evolution of fault but, rather, as a complex movement zone that Iapetus was reactivated continuously during most of the lifespan of the northern Appalachian Orogen. Most information on the pre-Silurian tectonic Marked contrasts in the Pb-isotope contents of history of the British Caledonide-northern the synvolcanic massive sulphide deposits of both Appalachian Orogen is preserved in the Dunnage subzones (Swinden & Thorpe 1984) and the Zone of Newfoundland (Williams 1978), where the palaeomagnetic data in general (van der Pluijm et post-Ordovician cover sequences are least exten- al. 1995) support this subdivision of the Dunnage sive and a large number of high quality U-Pb age Zone. The main lithological difference between the dates have been assembled as parts of detailed two major subzones is the presence of a Mid- mapping projects during the last ten years. Because Ordovician black shale and conformably overlying Newfoundland (Fig. 1) also forms the link between Late Ordovician/Early Silurian upward-coarsening the northern Appalachians and the British turbidites in the Exploits Subzone and their absence Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

214 r R. VAN STAAL ETAL. in the Notre Dame Subzone, which is characterized matic arc plutons, such as the Cape Ray and by a sub-Silurian unconformity. However, locally Hungry Mountain plutons, which range in compo- in the tectonic boundary zone between these two sition from tonalite to granite. The plutons contain subzones this stratigraphic contrast may not be xenocrystic zircons and have moderate to high ~Yd always diagnostic (Thurlow et al. 1992: Dec & values; hence they ascended through Laurentian Swinden 1994). crust (Whalen et al. 1987, 1997a,b; Dunning & On the basis of continuity of the specific Krogh, 1991; Dub6 et al. 1996). These post- geological characteristics, some elements of each Tremadoc arc plutons and associated volcanic of these subzones can be traced into Maritime rocks are referred to as the Notre Dame Arc. The Canada, New England (H. Williams 1995; van Bale Verte Oceanic Tract was progressively Staal 1994) and the British Isles (Colman Sadd et emplaced westward or northwestward onto the al. 1992a; Winchester & van Staal 1995), sug- Grenville margin during the Tremadoc to Llanvirn. gesting some degree of continuity of kinematically- Obduction onto the margin was diachronous and related belts along the length of the northern started in the Tremadoc opposite promontories on Appalachian-British Caledonide segment. the Laurentian margin. Obduction was synchronous with the generation of some supra-subduction-zone ophiolites (e,g. c. 484 ___ 5 Ma Bay of Islands Complex; Jenner et al. 1991; Cawood & Suhr Notre Dame Subzone 1992). Hence, the Bale Verte Oceanic Tract The Notre Dame Subzone as defined by H. contains pre- and syn-obduction ophiolites. Williams et al. (1988) is separated from the Furthermore, most of the calc-alkaline volcanic and Laurentian margin (Humber Zone) to the west by plutonic rocks postdate obduction, confirmed by a the Bale Verte Line and from the Exploits Subzone stitching relationship between Tremadoc arc to the east by the Red Indian Line (Fig. 1). How- plutons, and obducted ophiolites and their ever, the easterly-derived oceanic allochthons such underlying m61anges (Hall et al. 1994; van Staal et as the Bay of Islands Ophiolite Complex and al. 1996b; Whalen et al. 1997a,b). An implication associated ophiolitic m61anges emplaced on the of ophiolite obduction and interaction with an Laurentian margin rocks (Humber Zone) are also irregular margin is that obduction will be early and considered part of this subzone, i.e. they represent most pronounced opposite promontories, which large structural outliers (Fig. 1). This is consistent may explain the presence of numerous windows of with seismic reflection data that indicates that the metamorphosed continental margin rocks (Fleur de whole of the Notre Dame Subzone is allochthonous Lys Supergroup) south of the Little Grand Lake and structurally underlain by the Grenville margin fault (LGLF in Fig. 1) opposite the St Lawrence (Keen et al. 1986). Promontory. Such windows are less common or The ophiolitic rocks of the Notre Dame Subzone absent in the northern part of the Bale Verte have been subdivided into a western and eastern Oceanic Tract, although the isolated Ming's Bight belt (Fig. 1) on the basis of age and characteristics Group at the eastern end of Bale Verte may also be of the ophiolites (Colman-Sadd et al. 1992a). The a tectonic window rather than representing a close western belt contains mainly Upper Cambrian to flexure (cf. Hibbard 1982). Tremadoc (505-489 Ma) supra-subduction-zone The narrower, eastern ophiolite belt (e.g. ophiolites (e.g. St Anthony, Lushs Bight, Betts Annieopsquotch ophiolite) is composed mainly of Cove, Grand Lake) and juvenile ensimatic younger Arenig, fault-bounded, MORB-like ophio- volcanic-plutonic complexes, which commonly litic fragments, comprising various combinations of contain consanguineous, locally sheeted mafic pillow basalt, gabbros, sheeted dykes and plagio- dyke suites (Twillingate-Sleepy Cove-Moreton's granite (c. 481-478 Ma, Dunning et al. 1987). Harbour Group, Little Port) (Dallmeyer 1977; Mantle harzburgites have not been preserved any- Coish et al. 1982; Dunning & Krogh 1985; where in this belt, suggesting that these oceanic Swinden et al. 1997; Elliott et al. 1991; Kean et al. rocks represent scraped-off rather than obducted 1995; Cawood et al. 1996); characteristic are ophiolites. The uniform MORB character of the various combinations of boninite, IAT, trond- Annieopsquotch Ophiolite (Dunning 1987) indi- hjemite or plagiogranite and some MORB, OIB and cates that this piece of oceanic crust (layers 1,2 and alkalic basalt, which are collectively referred to 3) is potentially the only near-complete ophiolitic herein as the Bale Verte Oceanic Tract. Overall, fragment in the Newfoundland Appalachians that these characteristics suggest formation in an does not have a supra-subduction-zone character. extensional regime. The supra-subduction-zone The eastern ophiolite belt is, spatially, closely assemblages are overlain by Arenig to Llanvirn associated with coeval or slightly younger upper calc-alkaline volcanic rocks and are intruded by Tremadoc to upper Arenig (484-473 Ma) calc- Lower to Upper Ordovician (488-456 Ma) mag- alkaline volcanic and arc tholeiitic rocks of the Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 215

Roberts Arm, Cottrell's Cove, Chanceport and overprinted during a subsequent polyphase folding Buchans Groups (Dec et al. 1997). The original and faulting history in part related to dextral relationships between the ophiolitic rocks and transpression (e.g. Lafrance & Williams 1992). A volcanic arc rocks are poorly defined. This results. major unconformity, representing uplift and at least in part, from a poor understanding of the exhumation, separates the Mid-Ordovician and structural relationships between the ophiolitic older rocks from a generally thin cover of a bodies and associated volcanic rocks. However, discontinuous and probably diachronous sequence fault-bounded slivers of MORB pillow basalts and of Late Ordovician to Late Silurian (c. 453-422 cherts are, locally, structurally interleaved with Ma) dominantly terrestrial to shallow marine calc- island arc tholeiites (Dec & Swinden 1994) alkaline to alkaline volcanic and sedimentary rocks whereas, in the vicinity of the Buchans Mine and (Whalen et al. 1987; Chandler et al. 1987; Dub6 et Red Indian Lake, seismic and structural studies al. 1996) throughout the Notre Dame Subzone. have shown that the ophiolitic Skidder Basalt Sparse low-latitude, Laurentian, faunas occur in (Pickett, 1987) structurally underlies the Late sediments of both the Baie Verte Oceanic Tract and Arenig calc-alkaline volcanic rocks of the Buchans the Annieopsquotch Accretionary Tract (S. H. Group in a complex, Mid-Ordovician, east-directed Williams et al. 1995 and references therein). The thrust wedge (Thurlow et al. 1992). It is uncertain low-latitude faunas are consistent with the near- whether the ophiolitic bodies form part of the equatorial palaeomagnetic latitudes (c. l l~ basement to the Roberts Arm belt arc volcanic Johnson et al. 1991) obtained for the Upper rocks or the relationships are dominantly structural. Cambrian or Tremadoc Moreton's Harbour Group Negative eNd values for most of the calc-alkaline (Dec et al. 1997), which is regarded generally to volcanic rocks indicate involvement of continental form part of or to overlie the Upper Cambrian crust in their petrogenesis (Swinden et al. 1997). Sleepy Cove-Twillingate trondhjemite complex Therefore, we suspect a dominantly structural (Williams & Payne 1975) of the Baie Verte Oceanic relationship between most of the ophiolitic Tract. This palaeolatitude overlaps within error fragments and the bulk of the arc volcanic rocks. with the Early Ordovician latitude of the The Red Indian Line deformation zone in central Appalachian margin of Laurentia (c. 18~ and Newfoundland is marked by a section several suggests that the Baie Verte Oceanic Tract formed kilometres thick of structural mdlange and very close to the Laurentian margin. On the other deformed Mid-Ordovician siltsone, limestone and hand. the late Arenig volcanic rocks of the Roberts basalt (Harbour Round Formation) between the Arm and Chanceport groups yielded mainly overlying, intensely-imbricated Buchans Group intermediate latitudes (c. 30~ van der Voo et al. and Skidder Basalt and the structurally underlying 1991). Whether this difference is real and repre- Victoria mine sequence of the Exploits Subzone. sents significant Arenig latitudinal separation The overall character of this structural assemblage (15-20 ~ between the Baie Verte Oceanic Tract and suggests an accretionary complex. The Harbour Annieopsquotch Accretionary Tract, a southerly Round Formation was grouped into the Notre drift of Laurentia during the Arenig to Llanvim or Dame Subzone by Thurlow et al. (1992), although a combination of both is unclear. Dec and Swinden these rocks are atypical of this subzone. O'Brien (1994) claim that the oldest parts of the Cottrell's (1991) described the Red Indian Line in Notre Cove and Chanceport groups of the Annieo- Dame Bay as a 2-3 km wide zone of mfilange (Sops psquotch Accretionary Tract locally conformably Head-Boones Point Complex; Nelson 1981) overlie the Moreton's Harbour Group, which containing variably-sized exotic blocks including suggests that the Chanceport fault is not a terrane basaltic rafts with N-MORB compositions (Dec & boundary (cf. Lafrance 8,: Williams 1992). This Swinden 1994). The mdlange is imbricated with suggests a southerly drift of Laurentia to parts of the arc volcanic rocks of the Buchans- intermediate latitudes (20-25~ rather than a Roberts Arm belt and metamorphic assemblages major separation in the Arenig to Llanvirn. Such a with actinolite and pumpellyite occur in rocks of drift is permitted, within analytical error, by the the Chanceport Group close to the Lukes Arm Fault present palaeomagnetic dataset of Laurentia (Mac (S. Armstrong pers. comm.) These types of Niocaill & Smethurst 1994; Torsvik et al. 1996) relationship suggest that the eastern ophiolite belt (Fig. 8). One should also take into account that the and associated arc rocks were incorporated into an Laurentian margin was expanding southwards by east-facing subduction/accretionary complex, here the addition of the accreted Baie Verte Oceanic referred to as the Annieopsquotch Accretionary Tract. Other supporting arguments in favour of Tract. High-level thrusting, largely accommodated southerly drift rather than latitudinal separation of by m61ange formation, represents the earliest both ophiolite belts are (i) low-latitude faunas in history of the Red Indian Line deformation zone both the Bale Verte Oceanic- and Annieopsquotch but these structures have been complexly Accretionary Tract suggest development near the Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

216 c.R. VAN STAAL ET AL.

Equator -- Newfoundland, most oceanic fragments of the southern Quebec Ophiolite Belt are structurally intercalated with m61ange (St Daniel M61ange) and

-20 disconformably overlain by Llanvirn-Caradoc chromite-bearing flysch of the Magog Group. In this respect, the southern Quebec Ophiolite Belt resembles the Bay of Islands Ophiolite Complex of -40 Palaeo/atitude of Laurentian margin the Humber Arm Allochthon, with the syn-obduc- tion, Middle-Ordovician Crabb Brook Group (Casey & Kidd 1981) representing the -60 ' I ' I ' I ' I Newfoundland equivalent of the Magog Group. A 360 400 440 480 520 plagiogranite of the ophiolitic Thetford Mines ..4_~ Geological Time Complex yielded an age 479 _+ 3 Ma (Dunning & Pedersen 1988), which makes the latter the Fig. 8. Predicted palaeolatitude of the Newfoundland youngest known ophiolite in the southern Quebec margin of Laurentia (based on a present-day reference locality at 49~ 57~ versus time. The predicted Ophiolite Belt, provided the plagiogranite is palaeolatitudes are derived from the palaeomagnetic consanguineous with the remainder of the ophiolite compilation of Mac Niocaill & Smethurst (1994). Errors complex. Hence, the ophiolite may be older. on the predicted palaeolatitudes are of the order of 10~ Considering the position of the Thetford Mines within the southern Quebec Ophiolite Belt and the overlap within error between the age of the plagiogranite and a gabbro of the Bay of Islands Laurentian margin, and (ii) oceanic elements of Ophiolite Complex (484___5 Ma), we interpret the both tracts have been intruded by Tremadoc to Thetford Mines Complex as a syn-obduction Llanvirn magmatic arc plutons and/or include ophiolite. The volcanic-plutonic domains of the coeval calc-alkaline volcanic rocks that formed on Middle Ordovician Ascot Complex (460-455 Ma, Laurentian crust (Dunning et al. 1987: Swinden et David et al. 1993), which lack m61anges (Tremblay al. 1997; Whalen et al. 1997a,b; Dec et al. 1997). & St Julien 1990), are the Quebec analogues of the These relationships suggest that by late Tremadoc/ calc-alkaline Middle Ordovician volcanic/mag- early Arenig time all calc-alkaline rocks formed matic arc rocks of the Notre Dame Arc. Nd-isotope part of one coherent magmatic arc (Notre Dame data also indicate development on Laurentian crust Arc) built on the Laurentian margin. (Tremblay et al. 1994). The southern Quebec Ophiolite Belt continues into the Rowe-Hawley Correlatives of the Baie Verte Oceanic Belt in New England (Stanley & Ratcliffe 1985) with its relative abundance of boninite and Tract outside Newfoundland primitive arc volcanic rocks (Kim & Jacobi 1996). The Notre Dame Subzone can be traced into Intrusive arc plutons (488-462 Ma, Karabinos & Quebec and New England (e.g. Williams & St Tucker 1992; Williamson & Karabinos 1993; Julien 1982). Age and lithological equivalents of Karabinos et al. 1993) suggest that the oceanic the Baie Verte Oceanic Tract in southern Quebec elements are also Lower Ordovician or older in age, are represented mainly by the Southern Quebec consistent with Late Cambrian to Tremadoc Ophiolite Belt (Fig. 2). This belt comprises supra- 4~ ages (505--490Ma) of blueschists subduction-zone ophiolites that include a signifi- developed in ophiolitic rocks (Laird et al. 1993). A cant amount of boninitic rocks (Oshin & Crocket, characteristic feature of the boundary zone between 1986; Laurent & Hebert 1989). IAT and boninites, the oceanic element of the Notre Dame Subzone preserved as structural lenses in m61ange in the and rocks of the Humber Zone from Newfoundland Ascot structural complex (Tremblay & St Julien to Massachussets is a belt of continental margin- 1990; Tremblay 1992), are also considered derived albite schists tectonically interleaved with equivalents of the Baie Verte Oceanic Tract. In ultramafic pods (e.g. Rowe and Bennett schists, general, the complex magmatic assemblages Fleur de Lys Supergroup) in New England, Quebec preserved in the oceanic complexes closely and Newfoundland (Stanley & Ratcliffe 1985; resemble those in Newfoundland. A Late Cambrian Hibbard 1988). U-Pb zircon age of 504 + 3 Ma (David et al. 1993) Continuation of the Bale Verte Oceanic Tract for the Mt Orford Ophiolite overlaps with the oldest into the British Caledonides (Fig. 3) is also ages from Newfoundland. As in Newfoundland, probable (Dewey & Shackleton 1984; Colman- obduction probably started in the Tremadoc (Pinet Sadd et al. 1992a). The oceanic structural & Tremblay 1995; Whitehead et al. 1996). Unlike fragments of the Highland Border Complex, most parts of the Baie Verte Oceanic Tract in Midland Valley (Ballantrae) and Northwestern Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 217

Terrane (South Mayo and Tyrone) are associated Early Ordovician structures from the Dalradian with sedimentary rocks containing Laurentian margin into ophiolitic m61anges of the Clew Bay faunas (Cocks & Fortey 1982; S. H. Williams et al. Complex supports this conclusion (Harris 1995). 1995) and are characterized by low palaeomagnetic South Mayo is unique, compared with its equiv- latitudes (Torsvik et al. 1990). Of these three alents in the northern Appalachians, in preserving a terranes, the southern Midland Valley most closely continuous section from a primitive north-facing resembles the geology of the Notre Dame Subzone oceanic, Tremadoc-Arenig arc to a well developed of Newfoundland. The supra-subduction-zone fore-arc basin/accretionary complex (Dewey & Ballantrae Ophiolite Complex has a composite Ryan 1990; Clift & Ryan 1994), characterized by nature like those in the Baie Verte Oceanic Tract in sporadic blueschists and other high-pressure Newfoundland. Incorporation of volcanic rocks of assemblages (Gray & Yardley 1979). Another diverse tectonic settings may provide the explan- critical key to the geology of western Ireland is the ation for the wide variety of ages, which range recognition of an Arenig to late Llanvirn (c. 480- almost continuously from the Late Cambrian to late 460 Ma, Cliff et al. 1996) continental arc in the Tremadoc (505--484Ma; Bluck et al. 1980; Dalradian margin, syntectonic with D2-D3 Hamilton et al. 1984). Furthermore, the Ballantrae Grampian deformation and low-P metamorphism Ophiolite Complex, like the Bay of Islands (Yardley & Senior 1982; Leake 1989). Recent Ophiolite Complex, was obducted onto the margin detailed structural and U-Pb studies of the Cashel- during the Arenig. Obduction was followed Currywongaun calc-alkaline gabbros in Connemara immediately by intrusion of Arenig diabases with showed they also intruded syn-D2 at c. 468- island arc affinity (Holub et al. 1984). In general, 460Ma (Wellings 1998; Friedrich et al. 1997), the Arenig to Caradoc history of the Midland which is similar to the c. 468 Ma U-Pb zircon age Valley is dominated by formation of a south-facing of the probably correlative Aberdeenshire gabbro volcanic/magmatic arc (Midland Valley arc; Bluck suite in Scotland (Rogers et al. 1994). The 471 Ma 1984), which was unroofed during the Llanvirn- ophiolite-cutting tonalite pluton in Tyrone probably Caradoc. The Midland Valley arc terrane also also forms part of this continental arc. The age and appears to have a Grenville rather than a Dalradian geological characteristics of the Dalradian-margin basement (Aftalion et al. 1984; Parnell & arc and the Midland Valley arc bear a strong resem- Swainbank 1984; Upton et al. 1984), suggesting blance to the Notre Dame Arc developed on the that this terrane represents a far-travelled piece of Grenville margin in Newfoundland after initial the Grenville margin of Laurentia, i.e. the Midland ophiolite obduction in the Tremadoc and early Valley was carried by post-Ordovician, margin- Arenig. Furthermore, in all three areas parallel, sinistral strike slip faults into its present (Newfoundland-Ireland-Scotland), unroofing of position to the south of rocks underlain by the magmatic arc and associated metamorphic Dalradian basement (Dewey & Shackleton 1984; rocks started in the Mid- to Late Ordovician, Hutton 1987; Ryan et al. 1995), consistent with reflecting a once continuous tectonic setting and provenance studies that do not show a sedimentary kinematic history. linkage with Dalradian rocks until the Devonian (Bluck 1984). Early Ordovician northward-directed ophiolite obduction also characterized the Tectonic model for the Baie Verte Oceanic Dalradian part of the Laurentian margin. Shetland Ophiolite plagiogranite yielded a high-quality Tract and the Notre Dame Arc U-Pb zircon age of 492 _+ 3 Ma (Spray & Dunning The Late Cambrian-Tremadoc (508-486Ma) 1991), whereas the underlying hornblende schist oceanic fragments of the Baie Verte Oceanic Tract gave a slightly older 40Ar/39Ar plateau age of 503 _+ and correlatives throughout the northern 6 Ma (Flinn et al. 1991). This suggests that the Appalachians and British Caledonides appear to be plagiogranite either formed off-axis during the characterized mainly by an assemblage of supra- earliest stages of obduction in an oceanic realm or subduction-zone rocks which commonly include the older ages of the hornblende schist represent a IAT, boninite, MORB and minor bodies of juvenile remnant of an oceanic transform that was later silicic igneous rocks. Any tectonic model for the tectonically juxtaposed with a supra-subduction- Baie Verte Oceanic Tract has to explain the zone ophiolite. following: (1) it appears to have formed close to the In Tyrone, obduction took place before 471 Ma, Laurentian margin in an extensional setting; (2) the age of a cross-cutting tonalite pluton (Hutton et obduction took place shortly after formation; some al. 1985), whereas sediment provenance studies ophiolites formed while obduction was in process suggest that ophiolite obduction onto the Dalradian at nearby promontories; (3) the oldest oceanic margin was well advanced by late Arenig times in fragments locally contain mafic xenoliths with South Mayo (Dewey & Ryan 1990). Continuity of strong pre-entrainment foliations (e.g. Williams & Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

218 C. R. VAN STAAL ET AL.

Payne 1975) and were deformed during trans- foliation in the 507 Ma Twillingate trondhjemite current shear prior to and/or during obduction (e.g. (Williams & Payne 1975; Elliott et al. 1991) Elliott et al. 1991); (4) coeval arc, fore-arc or back- represent the oldest known ages in the Baie Verte arc volcanic-sedimentary assemblages are rare or Oceanic Tract and correlatives in the northern absent; (5) stitching Early Ordovician arc plutons Appalachian-Caledonian Orogen and may ascended through 'Grenville' crust; and (6) high represent remnants of such oceanic transform pressure assemblages (e.g. blueschists) occur in faults. The transform fault-subduction intiation several places in rocks structurally beneath the model could also explain the anomalous Late obducted ophiolitic rocks (New England, Laird et Cambrian ages (503 Ma) of hornblende schists in al. 1993; Quebec, Trzcienski 1976; Newfoundland, the Unst Ophiolite in Shetland (Flinn et al. 1991) Jamieson 1977; Ireland, Gray & Yardley 1979; that predate the Tremadoc plagiogranite. A dextral Harris 1995). obliquity of convergence, after the transform fault These criteria do not favour recent tectonic became a subduction zone, is consistent with the models that place the Baie Verte Oceanic Tract in structural data of Elliott et al. (1991), Goodwin & an arc-back-arc setting above a westward-directed Williams (1990) in other parts of Newfoundland, subduction zone underneath Laurentia, with the and Harris (1995) in western Ireland. However, we back-arc basin closing in Early to Mid-Ordovician deviate from the Karson & Dewey (1978) model, times (e.g. Swinden et al. 1997; van der Pluijm et where it concerns generation of the relatively al. 1995). Particularly, formation of syn-obduction young (c. 484 Ma), syn-obduction ophiolite of the ophiolites, blueschists and eclogites are hard to Bay of Islands Ophiolite Complex in an active reconcile with closure of a very young back-arc ridge segment of the transform fault. McCaig & basin. Furthermore, boninite occurs dominantly in Rex (1995) presented evidence that deformation the fore-arc areas of modern oceanic subduction and cooling of the Little Port Complex in the systems and is thought to be indicative of western Lewis Hills had taken place before, not subduction zone initiation (Casey & Dewey 1984; during, generation of the intruding ultramafic- Stern & Bloomer 1992). Some have suggested that mafic rocks of the Bay of Islands Ophiolite formation of the Bale Verte Oceanic Tract boninites Complex as is required by the Karson & Dewey took place during intra-arc rifting (e.g. Coish et al. (1978) model. Instead, we propose that the Bay of 1982). However, where boninite occurs in a modern Islands Ophiolite Complex formed in a dextral back-arc position such as in parts of the Lau basin, transtensional basin (rhombochasm) in the over- it is either regarded as products of fore-arc riding plate above a re-entrant after collision had lithosphere remelted during propagation of 'back- started. The former existence of such a re-entrant arc' spreading centres into the fore-arc, e.g. where may still be outlined by the tight bend in the Baie initial rifting of the Tonga arc occurred trenchward Verte Line near Grand Lake (marked by the trace of of the arc volcanic front, or by remelting of pre- the Grand Lake Complex) where it changes from a existing refractory back-arc mantle (Clift 1995). dominantly northeast trend into an east-west strike The Tonga Arc-Lau marginal basin system is, in marked by the Little Grand Lake Fault (LGLF in some ways, a pertinent example, because the Fig. 1). A possible modern analogue of ophiolite magmatic history of the Tonga Arc-Lau basin, generation within a colliding or post-collisional arc varying from alkaline volcanism at the still active complex is the Manus Basin north of New Guinea remnant arc to MORB or MORB transitional to IAT (Fig. 4). The northwards translation of the Bay of in the rift zone (Clift 1995) shows similarities with Islands Ophiolite Complex during its subsequent the magmatic assemblages preserved in the Baie Arenig emplacement onto the Laurentian margin Verte Oceanic Tract, although no good equivalent (Suhr & Cawood 1993) may explain the conflicting of the pre-rifting Tonga arc is present. Hence, a dextral and sinistral kinematic indicators observed rifted arc setting is not a satisfying analogue for the by McCaig & Rex (1995) in the Little Port Baie Verte Oceanic Tract. Complex of the western Lewis Hills, because it We concur that the original model of Karson & acted then as a sinistral lateral ramp. The more Dewey (1978), subduction initiation in a dextral MORB-like character of the Bay of Islands oceanic transform fault or fracture zone close to the Ophiolite Complex and the Arenig Snooks Arm Laurentian margin at c. 508 Ma, is the most appro- Group in Newfoundland, compared with the more priate to explain the formation of the Baie Verte typical supra-subduction-zone geochemistry of Oceanic Tract when comparing these rocks with most older basalts of the Baie Verte Oceanic Tract recent supra-subduction-zone equivalents (e.g. (Jenner & Fryer 1980; Jenner et al. 1991) is Stern & Bloomer 1992; Bloomer et al. 1995). The consistent with this model. syn-magmatic shear zones in the Little Port One of the characteristics of modern analogues Complex (505+3/-2 Ma; Jenner et al. 1991) and of supra-subduction-zone magmatism is the amphibolite xenoliths with a pre-entrainment absence of a well-defined volcanic front; instead, Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 219 based on studies in the Izu-Bonin-Mariana system, 1992; Williamson & Karabinos 1993). Such subduction initiation created a wide area (>200 kin) plutons and associated volcanic rocks occur from of extension and widely-distributed zone of fore- the British Isles to New England and signal a arc magmatism (infant arc magmatism of Stern & change in polarity of subduction along the Bloomer 1992) in the upper oceanic plate. Such Grenville and Dalradian margins in the Early juvenile supra-subduction-zone magmatism may Ordovician (Fig. 9). An alternative model, that the have lasted until c. 489 Ma in Newfoundland. Notre Dame Arc formed on a little-travelled, rifted- Equivalent supra-subduction-zone magmatism may off sliver of the Grenville margin is thought to be have lasted longer in the British Isles although geo- unlikely. Such a model requires that the auto- chronological control does not allow much chthonous, well-developed passive margin in diachroneity. With the possible exception of the Western Newfoundland and New England formed amphibolite xenoliths in the Twillingate trond- partly in a back-arc environment, for which there is hjemite, evidence for older, pre-subduction, no evidence, and does not provide a satisfactory oceanic crust on which the forearc supra-sub- explanation for the abrupt change from supra- duction-zone magmatism was built has not been subduction-zone oceanic to continental arc mag- detected yet in the Baie Verte Oceanic Tract; how- matism in the late Tremadoc to Arenig. ever, neither has much clear evidence been found The arc-polarity reversal along the Laurentian for pre-subduction oceanic crust in the Bonin- margin must have been very rapid and diachronous Mariana or Tonga forearcs. This suggests that most because westward-directed obduction and ophiolite existing oceanic crust was, largely, replaced or formation in parts of the Baie Verte Oceanic Tract displaced during supra-subduction-zone forearc postdate or overlap with generation of the oldest magmatism (Bloomer et al. 1995). Obduction of members of the Notre Dame Arc elsewhere. This is the Baie Verte Oceanic Tract onto promontories in particularly apparent in western Ireland where the the Laurentian margin must have started in the polarity reversal recorded by the synorogenic arc early Tremadoc between 500--488 Ma (Dallmeyer magmatism and related metamorphism in 1977; Cawood & Suhr 1990; Dewey & Ryan 1990; Connemara (480-465 Ma; Cliff et al. 1996) over- Pinet & Tremblay 1995; Swinden et al. 1997) soon laps, in part, with lower Arenig oceanic arc mag- after subduction was initiated, presumably while matism and fore-arc emplacement onto the fore-arc extension, due to rapid hinge retreat of the Dalradian margin in the adjacent South Mayo sinking slab, was still active and before stable Trough (see Figs 3 and 6). The first possible lithospheric subduction was established (see Stern elements of continental arc volcanism in the South & Bloomer 1992); hence supra-subduction-zone Mayo trough are the Llanvirn Mweelrea ignim- fore-arc magmatism may have had a lifespan of up brites (Figs 3 and 6); hence the juxtaposition of to c. 20 Ma in the Baie Verte Oceanic Tract, which Connemara with the South Mayo Trough took is not unlike the period of supra-subduction-zone place as a result of post-Llanvirn but pre-upper magmatism in the Izu-Bonin-Mariana fore-arc Llandovery transcurrent movement (Dewey & (10-15 Ma; Bloomer et al. 1995). The polarity Ryan 1990). The first possible sedimentary link direction of subduction associated with the between Connemara and the South Mayo Trough is Baie Verte Oceanic Tract therefore must have southerly-derived high-grade metamorphic detritus been towards the east or southeast in present in the Late Ordovician Derryveeny conglomerate coordinates. (Dewey & Ryan 1990). It is logical to link this The oldest known Notre Dame Arc pluton that detritus to strike-slip docking of the Connemara cuts obducted ophiolites of the Baie Verte Oceanic terrane because this terrane experienced rapid uplift Tract in Newfoundland is the 488 + 3 Ma (Dub6 et and erosion after 460 Ma (Elias et al. 1988). al. 1996) Cape Ray Granodiorite (Hall et al. 1994). After the polarity reversal was complete along This pluton, which ascended through Grenville the length of the orogen in the Arenig, obduction margin crust (Whalen et al. 1997b) is no exception onto the margin continued into the Mid- as similar, stitching continental arc plutons of Ordovician. Mutual cross-cutting relationships Tremadoc to Arenig age have been identified between Ordovician arc plutons, folds and throughout the Notre Dame Subzone of amphibolite-facies shear fabrics in the Fleur de Lys Newfoundland (Whalen et al. 1997a), while rocks of SW-Newfoundland also corroborate that Swinden et al. (1997) even describe crustally the Notre Dame Arc plutons were roughly syn- contaminated high-Mg andesite dikes with tectonic with the Mid-Ordovician Taconic Orogeny 4~ ages between 501 and 490 Ma that cut (van Staal et al. 1996b). Therefore, the Taconic deformed rocks of the Lushs Bight ophiolite. The west-directed thrusting mainly represents back- ages of the bulk of the Notre Dame Arc plutons fall thrusting behind an east or south-facing, compres- in the same range as the oldest arc plutons in New sive Notre Dame arc ( Bird & Dewey 1970) (Figs 6 England (c. 488-484Ma; Karabinos & Tucker and 9). Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

220 C.R. VAN STAAL ET AL.

Fig. 9, Early to Middle Ordovician tectonic evolution of the Grenville margin. Note that west-directed transport and associated deformation during middle Arenig to Caradoc in (b) is due to backthrusting. Evidence for Lower to Middle Ordovician, east-directed thrusting in the Notre Dame Subzone near the western boundary with the Annieopsquotch accretionary tract is locally well preserved (e.g. Hungry Mountain Thrust, see Thurlow et al. 1992 and references therein). Also note that the syntectonic Notre Dame Arc plutons are both deformed by these structures and also cut them (Hall et al. 1994). The general principles of this model are also applicable to the Dalradian margin although the deformation history generally is more complex.

In the northern Appalachians, the Early to Mid- whether it did or did not travel across the Moine/ Ordovician westward translation of the Baie Verte Dalradian complex onto the Durness carbonate Oceanic Tract led to a rather simple, narrow, domi- platform northwest of the Moine Thrust (Fig. 3) or nantly westward-vergent orogenic belt beneath the any successor foreland basin above it. The ophiolite and the transport of the latter together structural and metamorphic style and sequence of with subjacent continental rise rocks onto a the Moine and Dalradian are generally more foreland basin successor of the carbonate platform. complex and intense than the Fleur de Lys rocks In contrast, a more complex structural history is and its equivalents in Quebec and New England, preserved in the Dalradian margin of the British although there are also parallels (e.g. Tremblay & Isles. Although a suture is recognized in western Pinet 1994). Structural polarity and bulk vergence Ireland (Clew Bay Zone) and a high-level ophiolite are variable both on the Dalradian margin of the nappe and/or accretionary prism was probably Caledonides and Grenville margin of the emplaced above the Dalradian (Dewey & Appalachians but the rapid lateral changes from fiat Shackleton 1984), there is no evidence as to belts to intensely-deformed steep belts charac- Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 221 teristic of the central Highlands of Scotland, are represent their Taconic suture, implying that the rare or absent in the northern Appalachians. Bronson Hill Arc forms part of the Exploits Sub- Diachroneity and changes in style of structures zone, a conclusion entirely consistent with the within one type of margin are another complication. presence of Caradoc to Ashgill black shales and For example, most of the Early Grampian defor- volcanics. The Boil Mountain Ophiolite Complex mation (D1-D3) in Connemara, westem Ireland is in tectonic contact to the west with the Chain (Fig. 3), was characterized by northwards-directed Lakes Massif (Boudette 1982). Seismically, the tectonic transport, the main southward retro- Chain Lakes Massif forms basement to the Mid- charriage taking place during the Mid- to Late Ordovician Ascot Complex (Spencer et al. 1986) Ordovician (D4). On the other hand, tectonic and is intruded by Middle Ordovician, probably transport was both to the north and south during the consanguineous, arc-like plutons (e.g. Attean Early to Mid-Ordovician (D2) in the Dalradian of pluton: 463 Ma, Holtzman et al. 1996). The Chain the Grampian Highlands (Krabbendam et al. 1997). Lakes Massif contains typical Laurentian Grenville These contrasts suggest a complex response to the detrital zircons and experienced Taconic high-grade transpressional strains (D2-D4) superimposed on contact metamorphism (U-Pb monazite of 468 _ the initial obduction and collision structures (D1) 2 Ma) from the arc plutons; hence, it probably during and following the polarity reversal along the represents an exposed window of metamorphosed Dalradian margin. The marked ductile behaviour rocks of the Laurentian margin (Dunning & and thickening of the Dalradian contrasted with the Cousineau 1990; Trzcienski et al. 1992). These Grenville margin is perhaps due to a combination characteristics strongly resemble the relationships of a higher degree of underthrusting and converg- observed in structural inliers of Fleur de Lys ence with a more extended part of the Laurentian clastics in Southwest Newfoundland (Fig. 1), which crust with its thick sedimentary cover of the Moine/ experienced amphibolite and locally even granulite Dalradian. Syntectonic magmatic arc activity facies, Taconic metamorphism coeval with continued until the Caradoc both in the northern intrusion of the syntectonic Notre Dame Arc Appalachians and British Caledonides, after which plutons (Currie et al. 1992; Hall et al. 1994; van the Notre Dame-Midland Valley arc experienced Staal et al. 1996b). Thus, the bulk of the juvenile uplift, exhumation and a temporary magmatic shut- oceanic elements of the Bale Verte Oceanic Tract off. This coincides with accretion of the Bronson and equivalents in Quebec and New England were Hill-Popelogan-Victoria-Longford Down island obducted onto the Laurentian margin before arc system with the Notre Dame Arc approximately formation of the Early to Mid-Ordovician Notre along the present day trace of the Red Indian Line Dame Arc plutons and related volcanics along the (van Staal 1994; see below). whole length of the northern Appalachians, i.e. the Tremadoc and older elements of the Baie Verte Correlatives of the Annieopsquotch Oceanic Tract represent a rootless oceanic ldippe first obducted before the classic Mid-Ordovician Accretionary Tract in New England and Taconic Orogeny. Absence of the Attean contact the British Caledonides and their tectonic metamorphism in the Boil Mountain Ophiolite setting Complex and tonalite in the Chain Lakes Massif suggest that tectonic juxtaposition of the ophiolite Equivalents of the Annieopsquotch Accretionary complex with the massif, took place after 468 Ma Tract are less well defined in New England. A (Trzcienski et al. 1992; Kusky et al. 1997). recent U-Pb zircon age of 477 -4- 1 Ma for a tonalite Equivalents of the Annieopsquotch Accretionary in the Boil Mountain Ophiolite Complex in Maine Tract in the British Caledonides are not well (Kusky et aL 1997) suggest that the latter and on- defined and their existence is uncertain. Colman strike soapstone belt, immediately west of the Saddet al. (1992a) and S. H. Williams et al. (1992), Bronson Hill anticlinorium in the Connecticut correlated all of the Southern Uplands in Scotland Valley in New Hampshire and Massachusetts (Fig. with the Exploits Subzone in Newfoundland on the 2; Lyons et al. 1982) are their probable equivalents basis of the presence of Caradoc black shale, in New England. If the tonalite is consanguineous strictly adhering to the original definition of H. with the rest of the Boil Mountain Ophiolite Williams et al. (1988). However, we suggest that, Complex, the latter cannot represent the root zone although there may be a general spatial correlation of the older southern Quebec ophiolites as proposed along a zone that contains both the Exploits Sub- by Pinet & Tremblay (1995). As in Newfoundland, Zone and the Southern Uplands, and both contain we interpret this New England equivalent of the Caradoc black shale, the Southern Uplands has a Annieopsquotch Accretionary Tract to mark substantial southward-growing accretionary prism approximately the trace of the Red Indian Line. that has no direct correlative in Newfoundland. The Robinson & Hall (1980) inferred this line to dominantly Arenig-Llanvirn age rocks of the Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

222 r R. VAN STAAL ET AL.

Annieopsquotch Accretionary Tract, which lies while Nd-isotope studies also indicate a significant immediately west of the Red Indian Line in the juvenile component consisting of ophiolitic debris Notre Dame Subzone, may have British and Irish (Stone & Evans 1995) in the rocks of tract 1. equivalents in the northern Belt of the Southern Collectively, the most obvious source of most rocks Uplands in Scotland (Leggett et al. 1982; in tract 1 and 2 is provided by the obducted ophio- Armstrong et al. 1996) and Ireland (Morris 1987). lites of the Baie Verte Oceanic Tract, unroofed The northern Belt continues across Ireland to the Dalradian metamorphic and equivalent tracts, and South Connemara Group on the west coast (D. the Notre Dame Arc and its Grenville margin Williams et al. 1988; Harper & Parkes 1989), basement such as is exposed in Newfoundland and which exhibits a series of slices involving trench- probably also exists beneath the Upper Palaeozoic fill sediments and 'clipped-off' seamounts (Dewey cover of the Midland Valley, consistent also with & Ryan, in prep.). Palaeocurrent patterns and the presence of allochthonous Laurentian faunas provenance studies have shown that the rocks of the (Owen & Clarkson 1992) in tract 2. A comparison northern Belt had markedly variable source areas, of the tract 1 clasts with those in coeval conglom- although the bulk comes either from the northeast erates of the Midland Valley suggest that these had or northwest. Important provenance tracers are different source areas. Hence, Elders (1987) and early Caradoc (N. gracilis) conglomerates in tracts McKerrow & Elders (1989) suggested that the 1 and 2 in the Scottish Southern Uplands, which northern and Central Belts were part of an were derived from the northwest. These conglo- allochthonous terrane originally situated adjacent to merates are considered to represent trench deposits the Grenville margin of Newfoundland rather than formed when the uppermost part of oceanic crust the Midland Valley, and moved into its present represented by Early to Mid-Ordovician ocean position by sinistral strike-slip translation. If this is floor basalts (MORB to OIB, possibly seamounts, correct, the Midland Valley and parts of the Lambert et aL 1981; Phillips et al. 1995) and Southern Uplands were displaced independently by oceanic cherts entered a trench and were scraped- sinistral translation along the Laurentian margin. off and incorporated into an accretionary complex In summary, the remnants of a once-continuous resembling the Shimanto Belt in eastern Japan Shimanto-type accretionary complex (Annieop- (Taira et al. 1982). Similar conglomerates, with squotch Accretionary Tract; Figs 9 and 10) occur Dalradian metamorphic detritus, overlie the along the south or south-eastern edge of the Notre Arenig-Llanvirn South Connemara Group pillow Dame-Midland Valley magmatic arc in the basalts, which have MORB-Iike compositions northern Appalachians and British Caledonides. (Ryan et al. 1983; Winchester & van Staal 1995), The sedimentary component of the Annieop- cherts and turbidites. The Rb/Sr ages of the clasts in squotch Accretionary Tract in Newfoundland is the northern and Central Belts of Scotland suggest significantly less than in the Southern Uplands. a provenance that encompassed the Grenville However, overprinting by post-Middle Ordovician margin and an Early to Mid-Ordovician magmatic folding and faulting was much more severe in the arc (Elders 1987). Miller & O'Nions (1984) northern Appalachians (e.g.P.F. Williams et al. suggested, on the basis of Sm-Nd studies, that the 1988; Lafrance & Williams 1992; O'Brien 1991, Dalradian margin was not, in general, a source area 1993). Hence, its original architecture may have for the detritus but recent work by G. Oliver (pers. been largely destroyed and removed by erosion in comm.) on garnet chemistry suggests a possible the Newfoundland Appalachians; alternatively, and derivation from the Dalradian. The provenance of perhaps more realistically, parts of it may have been hornblende-biotite granite clasts of c. 600 +_ 40 Ma translated along the active Laurentian margin in the Corsewall conglomerate are problematic, but (Elders 1987), making the Newfoundland Appal- tonalites and granodiorites of the Lady Slipper achians an area of strike-slip excision and the Pluton of the Grenville margin of Newfoundland British Isles one of strike-slip duplication (Dewey have yielded a U-Pb zircon age of c. 555 Ma & Shackleton 1984). (Cawood et al. 1996). Although these rocks fall just outside the error range of the Rb/Sr ages, such Tectonic setting of the Exploits Subzone of rocks or undated, perhaps slightly older equivalents Newfoundland and New Brunswick elsewhere may have provided detritus to conglo- merates during the Caradoc. Rare Earth Element The Exploits Subzone (Figs 1 and 2) is composite studies (Williams et al. 1996) and detrital micas and contains elements that are unrelated (Kelley & Bluck 1989) in Caradoc sandstones of tectonically. These elements have in common that the Kirkcolm Formation of tract 2, which also has a they developed at high or intermediate latitudes and northerly provenance, confirm a source area mainly contain Gondwanan or peri-Gondwanan faunas in comprising an Early to Mid-Ordovician magmatic the Early Ordovician. During the Mid- to Late arc and associated metamorphic country rocks, Ordovician, fossils in sediments interlayered with Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 223

Fig. 10. Early Ordovician to Late Ordovician tectonic evolution of the Gander margin and its convergence with the Laurentian margin. See also Fig. 9. volcanic rocks of the Exploits Subzone in the 1994; Cocks & Fortey 1990; S. Williams et al. northern Appalachians began to acquire Laurentian 1992, 1995; Colman-Sadd et al. 1992a; van Staal or mixed affinities (Scoto-Appalachian), or contain 1994). low-latitude, deep-water faunas (e.g. Foliomena), The oldest known Exploits subduction-related, signalling arrival of Exploits elements at similar ensimatic rocks occur in the Middle Cambrian Lake latitudes to the Laurentian margin (Neuman 1984, Ambrose Volcanic Belt (513 +_ 2 Ma, Dunning et al. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

224 C. R. VAN STAAL ET AL.

1991) of the Victoria Lake Group east of the Red of a north- or northwest-facing arc-back-arc Indian Line in Newfoundland (Fig. 1). The Victoria complex (B. H. O'Brien et al. 1997; MacLachlan & Lake Group is a composite volcanic-sedimentary Dunning in press), implying a reversal in arc complex with an apparently conformable cover of polarity. Mafic calc-alkaline arc magmatism Mid-Ordovician limestone, shale and chert, and continued at least until c. 464 Ma, coeval with Late Ordovician to Silurian greywackes of the felsic calc-alkaline arc volcanism in the Victoria Badger Belt (Williams et al. 1993). The volcanic Lake Group (c. 462 Ma). component also includes Upper Cambrian / lower Formation of the Tremadoc arc volcanic rocks in Tremadoc (498 +6/-4 Ma ensimatic Tulks Hill the Victoria Lake, Exploits and Wild Bight Groups Belt, Evans et al. 1990) and Llandeilo (462 +4 / -2 partly overlaps formation of Tremadoc ophiolites Ma Victoria River rhyolite, Dunning et al. 1987) (e.g. 494 _ 2 Ma Pipestone Pond Ophiolite, calc-alkaline felsic volcanic rocks. Therefore, Dunning & Krogh 1985). These ophiolites have subduction of Iapetus oceanic crust had started by been interpreted as having formed during rifting of at least the Mid-Cambrian at high latitudes and an ensimatic arc (Jenner & Swinden 1993), continued for more than 50 Ma. Formation of ensi- remnants of which are probably represented by the matic and ensialic arc complexes at high latitudes Lake Ambrose and Tulks Hill Belts in the Victoria thus overlaps for a considerable period with Lake Group and temporal equivalents in the Wild formation of the Baie Verte Oceanic Tract and Bight and Exploits groups. Our interpretation is Notre Dame Arc at low latitudes; hence, Iapetus that all Tremadoc and older ensimatic magmatism must have contained several kinematically- in the Exploits Subzone represent part of a single independent subduction zones. It is significant that, late Mid-Cambrian to late Tremadoc south-east- towards the east, the Victoria Lake Group is tecton- facing, ensimatic arc-back-arc complex (Penobscot icaUy juxtaposed with two late Neoproterozoic Arc of van Staal, 1994; Fig. 10). The late Arenig- quartz-monzonite plutons (c. 563 Ma). The plutons Llanvirn (473-460 Ma), northwest-facing arc rocks experienced an earliest Cambrian metamorphism in the Victoria Lake, Wild Bight and Exploits (c. 545 Ma) and were amalgamated with the groups are referred to herein as the Victoria Arc Victoria Lake Group by at least the Early Silurian (see below). Penobscot age (493 +_ 2 Ma) arc-back- (Evans et al. 1990). The late Neoproterozoic ages, arc elements occur as a fault-bounded belt nature of plutonism and associated metamorphism (Annidale Belt) in the Gander Zone of southern are typical of Ganderian basement (S. J. O'Brien et New Brunswick and adjacent Maine (Fig. 6), al. 1996; van Staal et al. 1996a). These relation- immediately north of the boundary between the ships suggest pre-Silurian interaction between Gander and Avalon zones (McLeod et al. 1992, elements of the Victoria Lake Group and Ganderia. 1994). Extension of the Penobscot Arc complex Inherited pre-Cambrian zircons in the Victoria from the Red Indian Line to the line of the Gander River rhyolite (Dunning et al. 1987) and Pb-isotope River Ultrabasic Belt in eastern Newfoundland studies (Swinden & Thorpe 1984) suggest that (Fig. 1) and into the Annidale Belt of southeastern amalgamation took place during the Early New Brunswick and Maine (Fig. 2) requires that Ordovician Penobscot obduction (see below) the oceanic complex was obducted for at least because the older Cambrian and Tremadocian several hundred kilometres over the Gander volcanic rocks of the Victoria Lake Group appear to margin. Subsequent deformation, erosion and Mid- be ensimatic with no indication of interaction with to Late Ordovician cover sequences have largely continental crust in their petrogenesis (Dunning et destroyed or obscured this configuration, although al. 1991). mantles of Penobscot ophiolitic rocks and mrlanges Temporal correlatives of the Early and Mid- around large structural inliers of Gander Zone rocks Ordovician parts of the Victoria Lake Group occur in the Exploits Subzone of central Newfoundland in the Wild Bight and Exploits Groups along strike leave little doubt that this process has happened (Fig. 1) towards the north (B. H. O'Brien et al. (Fig. 1). Despite its considerable tectonic transport, 1997; MacLachlan & Dunning in press). Detailed obduction of the Penobscot arc was a relatively stratigraphic and petrological studies have revealed shallow 'soft' event, mainly characterized by the formation of a south- or southeast-facing mrlanges (Williams & Piasecki 1990; van Staal Tremadoc, juvenile volcanic-plutonic arc complex 1994) and minor shear zones and folding but not by (c. 490-486 Ma) with an ophiolitic base (B. H. any significant regional metamorphism. Uncon- O'Brien 1992). This volcanic-plutonic arc complex formities, stitching plutons and tightly-dated experienced rifting and was exhumed during the movement zones constrain this event to the late Arenig. After a marked decrease or absence in Tremadoc/early Arenig (485-478 Ma) in the volcanism during the early and middle Arenig northern Appalachians (Neuman 1967; Williams & (volcanic shut-off), volcanism flared up again in the Piasecki 1990; van Staal & Williams 1991; late Arenig (c. 472-470 Ma), recording formation Colman-Sadd et al. 1992b; Dunning et al. 1993; Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 225

Tucker et al. 1994). Originally, the Penobscot sets of felsic volcanic rocks in New Brunswick and Orogeny was defined in east-central Maine near the Newfoundland were deposited on Ganderia border with New Brunswick by an unconformity basement (Colman Sadd et al. 1992b; van Staal et that is also present in central and northem New al. 1996a and references therein). In New Brunswick. The unconformity separates Cambrian, Brunswick, the Tetagouche back-arc basin opened Oldhamia-bearing sandstones of the Gander Zone up sufficiently to produce oceanic crust (470-460 and m61ange (Boone et al. 1989) from overlying Ma MORB-like Fournier Ophiolite Complex (van middle Arenig-Llanvim volcanic and sedimentary Staal et al. 1991; Sullivan et al. 1990; Sullivan & rocks of the Exploits Subzone. The latter typically van Staal 1996). Although no Mid-Ordovician contain high latitude, peri-Gondwanan shelly ophiolite has been recognized as such in Exploits faunas (Neuman 1984). The Mid-Ordovician rocks back-arc basin in Newfoundland, supposedly Mid- (including their peri-Gondwanan fossils) of the Ordovician ocean floor basalts (E- or T-MORB) Exploits Subzone were also deposited uncon- occur together with silicic volcanic rocks in the formably above the Penobscot ophiolites and the Cold Spring Pond and North Steady Pond Gander sandstones in Newfoundland and thus form Formations of Central Newfoundland (Fig. 1; an important overstep sequence (van Staal & Swinden 1988). Their geological and geochemical Williams 1991; Colman-Sadd et al. 1992b). characteristics are reminiscent of the fragments After the Penobscot collision, arc magmatism, of transitional and oceanic crust preserved in the represented by ensialic arc plutonic-volcanic nappes that form the boundary between the complexes, flared up again during the Arenig (479- ensialic and ensimatic elements of the Tetagouche 474 Ma) in central and northern New Brunswick backarc basin in New Brunswick (van Staal et al. and adjacent Maine. Rifting and migration of this 1991). arc towards the northwest formed the Arenig- Accretion of the Popelogan-Victoria Arc with Llanvirn Popelogan Arc-Tetagouche back-arc the Annieopsquotch Accretionary Tract of the complex (c. 473-455 Ma) in New Brunswick (van Notre Dame Subzone and, by implication, Staal et al. 1991), demanding that subduction was Laurentia is reasonably well constrained in the towards the southeast and that subduction polarity Canadian Appalachians. First, there is the shut-off along the Gander margin had reversed after the and locally also uplift of the Popelogan Arc in the Penobscot collision (van Staal 1994). Equivalents Caradoc (van Staal et al. 1991) and, second, there is of the Arenig part of the Popelogan Arc in a sedimentary linkage represented by arc and southwestern Newfoundland are the large Baggs ophiolitic detritus and resedimented fossils derived Hill granite pluton (478 Ma, Tucker et al. 1994), from the Notre Dame Arc in the late Caradoc- which cuts the Penobscot structures and obducted Ashgill rocks of the adjacent Exploits Subzone ophiolites, and the nearly coeval tonalites, diorites (Nelson 1981; Arnott 1983; Colman-Sadd et al. and granodiorites of the Margaree Complex (c. 1992a; O'Brien et al. 1997). The Upper 474 Ma) near Port aux Basques (Van Staal et al. Ordovician, generally north-younging, upward- 1994; Valverde-Vaquero et al. 1998). The Baggs coarsening greywackes and conglomerates of the Hill Pluton was partially exhumed before the Mid- Badger Group that overlie the rocks of the Victoria Ordovician (van Staal et al. 1996c), probably as a Arc probably also date accretion. Their deposition result of arc rifting. The late Arenig-Llanvirn arc appears to be diachronous from west to east and and rifted arc rocks of the Victoria Lake, Exploits likely to have been structurally controlled because and Wild Bight Groups (Victoria Arc) in central there are consistent stratigraphic differences and northern Newfoundland (Fig. 1) are the between each thrust sheet (Arnott 1983; Arnott et obvious kinematic equivalents of the Mid- al. 1985; P. E Williams et al. 1988). We interpret Ordovician part of the New Brunswick Popelogan these rocks as clastic trench-fill wedges Arc. The Mid-Ordovician felsic volcanic rocks (McKerrow & Cocks 1976) deposited when the (and their contained syn-genetic massive sulphides) Victoria Arc together with an accreted seamount of the Bay du Nord and Baie d'Espoir Groups in (Summerford Group: E-MORB - limestone associ- southern and southeastern Newfoundland (Fig. 1) ation, Wasowski & Jacobi 1985) entered the trench are lithologically, temporally and isotopically in front of the Annieopsquotch Accretionary Tract. similar to coeval rocks in the Tetagouche Group We propose that the Summerford seamount was (Swinden & Thorpe 1984; Dunning et al. 1990; already accreted to the Victoria Arc prior to Colman Sadd et al. 1992b). Together with the collision with the Annieopsquotch Accretionary equivalent volcanic rocks in the Tetagouche Group Tract when it entered the west-facing subduction in New Brunswick they are considered to represent complex that must have fringed this arc magmatism generated during the early phases of immediately to the west (Fig. 10). Remnants of this ensialic back-arc basin formation (referred to as the accretionary complex are represented by the late Exploits back-arc basin in Newfoundland). Both Arenig-early Llanvirn Dunnage M61ange, which Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

226 c. R. VAN STAAL ET AL. contains blocks of the Summerford seamount and obtained from the Tetagouche Group by Liss et al. the Early Ordovician Exploits Group (Wasowski & (1993) who determined an average palaeolatitude Jacobi 1985). The presence of the Cambrian of 52 ~ + 21/-16~ mainly from Llanvirn-age (c. trilobite Baliella in a limestone block in the 470-466 Ma) rocks. We consider that their data Dunnage Mrlange is consistent with the Dunnage probably derives from the passive margin side of having formed at high latitudes near the the back-arc basin. Considering its large error, and Gondwanan rather than the Laurentian side of the that Avalon was consistently moving northwards Iapetus Ocean (Dean 1985). The Dunnage such that the Avalon peninsula of eastern accretionary complex was probably small, like the Newfoundland acquired a palaeolatitude of 32 ~ _ accretionary complex off the Mariana Arc, because 10~ (Hodych & Buchan 1994) at 441 Ma (Ashgill, the supply of clastic material to intraoceanic Greenough et al. 1993), suggests that the calculated trenches is generally low. The Dunnage Mrlange average of Liss et al. (1993) probably over- was intruded late syn- to post-kinematically by the estimates the palaeolatitude of Ganderia, consistent early Llanvirn (c. 467 Ma) Coaker Porphyry with a Caradoc palaeolatitude of c. 43~ for Avalon (Elliott et al. 1991). The Coaker Porphyry has been in the British Isles (Trench et al. 1992; Channell & interpreted as an S-type granite, although it McCabe 1992). The Tetagouche-Exploits back-arc contains many ultramafic xenoliths (Hibbard & basin thus achieved a width of at least 1100 km by Williams 1979; Lorentz 1984; Currie 1995). This the Caradoc (c. 10 ~ of latitudinal separation) but porphyry probably formed by melting of sediments not more than 2000 km. An upper value of 2000 km in the accretionary wedge, either during subduction is constrained by the occurrence of low-latitude, of a spreading ridge (Kiddet al. 1977) shortly after midcontinent, conodont faunas in late Caradoc to accretion of the Summerford seamount to the Ashgill limestones (Nowlan et al. 1997) deposited Victoria Arc or as a result of trenchward migration on Avalon Zone rocks (Ganderia) in southern New of this arc. Other possible remnants of accreted Brunswick (Fig. 2). A Tetagouche-Exploits basin seamounts or ridge subduction along the wider than 2000 km would also imply unrealistic approximate trace of the Red Indian Line are the convergence rates of 20 cm/a or higher. The large Caradoc Winterville and Munsungun within-plate width of the Tetagouche-Exploits Basin (Iapetus II alkalic and tholeiitic basalts in northern Maine near of van der Pluijm & van Staal 1988) was achieved the New Brunswick border (Winchester & van probably by slab rollback (hinge retreat; Dewey Staal 1994; Fig. 2), which are associated with 1980), dragging the Popelogan-Victoria Arc across Caradoc black shales typical of the Exploits Iapetus to the Laurentian margin in a manner Subzone. Their low palaeomagnetic latitudes analogous to the movement of the Cimmerian (10-20~ Potts et al. 1993, 1995) indicates that microcontinent across Palaeo-Tethys, opening up they must have formed close to the Laurentian Neo-Tethys in its wake (Sengrr 1979). margin in the Iapetus Ocean; hence, formation in Closure of the Tetagouche-Exploits Basin took the Tetagouche back-arc basin as earlier postulated place by west or northwestward-directed sub- by Winchester & van Staal (1994) is probably duction (Fig. 10) during the Ashgill and Lower incorrect. A Late Ordovician collision between the Silurian (van Staal 1994; Currie 1995); basically Popelogan-Victoria Arc and the Notre Dame Arc the northwest-dipping Benioff zone beneath the also explains the breakdown in faunal provinciality Notre Dame Arc/Annieopsquotch Accretionary in the late Llanvirn-Caradoc (replacement of peri- Tract, jumped eastward behind the accreted Gondwanan faunas by Scoto-Appalachian faunas in Popelogan-Victoria Arc. Hence, southeast-directed Exploits Subzone rocks) and poor preservation of thrusting continued in the accreted Victoria- the Exploits accretionary complex because most of Exploits Arc in Newfoundland during the Early the latter was probably subducted beneath the Notre Silurian (e.g. Lafrance & Williams 1992) and pro- Dame Arc. Modern analogues of face-to-face arc pagated southeastward when progressively more of collisions occur in the Molucca Sea (Fig. 4) and the Tetagouche-Exploits Basin was subducted (van immediately east of New Guinea where the Staal 1994; Currie 1995). Continuation of sub- Trobriand Trench and associated fore-arc are being duction is also supported by the flare-up of arc subducted diachronously beneath the Finistere- magmatic activity (now dominantly subaerial) in south Solomon arc (Pegler et al. 1995). the Notre Dame Subzone of Newfoundland and When the Notre Dame and Victoria Arcs had northern Quebec during the Ashgill and early collided in the Late Ordovician, the active Llandovery (453-430 Ma, Whalen 1989; David & Laurentian margin (Notre Dame Arc) was situated Gariepy 1990; Dunning et al. 1990; Dub6 et al. at intermediate latitudes (c. 20--30~ separated 1996). The suture zone of the Tetagouche-Exploits from the Gander margin by the Tetagouche- Basin is marked by Ashgill blueschists (van Staal et Exploits back-arc basin. The width of this marginal al. 1990) and Late Ordovician to Early Silurian basin is constrained solely by palaeomagnetic data ophiolitic mrlanges in New Brunswick and Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOI~U3"ION OF IAr'Ea'US 227

Newfoundland (e.g. Duder Complex-Carmanville be buried beneath the Connecticut Valley or MElanges in northeast Newfoundland and obscured/destroyed during Acadian tectonism. The Belledune River MElange in northern new best record of the Ordovician accretionary events is Brunswick, Williams et al. 1993; van Staal 1994; preserved in Maine, although the precise affiliation Currie, 1995; Lee & Williams 1995). of the numerous subduction-related melanges is not well understood. The Hurricane Mountain MElange in western Maine supposedly has been thrusted The Exploits Subzone in New England and above the Boil Mountain Ophiolite Complex the British Isles (Boone et al. 1989). It contains mafic clasts with low temperature-high pressure assemblages. A The Mid-Ordovician Popelogan Arc can be traced sodic-calcic amphibole in one of the entrained into adjacent Maine (Winchester & van Staal 1994) mafic bodies yielded an 4~ age of 485 + 4 and probably is equivalent to the Mid- to Late Ma, which predates the 477 Ma tonalite of the Boil Ordovician Bronson Hill magmatic arc further Mountain Ophiolite Complex (Kusky et al. 1997). along strike in New Hampshire and Massaschusetts Juxtaposition of the Hurricane Mountain MElange (e.g. Leo 1985; Tucker & Robinson 1990). Late and the Boil Mountain Ophiolite Complex probably Proterozoic basement (613 Ma) and detrital zircon took place after the Early Ordovician because suites in associated quartzites (Robinson & Tucker tonalite bodies or clasts do not occur in the melange 1996) are similar to those found in Ganderian and high pressure-low temperature assemblages do basement in New Brunswick and Newfoundland not occur in the ophiolite. The Hurricane Mountain (see van Staal et al. 1996a and references therein). MElange is unconformably overlain by Mid- Another link between Ganderian basement and arc Ordovician volcanic and sedimentary rocks. The plutons is preserved possibly in the Weeksburo- latter have been interpreted to form part of the Lunksoos inlier in central Maine (Fig. 2). The Popelogan-Tetagouche arc/back-arc system Upper Ordovician Rockabema Diorite (Neuman (Boone et al. 1989; Winchester & van Staal 1994). 1967) probably also forms part of the Popelogan The Hurricane Mountain MElange probably repre- Arc. It stitches Cambrian (Oldhamia) sandstones of sents a remnant of an accretionary complex formed the Grand Pitch Formation and Early Caradoc during the Penobscot Arc collision with Ganderia. tholeiitic basalts. The basalts acquired a Late An undated Ordovician subduction-related mEl- Ordovician palaeolatitude of 20~ (Wellensiek et ange (Fig. 2, Caucomgomoc MElange), exposed as al. 1990), which is consistent with their accretion to a basement window in the Connecticut Valley syn- and/or forming part of the Popelogan Arc, because clinorium, is associated with east-directed thrusting the arc had docked or nearly so by this stage with and has been correlated with the Hurricane the active Laurentian margin. Mountain MElange (Pollock 1993). However, it lies The Bronson Hill Arc was accompanied by a much further north and not on strike with the back-arc basin, the closure of which telescoped the Hurricane MElange, nor does it have the same Partridge and Ammonoosuc volcanic rocks in unconformable cover. This melange may be an Massachusetts with the magmatic arc (Hollocher equivalent of the melanges situated along the Red 1993). Rifting of the Bronson Hill magmatic arc Indian Line in Newfoundland (e.g. Boones Point and formation of oceanic-like back-arc crust must Complex) and, if correct, may mark the trace of the have started by at least 467 _+ 3 Ma in New Red Indian Line through Maine (Fig. 2). Hampshire (Fitz & Moench 1996), whereas A possible continuation of the Popelogan- volcanic rocks in Massachusetts are in part as Victoria Arc into the British Isles has important young as 453--449 Ma (Tucker & Robinson 1990). implications for understanding the tectonic setting Combined, these data suggest a tectonic setting of the Southern Uplands (e.g. McKerrow 1987). very similar to that represented by the Popelogan Some Caradoc turbiditic sandstones of tract 2 in the Arc-Tetagouche back-arc system in New northern Belt show a sudden influx of fresh Brunswick and Maine, although arc-back-arc andesitic detritus. Palaeocurrent directions in these activity appears to have continued slightly longer in sediments are variable (Leggett et al. 1982) but at southern New England. least some of the volcanic detritus appears to have The Boil Mountain Ophiolite and the on-strike been derived from the south. It has been argued that soapstone belt, immediately west of the Bronson the variability in palaeocurrent directions may have Hill Anticlinorium in the Connecticut Valley in been the result of meandering currents on the trench New Hampshire and Massachusetts (Lyons et al. floor (e.g. Leggett 1987) or the presence of a 1982), are inferred to mark the approximate trace of southerly island arc offshore (e.g. Hutton & the Red Indian Line in New England. Equivalents Murphy 1987; Stone et al. 1987; Morris 1987). of the Boones Point and Exploits subduction However, in the Southern Uplands, while there is complexes per se have not been identified but could evidence of the accretion of seamounts (e.g. Bail Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

228 C. R. VAN STAAL ET AL.

Hill, Wrae), there is no exposed evidence of a major outboard edge of the Gander margin. The oceanic arc. Furthermore the smooth, diachronous, progres- basin in which the Ashgill-Lower Silurian rocks of sively younger southwards facies change from the central and southem belts were deposited, i.e. oceanic shales to trench turbidites seems to pre- the basin south of the accreted but poorly preserved clude such an arc in the Southern Uplands. Longford Down Arc, is the logical continuation of Remnants of an arc may, however, be preserved in the Tetagouche-Exploits back-arc basin (Iapetus II) the Longford Down inlier of Ireland (Fig. 3). John in the British Caledonides. Hence, the closest Morris of the Irish Geological Survey has mapped analogue to the Silurian Navan-Silvermines Suture calc-alkaline basalts and andesites in close of the British Isles in Newfoundland is the Early proximity to Caradoc (N. gracilis) black shales in Silurian Dog Bay Line, just east of the Reach Fault Slieve Aughty and Tattinlieve (Winchester & van where Arnott et al. (1985) located their Iapetus Staal 1995), south of the Leadhills Line-Northern suture in Newfoundland. Continuation of north- or Belt Median Fault (LL-NBMF in Fig. 3). Implicit northwestward-directed subduction during the in the island arc models of Stone et al. (1987) and Ashgill and Early Silurian is much more widely Morris (1987) is that the northern Belt sediments accepted in the British Isles (e.g. Leggett et al. were deposited in a back-arc basin, contrary to the 1982) than in the northern Appalachians, despite evidence for their formation in a trench. If, on the preservation of high pressure-low temperature other hand, the southerly-derived andesite detritus metamorphic rocks (e.g, blueschists) and reflects an approaching island arc, similar to the subduction-related mrlanges of Ashgill to Early rapidly northwards-drifting Popelogan-Victoria arc Silurian age in New Brunswick (van Staal et al. in the northern Appalachians (Fig. 10), the sudden 1990) and Newfoundland (e.g. Williams et al. influx of andesite detritus may mark the unroofing 1993). Not surprisingly, similarly to the northern of this arc as a result of its accretion to the active Appalachians, calc-alkaline, dominantly subaerial Laurentian margin during the Caradoc. Such a volcanism also continued into at least the Wenlock model removes the necessity of invoking a back-arc on the active Laurentian margin of the British basin to explain the development of the northern Caledonides (McKerrow & Campbell 1960). Belt. Underthrusting of Avalonian crust beneath the The Caradoc Grangegeeth Arc-related basalts Southern Uplands accretionary prism began in the and andesites (Winchester & van Staal 1995) in Wenlock in England (Kneller et al. 1993; King eastern central Ireland (Fig. 3), a small terrane 1994) but may have started earlier in Ireland bounded on all sides by splays of the Navan- (Hutton & Murphy 1987) suggesting a west to east Silvermines Fault (the classical trace of the Iapetus diachroneity. Williams et al. (1993) also presented Suture in the British Isles), may represent a evidence for a Silurian closure of the last displaced part of the Popelogan-Victoria or vestiges of Iapetus along the Dog Bay Line in Longford Down Arc; that is, it may have been Newfoundland. Van Staal (1994), on the other situated originally to the north of the central and hand, argued that the first arrival of highly extended southern belts. We propose this model, because the Ganderian crust in the Brunswick subduction Grangegeeth Arc Terrane, like the Popelogan- complex in the Ashgill marked the onset of Victoria Arc in the Appalachians, shows a change collision between Ganderia and Laurentia in New in biogeographical affinity during the Ordovician. Brunswick. Although this is consistent with a west- It contains high-latitude Atlantic province grapto- to-east diachroneity, the Brunswick subduction lites in the Arenig but mixed Laurentian-Baltic complex did not breach sea level before the late brachiopods in the Caradoc (referred to as Scoto- Llandovery. Considering that the earliest-accreted Appalachian by Harper & Parkes 1989 and Owen piece of Ganderian crust was situated on the arc & Clarkson 1992). These faunas are different from side of the Tetagouche Basin (Rogers & van Staal the Anglo-Welsh faunas of the Lake District and 1997), the latter event rather than the first arrival of the Laurentian faunas of the Southern Uplands this highly-extended continental crust or a thick during the Caradoc, although the affinities are sediment pile on oceanic crust is, therefore, better much stronger with the Laurentian than the Anglo- regarded as the real start of collision. Moreover, the Welsh faunas. The adjacent Bellewstown Terrane extended crust that entered the subduction zone contains strikingly different within-plate mafic and during the Ashgill comprised a relatively large felsic volcanic rocks (Winchester & van Staal amount of E-MORB pillow lavas interlayered with 1995) and is characterized by Anglo-Welsh faunas felsic dacite (van Staal et al. 1991; Rogers & van during the Caradoc (cf. Harper & Rast 1964). The Staal 1997) and probably had properties compar- presence of a Llanvirn shelly fauna in the able with transitional crust. Analogous to the Bellewstown Terrane (Harper et al. 1990) similar to British Isles, an extensive foreland basin formed on those in the Exploits Subzone of Newfoundland the Ganderian margin during the Early to Late suggests that this terrane probably formed near the Silurian in the northern Appalachians (e.g. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 229

Fredericton Trough, Fyffe 1995; van Staal & de groups in northern England to represent equivalents Roo 1995), the sediments of which are litho- of the Ganderian margin in the British Isles. logically similar to those in the Windemere Group Common to all Lower Ordovician units of the in England. Like the Windermere Group, the Ganderian and Avalonian margin in the British Isles Fredericton Trough was inverted and strongly is a pre-Llanvirn deformation event represented by deformed before the end of the Silurian (West et al. ductile folding, mrlanges and/or olistostromes 1992; Fyffe 1995). (Kokelaar 1986; Bennet et al. 1989; Max et al. The Cambrian-Early Ordovician Penobscot Arc 1990; Cooper et al. 1995) and late Tremadoc/early and its accretionary complex probably also Arenig (c. 480 Ma) mylonites in the Rosslare continued into the British Isles (van Staal et al. Gneisses (Max & Roddick 1989). The locally- 1996a). In Leinster (southeast Ireland, Fig. 3) intense, pre-Llanvirn recumbent folding and olisto- ophiolitic, serpentinized ultramafic-mafic slivers stromes of the Skiddaw Group are generally inter- and pods are tectonically juxtaposed with the preted to have formed in response to significant Ribband and Duncannon Groups along the slope instability. We are not convinced that all the Wicklow Fault Zone (Gallagher 1989; Gallagher et strong folding took place when the rocks were still al. 1994; Max et al. 1990; Winchester & van Staal unlithified but we suggest that this deformation is 1995), while serpentinites, gabbros and basalts related to loading and obduction of the margin by occur in the New Harbour Group and Gwna the ensimatic Penobscot Arc-accretionary complex, mrlanges of the Monian Supergroup together with although remnants of the latter are not exposed or quartzite clasts of the underlying Cambrian South preserved in northern England. However, the Stack Group (Gibbons & Ball 1991; Gibbons et al. former existence of the Penobscot Arc is consistent 1994). The basalts have, at least in part, an island with the juvenile arc detritus, represented by among arc affinity (Thorpe et al. 1984). Lithostratigraphic others relatively high contents of Cr and Ni, in the correlations and sedimentary linkages suggest that upper Arenig sediments of the Skiddaw Group the ophiolitic lenses and the highly tectonised (Cooper et al. 1995). If early Ordovician Ribband Group structurally overlie a Cambrian southeastward ophiolite obduction took place along substrate mainly consisting of turbiditic quartzose the southeast margin of Iapetus, the ultramafic sandstones and shales, containing Oldhamia, of the rocks along the Wicklow Fault Zone must be part of Cahore, Cullenstown and Bray Groups (Max et al. an ophiolite nappe rooted to the west of the present 1990). On the basis of lithostratigraphic corre- Leinster Palaeozoic inlier, because the Bray Series lations with the above units in Ireland, Tietzsch- lies to the west of the fault zone yet belongs to the Tyler & Phillips (1989) also interpreted the South Gander Zone. Stack Group as Cambrian, consistent with the Deformation of the Avalonian margin in possible presence of Skolithos burrows (see mainland Wales occurred somewhat earlier than in Gibbons et al. 1994 for review). The lithological northern England and Ireland. Folding in the characteristics of the latter rock units invite Tremadoc was followed rapidly by the formation of correlation with the Cambrian-Lower Ordovician a late Tremadoc/early Arenig ensialic arc (Kokelaar Gander Zone sandstones and shales of the northern 1986), which may be correlated with the early Appalachians, while the mafic-ultramafic rocks Arenig part of the Popelogan/Victoria Arc in the and mrlanges are temporal and tectonic equivalents northern Appalachians prior to its rifting from of the obducted late Tremadoc/early Arenig Ganderia. This requires some diachroneity of the Penobscot Arc-accretionary complex. The thrust Penobscot collision and arc polarity reversal, which slices of the New Harbour Group and Gwna mrl- is permissible considering the strike-slip motion ange above the South Stack Group quartzites on along the Avalonian and Ganderian terranes in the Anglesey and ultramafic rocks along the Wicklow British Isles (Kokelaar 1988; Gibbons 1990; Horak fault zone are thus the British analogues of the et al. 1996). Gander River Ultrabasic Belt line in The late Llanvirn-Caradoc Lake District Arc in Newfoundland. The unexposed basement to the northern England (Kokelaar 1988) has no direct British Gander Zone equivalents is linked com- temporal equivalent in a similar tectonic setting in monly with the Rosslare and Coedana complexes of the northem Appalachians. Correlatives of the Lake the Monian terrane (Gibbons et al. 1994). The latter District volcanic rocks occur in southeastern complexes show similarities to exposed Ganderian Ireland, Wales and as concealed bodies (Fig. 3) in basement in the northern Appalachians (van Staal eastern England (Noble et al. 1993). However, the et al. 1996a) and, similarly with the northern Llanvirn-Caradoc volcanic rocks in both Wales and Appalachians, basement and structural cover are southeastem Ireland were formed in a supra- linked by an Arenig overstep sequence containing subduction extensional setting, probably an ensialic the 'Celtic' fauna. Hence, we consider these Lower backarc basin (Kokelaar et al. 1984; McConnell et Palaeozoic clastic rocks and the Skiddaw and Manx al. 1991; Winchester & van Staal 1995) rather than Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

230 c.R. VAN STAAL ET AL. in an arc senso stricto. Kokelaar (1988) argued that movement between Amazonia, Avalonia and the Lake District-Leinster Terrane was unlikely to Baltica during the Cambrian suggests that these have been adjacent to Wales in the Ordovician; were more or less fellow travellers. Spreading rates instead, he suggested that these two terranes were in the Tornquist Sea were therefore probably low or juxtaposed by sinistral strike-slip motion. The zero soon after the rift-drift transition. Baltica, alternative, that the Lake District Arc and Wales however, was far enough from Avalonia and by backarc basin were in their present relative posi- implication, Amazonia, during the Early to Mid- tions with respect to one another, would require that Cambrian to have significantly more diverse the late Tremadoc arc of Wales had migrated trilobite faunas than Avalonia (McKerrow et al. northwards for more than 150 km and had intruded 1992). On the other hand, Siberia moved inde- its own fore-arc-accretionary complex. There is no pendently and far enough east to allow faunal evidence for such an arc migration in England and exchange (McKerrow et al. 1992) with the Wales, nor does the Skiddaw Group resemble a Cadomian elements of southern Europe (Bohemia, fore-arc-accretionary complex. An alternative Amorica, Iberia). Because the south pole was solution is that the Lake District Arc does not have situated in northwest Africa during most of the a southwesterly strike continuing into Ireland but is Cambrian (Meert & Van der Voo 1997), we infer related to south- or southwesterly-directed sub- that the majority of Morocco (north of the South duction of the Tornquist Sea beneath Avalonia Atlas Fault) containing warm water Early (McKerrow et al. 1991; Noble et al. 1993). This Cambrian archaeocyathans was, at the time, at a model is attractive because the Late Ordovician lower latitude than adjacent Mauritania; hence cessation of the Lake District Arc coincides with Morocco was situated further eastward (present amalgamation of benthic faunas between Baltica coordinates) along the Gondwanan margin and may and Avalonia (Cocks & Fortey 1982), both sug- have been connected with Cadomia (Figs 1 and 7). gesting a Late Ordovician start of collision between Intra-oceanic subduction east of Siberia appears Baltica and Avalonia. This putative collision was to have been active during most of the Cambrian probably 'soft' because there is no obvious supply and formed the complex Kipchak Arc, an arm of of coarse clastics from Baltica during this time in which, the Tuva-Mongol Arc, may have extended Avalonia (Soper & Woodcock 1990). Collision into the Iapetus Ocean west of Siberia (Seng6r & probably continued into the late Llandovery when Natal'in 1996). Intra-Iapetus subduction on the all physical barriers (seaways) for benthic ostracods Gondwanan side must have started by at least were removed (Berdan 1990). The latter is consis- 513 Ma to account for the initiation of the tent with Upper Ordovician to Lower Silurian calc- Penobscot Arc. Soon after, subduction started along alkaline volcanic rocks in the Brabant massif of the Laurentian margin forming the Baie Verte Belgium (Andr6 et al. 1986) and SW-dipping Oceanic Tract, which may have linked up with, or seismic reflectors in the subsurface across the perhaps formed by southward propagation of, the Caledonian deformation front in Denmark and Tuva-Mongol Arc (Fig. 7b, c). Obduction of the northern Germany (Tanner & Meissner 1996). Baie Verte Oceanic Tract and Penobscot Arc onto, respectively, the Laurentian and the Ganderian/ Avalonian margins started nearly coevally between Synthesis of the tectonic evolution of 490 and 485 Ma (Tremadoc). Baltica also experi- Iapetus enced obduction at this time as evidenced by, for example, eclogite metamorphism (Andr6asson & The continental reconstructions of the southern Albrecht 1995). The Tremadoc was also marked by hemisphere between 555 and 435 Ma shown in Fig. a transgression. Combined, these events signal a 7 summarize the opening and closing history of the major plate reorganization during the Tremadoc, Iapetus and neighbouring oceans. These recon- which changed the dynamics of the Iapetus Ocean structions are based on a combination of palaeo- to one of destruction rather than one of opening. magnetic, faunal and geological data, most of Obduction on both margins was followed by an arc which has been discussed above. Opening of the polarity reversal which formed the Notre Dame and Iapetus Ocean took place in the latest Neo- Popelogan-Victoria Arcs on, respectively, the proterozoic (590-550 Ma) and continued to widen Laurentian and Ganderian margins. The Notre throughout the Cambrian at c. 7.5 cm/a across the Dame Arc was a compressive Andean-type Amazonian-Laurentian sector. While Amazonia, continental arc resulting from southward migration Avalonia and Baltica moved to the south and of Laurentia (Fig. 9). The Popelogan-Victoria Arc Siberia moved east with respect to Laurentia, the was extensional and rifted-off Gondwana shortly latter continued migrating further northwards after the arc-polarity reversal and opened the towards the equator (Mac Niocaill & Smethurst Tetagouche-Exploits back-arc basin in its wake at a 1994). The absence of any significant differential rate of c. 5 cm/a. Initiation of subduction beneath Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 231

Avalon in the Arenig approximately coincided with out the Silurian Laurentian ostracods were mostly its departure from Gondwana, which opened the distinct from the Avalonian/Baltic ostracod Rheic Ocean. Hence, the rapid roll-back that caused province (Berdan 1990). However, some ostracod the rifting of the Popelogan-Victoria Arc may have connection existed from the Llandovery, suggesting contributed also to the partial break-up of west geographic proximity of these two landmasses by Gondwana during the Lower Ordovician. the Llandovery, a view supported by the inter- Globally, the Early Ordovician and the Mid- to change of fish between Baltica and Laurentia in the Late Cretaceous were two unique periods in Silurian (Turner & Turner 1974). Persistence of Phanerozoic tectonic history (Dewey 1988). They narrow seaways after the start of collision such as were both characterized by high sea levels, with the foreland basin between present day Timor and carbonate platforms and oceans with abundant Australia (cf. van Staal & de Roo 1995) can explain black shales, widespread supra-subduction-zone the ostracod biogeography. Breakdown of the ophiolites obducted shortly after generation, and faunal provincialities of ostracods and fish between abundant blueschists and arc magmatism. Both Baltica and Gondwana in the Emsian coincide with were times of substantial plate boundary reorganiz- the formation of two short-lived brachiopod faunal ation, perhaps caused by high rates of relative plate provinces (Rhenish-Bohemian versus the motion with widely-dispersed continents and little Appalachian) following cosmopolitan brachiopod continental collision (Dewey 1988). conditions during most of the Silurian. These By the middle to late Caradoc, the Notre Dame relationships have been explained by impingement and the Popelogan-Victoria Arcs had collided, of part of Gondwana with Laurentia during the implying rapid closure (c. 12.5 cm/a) of the main Emsian creating a narrow land connection allowing tract of the Iapetus Ocean during the late Arenig- interchange of fish and ostracods but separating the early Caradoc (475-455 Ma). This is not surprising Rheic Ocean into two major seas each with a considering that Iapetus was subducting beneath distinctive brachiopod fauna, namely the Rhenish- both margins, i.e. both the Laurentian and Bohemian fauna in the east and the Appalachian Avalonian/Ganderian margins were initially active fauna in the west (Scotese & McKerrow 1990). The (Fig. 10). Such a tectonic scenario was originally geographical distribution of these two provinces proposed by Dewey (1969), supported by Liss et al. was further influenced by the presence of a (1993), van der Pluijm et al. (1995) and Mac mountain range between the Gasp6-Connecticut Niocaill et al. (1997) based on their palaeomagnetic Valley seaway and Avalon (van Staal & de Roo data and is also supported by palaeontological 1995). evidence (e.g. Cocks & Fortey 1990). In the southern Appalachians, the Amazonian After the Caradoc, convergence between margin of Gondwana may have collided with Laurentia and Avalon continued with north- Laurentia during the Silurian (Fig. 7f), possibly westerly-directed subduction but now involving following an earlier arrival of the Carolina micro- mainly young oceanic crust of the Tetagouche- continent. Such a scenario is consistent with the Exploits basin (Iapetus II). At the same time, sub- evidence of distinct Ordovician and Early Silurian duction of Tornquist oceanic lithosphere beneath loading cycles in the Taconian foredeep of the Avalon is thought to have been responsible for the central and southern Appalachians (Dorsch et al. Caradoc arc-back-arc volcanism in England, Wales 1994). Subsequent closure of the Rheic Ocean and southeast Ireland. This volcanism ceased in the probably involved a significant anticlockwise Ashgill, presumably by the onset of collision rotation of Gondwana to allow convergence between Baltica and Avalonia. between northwest Africa and Laurentia; hence the Geological data indicate that Laurentia and Laurentian margin became the locus of significant Avalon collided diachronously during the Silurian. dextral transpression during the Devonian (Dewey In the British Isles the progressive underthrusting 1982; Dalziel et al. 1994; van Staal & de Roo, of Avalonia beneath Laurentia is best reflected in 1995). The polarity of subduction of the Rheic the distribution of the Wenlock and Ludlow Ocean was probably also towards the northwest. foreland basin deposits in Central Ireland and Late Ordovician/Silurian arc-like magmatism in northern England (Hutton & Murphy, 1987; King, Avalonia (Greenough et al. 1993) and Ganderia 1994), and uplift of the Midland Valley. Defor- (Doig et al. 1990: Hepburn et al. 1995) is most mation of the foreland basin rocks during the simply explained by such a tectonic setting because Emsian in northern England and Wales may Laurentia, with its accreted terranes, represents the represent the final stage of Avalonian/Laurentian upper plate. Cessation of this arc magmatism collision. The presence of deep foreland basins on during the Late Silurian coincides with the onset of the Avalonian and Ganderian margins (van Staal & collision of Meguma (lower plate), either as a de Roo 1995) may also explain the patterns of microcontinent or as a promontory on the provinciality of Silurian ostracod fauna. Through- Gondwanan continent, with Avalonia during the Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

232 C.R. VAN STAAL ET AL.

Devonian. This collision probably was responsible Newfoundland: separate fault-controlled basins? for the Acadian Orogeny. Canadian Journal of Earth Sciences, 20, 345-354. --, MCKERROW, W. S. & COCKS, L. R. M. 1985. The Much of the data and ideas presented herein were tectonics and depositional history of the Ordovician developed by discussions and fieldtrips with our and Silurian rocks of Notre Dame Bay, Appalachian/Caledonian colleagues. We thank, particu- Newfoundland. Canadian Journal of Earth larly, Sandra Barr, Garry Boone, Robin Cocks, Steve Sciences, 22, 607-618. Colman-Sadd, Ken Currie, Benoit Dubr, Greg Dunning, BARR, S. M. & WHITE, C. E. 1996. Contrasts in Late Richard Fortey, Les Fyffe, Wes Gibbons, Lindsay Hall, Precambrian-Early Paleozoic Tectonothermal Chris Hepburn, Jim Hibbard, George Jenner, Paul History between Avalon Composite Terrane sensu Karabinos, Jonathan Kim, Tim Kusky, Shoufa Lin, stricto and other possible Peri-Gondwanan Terranes Michel Malo, John Morris, Bob Neuman, Brian O'Brien, in Southern New Brunswick and Cape Breton Paul Ryan, Dave Scofield, Scott Swinden, Alain Island, Canada. In: NANCE, D. & THOMPSON, M. Tremblay, Pablo Valverde, Ben van der Pluijm, Rob Van (eds) Avalonian and related Peri-Gondwanan der Voo, Joe Whalen, Hank Williams, Paul Williams, and Terranes of the Circum-North Atlantic. Geological John Winchester. We realize that not all of them will agree Society of America Special Paper, 304, 95-108. with our interpretations! Paul Hoffman, Steve Lucas, Jim ~, BEVIER, M. L., WHITE, C. E. & DOIG, R. 1994. Hibbard and Alain Tremblay are thanked for critically Magmatic history of the Avalon Terrane of Southern reading this manuscript. We would like to thank Peter New Brunswick, Canada, Based on U-Pb (Zircon) Coney, Rod Gayer and Robert Hall for their thoughtful Geochronology. Journal of Geology, 102, 399-409. reviews. We also wish to especially thank Claire Grainger BENNET, M. C., DUNNE, W. M. 8s TODD, S. P. 1989. and Deborah Lemkow for their skill and immense Reappraisal of the Cullenstown Formation: patience in drafting several of the figures. Van Staal implications for the Lower Paleozoic tectonic thanks the Department of Earth Sciences of Oxford history of SE Ireland. Geological Journal, 24, University for hospitality during his 1996 sabbatical. Mac 317-329. Niocaill gratefully acknowledges funding from an EU BERDAN, J. M. 1990. The Silurian and Early Devonian Marie Curie Research Fellowship. Much of this work was biogeography of ostracodes in North America. In: done through funding provided by the Canada-New MCKERROW, W. S. 8z SCOTESE, C. R. (eds) Brunswick cooperation agreement on Mineral develop- Palaeozoic Palaeogeography and Biogeography, ment 1990-1994: The Canada-Newfoundland cooper- Geological Society of London Memoir, 12, ation agreement on Mineral development 1990-1994, 223-231. 1994-1995. This is Geological Survey of Canada BIRD, J. M. & DEWEY, J. E 1970. Lithosphere Plate- contribution 1997229. continental margin tectonics and the evolution of the Appalachian Orogen. Geological Society of America Bulletin, 81, 1031-1060. BLOOMER, S. H., TAYLOR, B., MACLEOD, C. J., STERN, R. References J., FRYER, P., HAWKINS, J. W. & JOHNSON, L. 1995. AFrALION, M., VAN BREEMEN, O. 8Z BOWLS, D. R. 1984. Early arc volcanism and the ophiolite problem: a Age constraints on basement of the Midland Valley perspective from drilling in the Western Pacific. In: of Scotland. Transactions of the Royal Societ3' of TAYLOR, B. & NATLANDJ. (eds) Active margins and Edinburgh: Earth Sciences, 75, 53-64. marginal basins of the western Pacific, Geophysical ALEINIKOFF, J. N., WALTER, M. 8z FANNING, C. M. 1995. Monograph, 88, 1-30. U-Pb ages of zircon, monazite and sphene from BLUCK, B. J. 1984. Pre-Carboniferous history of the rocks of the Massabesic gneiss complex and Midland Valley of Scotland. Transactions of the Berwick Formation, New Hampshire and Royal Society of Edinburgh, 75, 275-295. Massachusetts. Geological Socie~ of America, --, HALLIDAY, A. N., AFrALION, M. & MACINTYRE, R. Abstracts with Programs, 27, 26. M. 1980. Age and origin of the Ballantrae ophiolite ANDRr~, L., HERTOGEN, J. & DEUTSCH, S. 1986. and its significance to the Caledonian Orogeny and Ordovician-Silurian magmatic provinces in Ordovician time scale. Geology, 9, 331-333. Belgium and the Caledonian orogeny in Middle BOILLOT, G., GIRARDEAU, J. & KORNPROBST, J. 1988. Europe. Geology, 14, 879-882. Rifting of the Galicia margin: crustal thinning and ANDRI~ASSON, P. G. & ALBRECHT, L. 1995. Derivation of emplacement of mantle rocks on the seafloor. 500 Ma eclogites from the passive margin of Baltica Proceedings of the Ocean Drilling Program, and a note on the tectonometamorphic heterogeneity Scientific Results, 103, 741-756. of eclogite-bearing crust. Geological Magazine, BOND, G. C., NICKESON, P. A. & KOMINZ, M. A. 1984. 132, 729-738. Break-up of a supercontinent between 625 Ma and ARMSTRONG, H. A., OWEN, A. W., SCRUTIrON, C. T., 555 Ma: New evidence and implications for CLARKSON, E. N. K. & TAYLOR, C. M. 1996. continental histories. Earth and Planetary Science Evolution of the Northern Belt, Southern Uplands: Letters, 70, 325-345. implications for the Southern Uplands Controversy. BOONE, G., DOTY, D. T. & HEIZLER, M. T. 1989. Hurricane Journal of the Geological Society, London, 153, Mountain Formation Mrlange: description and 197-205. tectonic significance of a Penobscottian ARNo~ R. J. 1983. Sedimentology of Upper Ordovician- accretionary complex. In: Studies in Maine Silurian sequences on New World Island, Geology, Maine Geological Survey, 2, 33-83. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 233

BOUDETrE, E. L. 1982. Ophiolite assemblage of Early COISH, R. A., HICKEY, R. & FREY, F. A. 1982. Rare earth Paleozoic age in central Maine. In: ST JULmN, E & element geochemistry of the Betts Cove ophiolite, B~LAND, J. (eds) Major Structural Zones and Faults Newfoundland: complexities in ophiolite formation. of the Northern Appalachians, Geological Geochimica et Cosmochimica Acta, 46, 2117-2134. Association of Canada Special Paper, 24, 43-66. COLMAN-SADD, S. P., HAYES, J. P. & KNIGHT, I. 1990. BOWRING, S. A., GROTZINGER, J. P., ISACHSEN, C. E., Geology of the Island of Newfoundland. KNOLL, A. H., PELECHATY, S. M. & KOLOSOV, E Newfoundland Department of Mines and Energy, 1993. Calibrating rates of Early Cambrian Map 90-01 evolution. Science, 261, 1293-1298. --, STONE, P., SWINDEN, H. S. & BARNES, R. E 1992a. CASEY, J. E & DEWEY, J. E 1984. Initiation of subduction Parallel geological development in the Dunnage zones along transform and accreting plate Zone of Newfoundland and the Lower Paleozoic boundaries, triple junction evolution, and forearc terranes of southern Scotland: an assesment. spreading centres-implications for ophiolitic Transactions of the Royal Society of Edinburgh, geology and obduction. In: GASS, I. G., LIPPARD, S. Earth Sciences, 83, 571-594. J. 8z SHELTON, A. W. (eds) Ophiolites and Oceanic --, DUNNING, G. R. & DEC, T. 1992b. Dunnage-Gander Lithosphere, Geological Society of London Special relationships and Ordovician orogeny in central Paper, 13, 131-144. Newfoundland: a sediment provenance and U/Pb -- & KIDD, W. S. E 1981. A parallochthonous group of study: American Journal of Science, 292, 317-355. sedimentary rocks unconformably overlying the COOPER, A. H., RUSHTON, A. W. A, MOLYNEUX, S. G., Bay of Island ophiolite complex, north Arm HUGHES, R. A., MOORE, R. M. & WEBB, B. C. 1995. Mountain, Newfoundland. Canadian Journal of The stratigraphy, correlation, provenance and Earth Sciences, 18, 1035-1050. palaeogeography of the Skiddaw Group CAWOOD, P. A. 8z SUHR, G. 1992. Generation and (Ordovician) in the English Lake District. obduction of ophiolites: constraints from the Bay of Geological Magazine, 132, 185-211. Islands Complex, Western Newfoundland. CURRIE, K. L. 1995. The northeastern end of the Dunnage Tectonics, 11, 884-897. Zone in Newfoundland. Atlantic Geology, 31, , VAN GOOL, J. A. M. & DUNNING, G. R. 1996. 25-38. Geological development of eastern Humber and --, VAN BREEMEN, O., HUNT, P. A. & VAN BERKEL, J. T. western Dunnage zones: Comer Brook - Glover 1992. The age of high-grade gneisses south of Island region, Newfoundland. Canadian Journal of Grand Lake, Newfoundland. Atlantic Geology, 28, Earth Sciences 33, 182-198. 153-161. CHANDLER, E W., SULLIVAN,R. W. & CURRIE, K. L. 1987. DALLMEYER, R. D. 1977. Diachronous ophiolite The age of the Springdale Group, western obduction in western Newfoundland: evidence from Newfoundland, and correlative rocks-evidence for a 40Ar/39Ar ages of the Hare Bay metamorphic Llandovery overlap assemblage in the Canadian aureole. American Journal of Science, 277, 61-72. Appalachians. Transactions of the Royal Society of DALZlEL, I. W. D. 1992. On the organization of American Edinburgh, Earth Sciences, 78, 41-49. plates in the Neoproterozoic and the breakout of CrtANNELL, J. E. T. & MCCABE, C. 1992. Palaeomagnetic Laurentia. GSA Today, 2, 237-241. data from the Borrowdale Volcanic Group:volcano- --., DALLA SALDA, L., CINGOLANI, C. • PALMER, P. tectonics and Late Ordovician palaeolatitudes. 1996. The Argentine Precordillera: A Laurentian Journal of the Geological Socier), London, 149, Terrane? Penrose Conference Report. GSA Today, 881-888. 6(2), 16-18. CLIFF, R. A., YARDLEY, B. W. D. & BUSSY, E R. 1996. ---, DALLA SALDE, L. H. & GAHAGAN, L. M. 1994. U-Pb and Rb-Sr geochronology of magmatism and Paleozoic Laurentia-Gondwana interaction and the metamorphism in the Dalradian of Connemara, origin of the Appalachian-Andean mountain western Ireland. Journal of the Geological Societ3; system. Geological Society of America Bulletin, London, 153, 109-120. 106, 243-252. CLIFT, P. & RYAN, P. D. 1994. Geochemical evolution of DAVID, J. 8z GARIEPY, C. 1990, Early Silurian orogenic an Ordovician island arc. Journal of the Geological andesites from the central Quebec Appalachians. Society, London, 151, 329-342. Canadian Journal of Earth Sciences, 27, 632-643.

-- 1995. Volcaniclastic sedimentation and volcanism --, MARQUIS, R. t~ TREMBLAY, A. 1993. U-Pb during rifting of western Pacific backarc basins. In: geochronology of the Dunnage Zone in the TAYLOR, B. & NATLANDJ. (eds) Active margins and southwestern Quebec Appalachians. Geological marginal basins of the western Pacific, Geophysical Society of America, Annual Meeting Program with Monograph, 88, 67-96. Abstracts, A-485. COCKS, L. R. M. & FORTEY, R. A. 1982. Faunal evidence DE METS, C., GORDON, R. G, ARGUS, D. E & STEIN, S. for oceanic separations in the Paleozoic of Britain. 1990. Current plate motions. Geophysical Journal Journal of the Geological Society, London, 139, International, 101,425-478. 467-480. DEAN, W. T. 1985. Relationships of Cambro-Ordovician

-- & -- 1990. Biogeography of Ordovician and faunas in the Caledonide-Appalachian region, with Silurian faunas. In: MCKERROW, W. S. & SCOTESE, particular reference to trilobites. In: GAYER, R. A. C. R. (eds) Paleozoic Paleogeography and (ed.) The Tectonic Evolution of the Caledonide- Biogeography, Geological Society of London Appalachian Orogen, Braunschweig/Wiesbaden, Memoir, 12, 97-104. Fredrich Vieweg & Sohn, 17-47. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

234 C.R. VAN STAAL ET AL.

DEC, T. & SWINDEN,H. S. 1994. Lithostratigraphic model, DUNNING, G. R. 1987. Geology of the Annieopsquotch geochemistry and sedimentology of the Cotrells Complex, southwest Newfoundland. Canadian Cove Group, Buchan-Roberts Arm volcanic belt, Journal of Earth Sciences, 24, 1162-1174. Notre Dame Subzone. Current Research, -- & COUSINEAU. P. A. 1990. U/Pb ages of single Newfoundland Department of Mines and Energ 3, zircons from Chain Lakes Massif and a correlative Geological Survey Branch Report. 94-1, 77-100. unit in ophiolitic mrlange in Quebec. Geological , & DUNNING, R. G. 1997 Lithostgratigraphy Society of America, Abstracts with Programs, and geochemistry of the Cotrells Cove Group, Northeastern Section, 22, 13.

Buchans-Robert Arm volcanic belt: new constraints -- & KROGH, T. E. 1985. Geochronology of ophiolites for the paeotectonic setting of the notre dame of the Newfoundland Appalachians. Canadian subzone, Newfoundland Appalachians. Canadian Journal of Earth Sciences, 22, 1659-1670. Journal of Earth Sciences, 34, 86-103. & 1991. Stratigraphic correlation of the DEMPSTER, T. J. & BLUCK, B. J. 1991. Age and tectonic Appalachian Ordovician using advanced U-Pb significance of the Bute amphibolite, Highland zircon geochronology techniques. In: BARNES, C, R. Border Complex, Scotland. Geological Magazine, & WILLIAMS, S. H. (eds) Advances in Ordovician

152, 77-80. Geology. Geological Survey of Canada Paper, 90-9, DEWEY, J. F. 1969. Evolution of the Appalachian- 85-92. Caledonian Orogen, Nature, 222, 124-129, & O'BRIEN, S. J. 1989. Late Proterozoic-Early

-- 1975. Finite plate evolution: Implications for the Paleozoic crust in the Hermitage flexure, evolution of the rock masses at plate margins. Newfoundland Appalachians: U/Pb ages and American Journal of Science, 275-A, 260-284. tectonic significance. Geology, 17, 548-551.

-- -- 1980. Episodicity, sequence and style at convergent & PEDERSEN, R. B. 1988. U/Pb ages of ophiolites plate boundaries. In: STRANGWAY, D. W. (ed.) The and arc-related plutons of the Norwegian Continental Crust and its Mineral Deposits. Caledonides: Implications for the development of Geological Association of Canada Special Paper, Iapetus. Contributions to Mineralogy and 20, 553-573. Petrolog3, 98, 13-23. 1982. Plate tectonics and the evolution of the British --, BARR, S. M., RAESIDE, R. P. & JAMIESON, R. A. Isles. Journal of the Geological Societ3; London, 1990. U-Pb zircon, titanite and monazite ages in the 139, 317-414. Bras d'Or and Aspy terranes of Cape Breton Island, 1988. Lithospheric stress, deformation and tectonic Nova Scotia: implications for magmatic and cycles: the disruption of Pangea and the closure of metamorphic history. Geological Socie O" of America Tethys. In: AUDLEY-CHARLES,M. G. & HALLAM, A. Bulletin, 102, 322-330. (eds) Gondwana and Tethvs. Geological Society of --. I~AN, B. F., THURLOW,J. G. & SWlNDEN H. S. 1987. London Special Publications, 37, 23-40. Geochronology of the Buchans, Roberts Arm, and

-- & LAMB, S. H. 1992. Active tectonics of the Andes. Victoria Lake Groups and Mansfield Cove Tectonophysics, 205, 79-95. Complex, Newfoundland. Canadian Journal of

-- & RYAN, E D. 1990. The Ordovician Evolution of Earth Sciences. 24, 1175-1184. the South Mayo Trough, Western Ireland. Tectonics, --. O'BRIEN S. J., COLMAN-SADD, S. P., BLACKWOOD, 9, 887-901. R. E. DICKSON, W. L.. O'NEILL. E E & KROGH, T.

-- & SHACrd.ETON, R. M. 1984. A model for the E. 1990. Silurian orogeny in the Newfoundland evolution of the Grampian tract in the early Appalachians. Journal of Geolog3, 98, 895-913. Caledonides and Appalachians. Nature, 312, --, O'BRIEN, B. H., HOLDSWORTH, R. E. & 115-121. TUCKER, R. D. 1993. Chronology of Pan-African, , HELMAN, M. L., TURCO, E., HUTTON, D. H. W. & Penobscot and Salinic shear zones on the KNorr, S. D. 1989. Kinematics of the western Gondwanan margin, Northern Appalachians. Mediterranean. In: COWARD, M. E, DIETRJCH, D. & Geological Socieo' of America Annual Meeting, PARK, R. G. (eds) Alpine Tectonics. Geological Abstracts with Programs, 421-422. Society of London Special Publications, 45, --. SWlNDEN, H. S., KLAN, B. E, EVANS, D. T. W. & 265-283. JENNER. G. A. 1991. A Cambrian island arc in DOIG, R., NANCE, R. D., MURPHY,J. B. & CASSEDAY, R. P. Iapetus; geochronology and geochemistry of the 1990. Evidence for Silurian sinistral accretion of Lake Ambrose volcanic belt, Newfoundland Avalon composite terrane in Canada. Journal of the Appalachians. Geological Magazine, 128, 1-17. Geological Society, London, 147, 927-930 EGAL. E., GUERROT, C., LE GOFF, E., THII~BLEMONT,D. & DORSCH, J., BAMBACH,R. K. & DRIESE, S. G. 1994. Basin- CHANTRAINE, J. 1996. The Cadomian Orogeny rebound origin for the "Tuscarora unconformity" in revisited in northern Brittany (France). bt: NANCE, southwestern Virginia and its bearing on the nature D. & THOMPSON. M. (eds) Avalonian and related of the taconic Orogeny. American Journal of Peri-Gondwanan Terranes of the Circum-North Science, 294, 237-255. Atlantic. Geological Society of America Special DUBE, B., DUNNING, G. R., LAUZJI~RE,K. & RODDICK, J. C. Paper, 304, 281-318. 1996. New insights into the Appalachian Orogen ELDERS, C. E 1987. The provenance of granite boulders in from geology and geochronology along the Cape conglomerate of the Northern and Central Belts of Ray fault zone, southwest Newfoundland. the Southern Uplands of Scotland. Journal of the Geological Society of America Bulletin, 108, Geological Socieo; London, 144, 853-863. 101-116. ELIAS, E. M., MACINTYRE, R. N. & LEAKE, B. E. 1988. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 235

The cooling history of Connemara western Ireland, dispersal of the Avalon superterrane. In: D'LEMOS, from K-Ar and Rb-Sr studies. Journal of the R. S., STRACHAN, R. A. & TOPLEY, C. G. (eds) The Geological Societ), London, 145, 649-660. Cadomian Orogeny. Geological Society of London ELLIOTT, C. G., DUNNING, G. R. & WILLIAMS, E E 1991. Special Publication, 51,407-423. New constraints on the timing of deformation in --, TIETZSCH-TYLER, D., HORAK, J. M. & MURPHY,E C. eastern Notre Dame Bay, Newfoundland, from U/Pb 1994. Precambrian rocks in Anglesey, southwest zircon ages of felsic intrusions. Geological Society Llyn and southeast Ireland. In: GIBBONS, W. & of America Bulletin, 103, 125-135. HARRIS, A. L. (eds) A revised correlation of EVANS, D. Y. W., KLAN, B. F. & DUNNING, G. R. 1990. Precambrian rocks in the British Isles. Geological Geological studies, Victoria Lake Group, Central Society of London, Special report, 22, 75-84. Newfoundland. Current Research, Ne~,foundland GOODWIN, L. B. & WILLIAMS, P. F. 1990. Strike-slip Department of Mines and Energy, Geological motion along the Baie Verte Lone. Atlantic Survey Branch, Report 90-1, 131-144. Geology, 26, 170. FITCH, T. J. 1972. Plate convergence, transcurrent faults, GRAY, J. R. & YARDLEY, B. W. D. 1979. A Caledonian and internal deformation adjacent to south-east Asia blueschist from the Irish Dalradian. Nature, 278, and the western Pacific. Journal of Geophysical 736-737. Research, 77, 4432--4460. GREENOUGH, J. D., KAMO, S. L. & KROGH, T. E. 1993. A FITZ, T. J. & MOENEH, R. H. 1996. Tectonic significance Silurian U-Pb age for the Cape St. Mary's sills, of the sheeted dike complex in the Chickwolnepy Avalon Peninsula, Newfoundland, Canada: intrusions in northern New Hampshire. Geological implications for Silurian orogenesis in the Avalon SocieO, of America, Annual Meeting Program with Zone. Canadian Journal of Earth Sciences, 30, Abstracts, 28, 54. 1607-1612. FLINN, D., MILLER, J. A. & RODDOM, D. 1991. The age of GRUNOW, A., HANSON, R. & WILSON, T. 1996. Were the Norwick hornblendic schists of Unst and Fetlar aspects of Pan-African deformation linked to and the obduction of the Shetland ophiolite. Scottish Iapetus opening? Geology, 24, 1063-1066. Journal of Geology, 27, 11-19. HALL, L. M., VAN STAAL, C. R. & WILLIAMS, H. 1994. FORTEY, R. A. & COCKS, L. R. M. 1992. The Early Ordovician structural evolution of SW- Paleozoic of the north Atlantic Region as a test case Newfoundland. Geological Association of Canada / for the use of fossils in continental reconstruction. Mineralogical Association of Canada Annual Tectonophysics, 206, 147-158. Meeting Program with Abstracts, Waterloo, , HARPER, D. A. T., INGHAM, J. K.. OWEN, A. W. & Ontario, A46. RUSHTON, A. W. A. 1995. A revision of Ordovician HALL, R. 1996. Reconstructing Cenozoic South-East series and stages from the historical type area. Asia. In: HALL, R. & BLUNDELL,D. J. (eds) Tectonic Geological Magazine, 132, 15-30. Evolution of Southeast Asia, Geological Society of FRIEDRICH, A. M., BOWRING, S. A. & HODGES, K. V. 1997. London Special Publications, 106, 153-184. U-Pb geochronological constraints on the duration -- 1997. Cenozoic Plate Reconstructions of Southeast of arc magmatism and metamorphism from Asia. Geological Society of London Special Connemara, Irish Caledonides. Terra Nova, 9, Publications, 126, 11-23. abstract supplement 1, p.331. HAMILTON, P. J., BLUCK, B. J. & HALLIDAY, A. N. 1984. FYFFE, L. R. 1995. Fredericton Belt. In: WILLIAMS, H. Sm-Nd ages from the Ballantrae complex, SW (ed.) Geology of the Appalachian-Caledonian Scotland. Transactions of the Royal Society of Orogen in Canada and Greenland, Geological Edinburgh: Earth Sciences, 75, 183-187. Survey of Canada, Geology of Canada, 6, 351-354. HAMILTON, W. 1970. The Uralides and the motion of the -- & PICKERILL, R. K. 1993. Geochemistry of Upper Russian and Siberian platforms. Geological Society Cambrian Lower Ordovician black shale along a of America Bulletin, 81, 2553-2576. northeastern Appalachian transect. Geological HARPER, D. A. T. & PARKES, M. A. 1989. Palaeontoiogical Society of America Bulletin, 105, 897-910. constraints on the definition and development of the GALLAGHER, V. 1989. The occurrence, textures, Irish Caledonide terranes. Journal of the Geological mineralogy and geochemistry of a chromite-bearing Societ), London, 146, 413-4 15. serpentenite. Geological Survey of Ireland Bulletin. & HOLY, A. N. 1990. Intra-Iapetus 4, 89-98. brachiopods from the Ordovician of eastern Ireland: , O'CONNOR, P. J. & AF'rALION M. 1994. Intra- implications for Caledonide correlation. Canadian Ordovician deformation in southeast Ireland: Journal of Earth Sciences, 27, 1757-1761. evidence from the geological setting, geochemical HARPER, J. C. & RAST, N. 1964. The faunal succession and affinities and U-Pb zircon age of the Croghan volcanic rocks of the Ordovician near Bellowstown, Kinshelagh granite. Geological Magazine, 131. County Meath. Proceedings of the Royal Irish 669-684. Academy, B64, 1-23. GEOLOGICAL WORLD ATLAS 1976. Commision for the HARRIS, A. L. 1969. The relationship of the Leny Geological Map of the World and UNESCO, Paris. Limestone to the Dalradian. Scottish Journal of GIBBONS, W. & BALL, M. J. 1991. A discussion of Monian Geology, 5, 187-190. Supergroup stratigraphy in northwest Wales. HARRIS, D. H. M. 1995. Caledonian transpressional Journal of the Geological Societ3, l~mdon, 148, terrane accretion along the Laurentian margin in Co. 5-8. Mayo, Ireland. Journal of the Geological Socieo; 1990. Transcurrent ductile shear zones and the London, 152, 797-806. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

236 C.R. VAN STAAL ET AL.

HEPBURN, J. C., DUNNING, G. R. & HON, R. 1995. amphibole identified from the Newfoundland Geochronology and regional tectonic implications Appalachians-its occurrence, composition and of Silurian deformation in the Nashoba terrane, possible tectonic implications. Nature, 165, southeastern New England, USA. In: HIBBARD, J., 428--430. VAN STAAL, C. R. • CAWOOD, P. (eds) Current JENNER, G. A. & FRYER, B. J. 1980. Geochemistry of the Perspectives in the Appalachian-Caledonian upper Snooks Arm basalts, Burlington Peninsula, Orogen. Geological Association of Canada Special Newfoundland: evidence against formation in an Paper, 41, 349-366. island arc. Canadian Journal of Earth Sciences, 17, HIBBARD, J. & WILLIAMS, H. 1979. Regional setting of the 888-900. Dunnage Melange in the Newfoundland & SW~NDEN, H. S. 1993. The Pipestone Pond Appalachians. American Journal of Science, 279, Complex, central Newfoundland: complex 993-1021. magmatism in an eastern Dunnage Zone ophiolite. 1982. Significance of the Bale Verte Flexure, Canadian Journal of Earth Sciences, 30, Newfoundland. Geological Society of America 434--448. Bulletin, 93, 790-797. , DUNNING, G. R., MALPAS, J., BROWN, M. & BRACE, 1988. Stratigraphy of the Fleur de Lys Belt, T. 1991. Bay of Islands and Little Port complexes, northwest Newfoundland. In: WINCHESTER, J. A. revisited: age, geochemical and isotopic evidence (ed.) Later Proterozoic stratigraphy of the Northern confirm suprasubduction-zone origin. Canadian Atlantic Regions. Blackie, London, 20(O21 I. Journal of Earth Sciences, 28, 1635-1652. HODYCH, J. P. & BUCHAN, K. L. 1994. Paleomagnetism of JOHNSON, R. J. E., VAN DER PLUIJM, B. A. & VAN DER VOO, the Early Silurian Cape St. Mary's Sills of the R. 1991. Paleomagnetism of the Moreton's Harbour Avalon Peninsula of Newfoundland, Canada. EOS, Group, Northeastern Newfoundland Appalachians: 75, 128. Evidence for an Early Ordovician island Arc near HOFFMAN, P. E 1996. Evolutionary model of the Southern the Laurentian margin of Iapetus. Journal of Brazilide Ocean. EOS, 77, $87. Geophysical Research, 96, 689-701. 1991. Did the breakout of Laurentia turn KARABINOS, P. & TUCKER, R. D. 1992. The Shelburn Falls Gondwanaland inside-out? Science, 252, 1409- arc in western Massachusetts and Connecticut: the 1412. lost arc of the Taconic orogeny. Geological Society HOLDSWORTH, R. E. 1994. Structural evolution of the of America Annual Meeting, Abstracts with Gander-Avalon terrane boundary: a reactivated Programs, 24, 288-289. transpression zone in the NE Newfoundland ~, STOLL, H. M., HEPBURN, J. C. & TUCKER, R. D. Appalachians. Journal of the Geological Society, 1993. The New England segment of the Laurentian London, 151,629-646. margin: continuous tectonic activity from Early HOLLOCHER, K. 1993. Geochemistry and origin of Ordovician through Early Devonian. Geological volcanics in the Ordovician Partridge Formation, Society of America Annual Meeting, Abstracts with Bronson Hill Anticlinorium, west-central Programs, A- 178. Massachusetts. American Journal of Science, 293, KARSON, J. & DEWEY, J. F. 1978. Coastal complex, 671-721. western Newfoundland: an Early Ordovician HOLTZMAN, B., TRZCIENSKI, W. & GROMET, L. P. 1996. oceanic fracture zone. Geological Society of New tectonic constraints on the Boil Mountain America Bulletin, 89, 1037-1049. Complex/Chain lakes unit contact, central western KEAN, B. E, EVANS, D. T. W. & JENNER, G. A. 1995. Maine. Geological Society of America Annual Geology and mineralization of the Lushs Bight Meeting, Abstracts with Programs, 28, 65. Group. Newfoundland Department of Natural HOLUB, E V., KLAPOVA,H., BLUCK, B. J. & BOWES, D. R. Resources, Geological Survey Report, 95-2. 1984. Petrology and geochemistry of post- KEEN, C. E., KEEN, M. J., NICHOLS, B., ETAL. 1986. Deep obduction dykes of the Ballantrae complex, SW seismic reflection profile across the northern Scotland. Transactions of the Royal Society of Appalachians. Geology, 14, 141-145. Edinburgh, Earth Sciences, 75, 211-224. KELLEY, S. & BLUCK, B. J. 1989. Detrital mineral ages HORAK, J. M., DOIG, R., EVANS., J. A. & GIBBONS, W. from the Southern Uplands using Ar laser probe. 1996. Avalonian magmatism and terrane linkage: Journal of the Geological Society, London, 146, new isotopic data from the Precambrian of North 401--403. Wales. Journal of the Geological Society, London, KIDD, W. S. F., DEWEY, J. E & NELSON, K. D. 1977. 153, 91-100. Medial Ordovician Ridge subduction in central HUTTON, D. H. W. 1987. Strike-slip terranes and a model Newfoundland. Geological Society of America for the evolution of the British and Irish Annual Meeting, Abstracts with Programs, 9, Caledonides. Geological Magazine, 124, 405--425. 283-284. HUTFON, D. H. W., AFTALION, M. & HALLIDAY, A. N. KIM, J. & JACOBI, R. D. 1996. Geochemistry and tectonic 1985. An Ordovician ophiolite in County Tyrone, implications of Hawley Formation meta-igneous Ireland. Nature, 315, 210-212. units: northwestern Massachusetts. American -- & MURPHY, E C. 1987. The Silurian of the Southern Journal of Science, 296, 1126-1174. Uplands and Ireland as a successor basin to the end- KING, L. M. 1994. Subsidence analysis of eastern Ordovician closure of Iapetus. Journal of the Avalonian sequences: implications for Iapetus Geological Societ); London, 144, 765-772, 1987. closure. Journal of the Geological Society, London, JAMIESON, R. A. 1977. The first metamorphic sodic 151,647--657. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 237

KNELLER, B. C., KING, L. M. & BELL, A. M. 1993. perspectives in the Appalachian-Caledonian Foreland basin development and tectonics on the Orogen. Geological Association of Canada Special northwest margin of western Avalonia. Geological Paper, 41, 147-160. Magazine, 130, 691--697. LEGGE'rr, J. K. 1987. The Southern Uplands as an KOKELAAR, P. 1986. Petrology and geochemistry of accretionary prism: the importance of analogues in the Rhobell Volcanic Complex: amphibole- reconstructing palaeogeography. Journal of the dominated fractionation at an early Ordovician Arc Geological SocieO', London, 144, 737-752. Volcano in North Wales. Journal of Petrology, 27, --, MCKERROW, W. S. • CASEY, D. M. 1982. The 887-914. anatomy of a Lower Paleozoic accretionary 1988. Tectonic controls of Ordovician arc and forearc:the Southern Uplands of Scotland. marginal basin volcanism in Wales. Journal of the Geological Society London Special Publication, 10, Geological Society, London, 145, 759-775. 494-520. , HOWELLS, M. F., BEVINS, R. E., ROACH, R. A. & LEO, G. H. 1985. Trondhjemite and metamorphosed DUNKLEY, P. N. 1984. The Ordovician marginal quartz keratophyre tuff of the Ammonoosuc basin of Wales. In: KOKELAAR,P. & HOWELLS, M. F. volcanics (Ordovician), western New Hampshire (eds) Marginal Basin Geology, Geological Society and adjacent Vermont and Massachusetts. of London Special Publication, 16, 245-269. Geological Socie O' of America Bulletin, 96, KRABBENDAM, M., LESLIE, A. G., CRANE, A. & GOODMAN, 1493-1507. S. 1997. Generation of the Tay nappe, by large-scale LIN, S., VAN STAGE, C. R. c9~ DUaL B. 1994, Promontory- SE-directed shearing. Journal of the Geological promontory collision in the Canadian Appalachians. Society, London, 154, 15-24. Geology, 22, 897-900. KUMARAPELI, P. S., DUNNING, G. R., PINTSON, H. & LISS, M. J., VAN DER PLUIJM, B. A. & VAN DER Voo, R. SHAVER, J. 1989. Geochemistry and U-Pb zircon 1993. Avalonian proximity of the Ordovician age of comenditic metafelsites of the Tibbit Hill Miramichi terrane, northern New Brunswick, Formation, Quebec Appalachians. Canadian northern Appalachians: Palaeomagnetic evidence Journal of Earth Sciences, 26, 1374-1383. for rifting and back-arc basin formation at the KUSKY, T. M., CHOW, J. S. & BOWRING, S. A. 1997. Age southern margin of Iapetus. Tectonophysics, 227, and origin of the Boil Mountain Ophiolite and 17-30. Chain Lakes Massif, Maine: implications for the LORENTZ, B. E. 1984. Mud-magma interactions in the Penobscotian Orogeny. Canadian Journal of Earth Dunnage melange, Newfoundland. In: NOKELAAR, Sciences, 34, 6464554. P. & HOWEEES, M. E (eds) Marginal Basin Geology. LAERANCE, B. t~ WILLIAMS, P. F. 1992. Silurian Geological Society London Special Publications, deformation in eastern Notre Dame Bay, 16, 271-277. Newfoundland: Canadian Journal of Earth LYONS, J. B., BOUDETI'E, E. L. & ALEINIKOFF, J. N. 1982. Sciences, 29, 1899-1914. The Avalonian and Gander zones in Central Eastern LAIRD, J., TRZCIENSKI, W. E. JR & BOTHNER, W. A. 1993. New England. In: ST JULIEN, P. & BI~LAND, J. (eds) High-pressure, Taconian and subsequent Major Structural Zones and Faults of the Northern polymetamorphism of southern Quebec and Appalachians. Geological Association of Canada northern Vermont. Department of Geology and Special Paper, 24, 43-66. Geography, University of Massachusetts, MAC NIOCAILL, C. 8z SMETHURST, M. A. 1994. Palaeozoic Contribution, 67-2, 1-32. palaeogeography of Laurentia and its margins: a LAMBERT, R. STJ., HOLLAND, J. G. & LEGGET, J. K. 1981. reassessment of palaeomagnetic data. Geophysical Petrology and tectonic setting of some Ordovician Journal International, 116, 715-725. volcanic rocks from the Southern Uplands of --, VAN DER PLUIJM, B. A. & VAN DER VOO, R. 1997. Scotland. Journal of the Geological Societ3', Ordovician paleogeography and the evolution of the London, 138, 421-436. Iapetus ocean. Geology, 25, 159-162. LANDING, E., BOWRING, S. M., FORTEY, R. A. & DAVIDEK, MACLACHLAN, K. & DUNNING, G. R. 1998. U-Pb ages K. 1997. U-Pb date from Avalonian Cape Breton and tectonic setting of Middle Ordovician calc- Island and geochronology calibration of the Early alkaline and within plate volcanic rocks of the Wild Ordovician. Canadian Journal of Earth Sciences, Bight Group, northeastern Dunnage Zone, 34, 724-730. Newfoundland Appalachians: implications for the LAURENT, R. & HEBERT, R. 1989. The volcanic and evolution of the Gondwanan margin. Canadian intrusive rocks of the Quebec Appalchians Journal of Earth Sciences (in press). ophiolites and their island-arc setting. Chemical MAX, M. D. & RODOICK, J. C. 1989. Age of Geology, 77, 265-286. metamorphism in the Rosslaare Complex, S.E. LEAKE, B. E. 1989. The metagabbros, orthogneisses and Ireland. Proceedings of the Geologists' Association, paragneisses of the Connemara complex, western 100, 113-121. Ireland. Journal of the Geological SocieD', London, ---, BARBER, A. J. & MARTINEZ, J. 1990, Terrane 146, 575-596. assemblage of the Leinster Massif, SE Ireland, LEE, C. B. & WILLIAMS, H. 1995. The Teakettle and during the Lower Paleozoic. Journal of the Carmanville m61anges in the Exploits subzone of Geological SocieO', London, 147, 1035-1050. northeast Newfoundland: recycling and diapiric MCCAFFREY, R. 1996. Slip partitioning at convergent plate emplacement in an accretionary prism. In: HIBBARD, boundaries of southeast Asia. In: HALL, R. & J., VAN STAAL, C. R. & CAWOOD, P. (eds) Current BEUNDELL, D. J. (eds) Tectonic Evolution of Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

238 C.R. VAN STAAL ETAL.

Southeast Asia, Geological Society of London Mechanics and Transport Properties of Rocks, Special Publications, 106, 3-18. Academic Press, 435--459. MCCAIG, A. & REX, D. 1995. Argon-40/Argon-39 ages MORRIS, J. H. 1987. The Northern Belt of the Longford- from the Lewis Hills massif, Bay of island complex, Down Inlier, Ireland and Southern Uplands: an west Newfoundland. In: HIBBARD, J., VAN STAAL, C. Ordovician back-arc basin. Journal of the R. & CAWOOD, P. (eds) Current perspectives in the Geological SocieO', London, 144, 773-786. Appalachian-Caledonian Orogen. Geological MURPHY, J. B. & NANCE, R. D. 1989. Model for the Association of Canada Special Paper, 41, evolution of the Avalonian-Cadomian belt. Geology, 137-146. 17, 735-738. MCCAUSLAND, P. J. A., HODYCH, J. P. & DUNNING, G. R. NELSON, K. D. 1981. Mrlange development in the Boones 1997. Evidence from western Newfoundland for the Point Complex, north-central Newfoundland. final breakup of Rodinia? U-Pb age and Canadian Journal of Earth Sciencess, 18, 433-442. palaeolatitude of the Skinner Cove volcanics. NEUMAN, R. B. 1967. Bedrock geology of the Shin Pond Geological Association of Canada~Mineralogical and Stacyville Quadrangles, Penobscot County, Association of Canada Annual Meeting, Program Maine. United States Geological Surve); with Abstracts, Ottawa, Ontario, A-96. Professional Paper, 524-1, 37pp.

MCCONNELL, B. J., STILLMAN, C. J. & HERTOGEN, J. 1991. -- 1984. Geology and paleobiology of islands in the An Ordovician basalt to peralkaline rhyolite Ordovician Iapetus Ocean: review and implications. fractionation series from Avoca, Ireland. Journal of Geological SocieD' of America Bulletin, 94, the Geological Socie~,, London, 148, 711-718. 1188-1201.

MCKENZIE, D. P. & MORGAN, W. J. 1969. Evolution of -- 1994. Late Ordovician (Ashgill) Foliomena Fauna triple junctions. Nature, 224, 125-133. brachiopods from northeastern Maine. Journal of MCKERROW, W. S. & CAMPBELL, C. J. 1960. The Paleontology, 68, 1218-1234. stratigraphy and structure of the Lower Paleozoic NOBLE, S. R., TUCKER, R. D. & PHARAOH, T. C. 1993. rocks of north-west Galway. Scientific Proceeding Lower Paleozoic and Precambrian igneous rocks of the Royal Dublin Socie~.', 1, 27-52. from eastern England and their bearing on late & COCKS, L. R. M. 1976. Progressive faunal Ordovician closure of the Tornquist Sea: constraints migration across the Iapetus Ocean. Nature, 263, from U-Pb and Nd isotopes. Geological Magazine, 304-306. 130, 835-846. & ELDERS, C. F. 1989. Movements on the Southern NOBLET, C. & LEFORT, L. P. 1990. Sedimentological Upland Fault. Journal of the Geological SocieO; evidence for a limited separation between Armorica London, 146, 393-395. and Gondwana during the Early Ordovician. -- 1987. The Southern Uplands controversy. Journal of Geology, 18, 303-306. the Geological Societ); London, 144, 735-736. NOWLAN, G. S., MCCRACKEN, A. D. & McLEoO, M. J. , DEWEY, J. F. & SCOTESE, C. R. 199l. The 1997. Tectonic and palaeogeographic significance Ordovician and Silurian development of the Iapetus of Ordovician conodonts from the Letang area, Ocean. In: BASSETT, M. G., LANE, P. & EDWARDS, D. Avalon Terrane, southwestern New Brunswick. (eds) The Murchison Symposium. Special Paper in Canadian Journal of Earth Sciences, 34, Palaeontology, 44, 165-178. 1521-1537. , SCOTESE, C. R. & BRASIER, M. D. 1992. Early O'BRIEN, B. H. 1991. Geological development of the Cambrian reconstructions. Journal of the Exploits and Notre Dame Subzones in the New Geological Socie~, London, 149, 599-606. Bay area (parts of NTS 2E/6 and 2E/ll), Notre MCLEOD, M. J., RUITENBERG, A. A. & KROGH, T. E. 1992. Dame Bay, Newfoundland. Current Research, Geology and U-Pb geochronology of the Annidale Newfoundland Department of Mines and Energy, Group, southern New Brunswick: Lower Geological Survey Branch Report, 91-1, 155-166. Ordovician volcanic and sedimentary rocks formed 1992. Internal and external relationships of the near the southeastern margin of Iapetus Ocean. South Lake Igneous complex, north-central Atlantic Geology, 28, 181-192. Newfoundland: Ordovician and later tectonism in , WINCHESTER, J. A. & RUITENBERG, A. A. 1994. the Exploits Subzone? Current Research, Geochemistry of the Annidale Group: implications Newfoundland Department of Mines and Energy, for the tectonic setting of Lower Ordovician Geological Survey Branch Report, 92-1, 159-69. volcanism in southwestern New Brunswick. 1993. A mapper's guide to Notre Dame Bay's folded Atlantic Geology, 30, 87-95. thrust faults: evolution and regional development. MEERT, J. G. & VAN DER VOO, R. 1997. The assembly of Current Research, Newfoundland Department of Gondwana (800-550 Ma), Journal of Geodynamics, Mines and Energy, Geological Survey Branch 23, 223-235. Report, 93-1,279-291. MINCER, R. G. & O'NIONS, R. K. 1984. The provenance --, SW1NDEN, H. S., DUNNING, G. R., WILLIAMS, S. H. and crustal residence age of British sediments in & O'B~EN, E H. C. 1997. A peri-Gondwanan arc- relationship to palaeogeographic reconstructions. backarc complex in Iapetus: Early-Mid Ordovician Earth and Planetary Science Letters, 68, 459--470. evolution of the Exploits Group, Newfoundland. MOLNAR, P. 1992. Bruce-Goetze strength profiles, the American Journal of Science, 297, 220-272. partitioning strike-slip and thrust faulting at zones O'BRIEN, S. J., O'BRIEN, B. H., DUNNING, G. R. & of oblique convergence, annd the stress-heat flow TUCKER, R. D. 1996. Late Neoproterozoic paradox of the San Andreas Fault. In: Fault Avalonian and related peri-Gondwanan rocks of the Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 239

Newfoundland Appalachians. In: NANCE, D. & America, The Geology of North America, Boulder, THOMPSON, M. (eds) Avalonian and related Peri- Colorado, F-2. Gondwanan Terranes of the Circum-North Atlantic. ROBINSON, P. & HALL, L. M. 1980. Tectonic synthesis of Geological Society of America Special Paper, 304, southem New England. In: WONES, D. R. (ed.) 9-28. Proceedings 'The Caledonides in the U.S.A. ', IGCP OSH[N, I. O. & CROCKET, H. 1986. The geochemistry and Project 27: Caledonide Orogen, Virginia petrogenesis of ophiolitic volcanic rocks from Lac Polytechnic & State University Memoir, 2, 73-82. de l'Est, Thetford Mines Complex, Quebec, --. & TUCKER, R. D. 1996. The Bronson Hill magmatic Canada. Canadian Journal of Earth Sciences, 23, arc, New England: myth and reality. Geological 202-213. Societ3' of America, Northeastern Section, Annual OWEN, A. W. & CLARKSON, E. N. K. 1992. Trilobites from meeting Program with Abstracts, 28, 94. Kilbucho and Wallace's Cast and the location of the ROGERS, G., DEMPSTER, T. J., BLUCK, B. J. & TANNER, P. Northern Belt of the Southern Uplands during the W. G. 1989. A high-precision U-Pb age for the Ben Late Ordovician. Scottish Journal of Geology, 28, Vurich granite: implications for the evolution of the 3-17. Scottish Dalradian Supergroup. Journal of the PARNELL, J. t~ SWAINBANK, I. 1984. Interpretation of Pb Geological Societ)', London, 146, 789-798. isotope compositions of galenas from the Midland --, PATERSON, B. A., DEMPSTER, T. J. & REDWOOD, S. Valley of Scotland and adjacent regions. D. 1994. U-Pb geochronology of the 'newer' Transactions of the Royal Society of Edinburgh, gabbros, NE Grampians. Symposium on Caledonian Earth Sciences, 75, 85-96. Terrane Relationships in Britain, September, 1994 PEGLER, G., DAS, S. & WOODHOUSE, J. 1995. A ROGERS, N. & VAN STAAL, C. R. 1997. Comparing the seismological study of the eastern New Guinea and Bathurst Mining Camp to the Japan Sea and the western Solomon Sea regions and its tectonic Okinawa Trough: ancient, recent and active back- implications. Geophysical Journal International, arcs. Geological Association of Canada/ 122, 961-981. Mineralogical Association of Canada Annual PELECHATY, S. M. 1996. Stratigraphic evidence for the Meeting, Program with Abstracts, Ottawa, Ontario, Siberia-Laurentia connection and Early Cambrian A-127. rifting. Geology, 24, 719-722. RYAN, P. D. & DEWEY, J. E 1991. A geological and PHILLIPS, E. R., BARNES, R. P., MERRIMAN, R. J. & FLOYD, tectonic cross-section of the Caledonides of western J. D. 1995. The tectonic significance of Ordovician Ireland. Journal of the Geological SocieO; London, basic igneous rocks in the Southern Uplands, 148, 173-180. Southwest Scotland. Geological Magazine, 132, --, MAX, M. D. & KELLY, T. J. 1983. The 549-556. petrochemistry of the basic volcanic rocks of the PICKETT, J. W. 1987. Geology and geochemistry of the South Connemara Group (Ordovician), Western Skidder basalt. In: KIRKHAM, R. V. (ed.) Buchans Ireland. Geological Magazine, 120, 141-152. Geology. Geological Survey of Canada Special --, SOPER, N. J., SNYDER, D. B., ENGLAND, R. W. & Paper, 86-24, 195-218. HtrrrON, D. H. W. 1995. The Antrim-Galway Line: PINET, N. & TREMBLAY,A, 1995. Tectonic evolution of the a resolution of the highland Border Fault enigma of Quebec-Maine Appalachians: from oceanic the Caledonides of Britain and Ireland. Geological spreading to obduction and collision in the Northern Magazine, 132, 171-184. Appalachians. American Journal of Science, 295, SANDERS, I. S., VAN CALSTEREN, P. W. C. & 173-200. HAWKESWORTH, C. J. 1984. A Grenville Sm-Nd age POLLOCK, S. G. 1993. Terrane sutures in the Maine for the Glenelg eclogite in northwest Scotland. Appalachians, USA and adjacent areas. Geological Nature, 312, 439--440. Journal, 28, 45-67. SCOTESE, C. R. & MCKERROW, W. S. 1990. Revised world POTTS, S. S., VAN DER PLUIJM, B. A. & VAN DER WOO, R. maps and introduction. In: MCKERROW, W. S. & 1993. Paleomagnetism of the Ordovician Bluffer SCOTESE, C. R. (eds) Palaeozoic Palaeogeography Pond Formation: Palaeogeographic Implications for and Biogeography. Geological Society of London the Munsungun Terrane of Northern Maine. Journal Memoir, 12, 1-21. of Geophysical Research, 98, 7987-7996. SENGOR, A. M. C. & NATAL'IN, B. A. 1996. Paleotectonics , & 1995. Paleomagnetism of the of Asia: fragments of a synthesis. In: YIN A. & Pennington Mountain terrane: A near Laurentian HARRISON, M. (eds) Tectonic evolution of Asia. back arc basin in the Maine Appalachians. Journal Cambridge Press, 486--640. of Geophysical Research, 100, 10 003-10 011. 1979. Mid-Mesozoic closure of permo-Triassic PRIGMORE, J. K., BULLER, A. J. & WOODCOCK, N. H. 1997. Tethys and its implications. Nature, 279, 590-593. Rifting during separation of eastern Avalonia from SMETHURST, M. A., KHRAMOV, A. N. & TORSVIK, T. H. Gondwana: evidence from subsidence analysis. 1998. The Neoproterozoic and Palaeozoic Geology, 25, 203-206. palaeomagnetic data for the Siberian Platform: from RaNKIN, D. W., DRAKE, A. & RATCLIFFE, N. M. 1990. Rodinia to Pangea. Earth Science Reviews, 43, Plate 2: Geologic map of the U. S. Appalachians 1-24. showing the Laurentian margin and the Taconic SOPER, N. J. & HARRIS, A. L. 1997. Proterozoic orogeny Orogen. In: HATCHER, R. D. JR, THOMAS, W. A. & questioned: a view from Southern Highland VIELE, G. W. (eds) The Appalachian-Ouachita fieldtrips and workshops, 1995-1996. Scottish Orogen in the United States. Geological Society of Journal of Geology, 33, 187-190. Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

240 C. R. VAN STAAL ET AL.

& WOODCOCK, N. 1990. Silurian collision and TAIRA, A., OKADA, H., WHITAKER,J. H. McD & SMITH, A. sediment dispersal patterns in southern Britain. J. 1982. The Shimanto Belt of Japan: Cretaceous to Geological Magazine, 127, 527-542. Lower Miocene active margin sedimentation. In: 1994. Was Scotland a Vendian RRR junction? LEGGETT, J. K. (ed.) Trench-forearc Geology: Journal of the Geological Societ3; London, 151, Sedimentation and Tectonics on Modern and 579-582. Ancient Active Margins. Geological Society of SPENCER, C., GREEN, A., MOREL, A. ET AL 1986. The London Special Publications, 10, 5-26. extension of Grenville basement beneath the TANNER, B. & MEISSNER, R. 1996. Caledonian northern Appalachians: results from the Quebec- deformation upon southwest Baltica and its tectonic Maine seismic reflection and refraction surveys. implications: alternatives and consequences. Tectonics, 8, 677-696. Tectonics, 15, 803-812. SPRAY, J. G. & DUNNING, G. R. 1991. A U/Pb age for the TECTONICMAP OF THE WORLD 1985. Exxon Production and Shetland Islands oceanic fragment, Scottish Research Company. Caledonides: evidence from anatectic plagiogranites THOROCOOD, E. J. 1990. Provenance of the pre-Devonian in layer 3 shear zones. Geological Magazine, 128, sediments of England and Wales: Sm-Nd isotopic 667-671. evidence. Journal of the Geological Society, STANLEY, R. & RATCU~E, N. M. 1985. Tectonic synthesis London, 147, 591-594. of the Taconian Orogeny in western New England. THORPE, R. S., BECKINSALE, R. D., PATCHETT,P. J., PIPER, Geological Society of America Bulletin, 96, J. D. A., DAVIES, G. R. & EVANS, J. A. 1984. Crustal 1227-1250. growth and late Precambrian-early Palaeozoic plate STERN, R. J. & BLOOMER, S. H. 1992. Subduction zone tectonic evolution of England and Wales. Journal of infancy: examples from the Eocene Izu-Bonin- the Geological Socieo', London, 141,521-536. Mariana and Jurassic California arcs. Geological THURLOW, J. G., SPENCER, C. P., BOERNER, D. E., REED, L. SocieD' of America Bulletin, 104, 1621-1636. E. & WRIGHT, J. A. 1992. Geological interpretation STONE, P. & EVANS, J. A. 1995. Nd-isotopic study of of a high resolution reflection seismic survey at the provenance patterns across the British sector of Buchans Mine, Newfoundland. Canadian Journal the Iapetus Suture. Geological Magazine, 132, of Earth Sciences, 29, 2022-2037. 571-580. TIETZSCH-TYLER, D. & PHILLIPS, E. 1989. Correlation of , FLOYD, J. D., BARNES, R. P. & LINTERN, B. C. 1987. the Monian Supergroup in NW Anglesey with the A sequential back-arc and foreland basin thrust Cahore Group in SE Ireland. Journal of the duplex model for the Southern Uplands of Scotland. Geological Societ3; London, 146, 417-418. Journal of the Geological Societ3; London, 144, TORSVIK. T. H., SMETHURST, M. A., BRIDEN, J. C. & 753-764. STURT, B. A. 1990. A review of Palaeozoic data SUHR, G. & CAWOOD, P. A. 1993. Structural history of from Europe and their palaeogeographical ophiolite obduction, Bay of Islands, Newfoundland. implications. In: MCKERROW, W. S. & SCOTESE, C. Geological Society of America Bulletin, 105, R. (eds) Palaeozoic Palaeogeography and 399-4 10. Biogeography,.Geological Society of London SULLIVAN, R. W. & VAN STAAL, C. R. 1996. Preliminary Memoir, 12, 25-41. chronostratigraphy of the Tetagouche and Fournier --, MEERT, J. G., VAN DER VOO, R., MCKERROW, Groups in northern New Brunswick, Canada; In: W. S., BRASIER, M. D., STURT, B. A. & Radiogenic Age and Isotopic Studies, Report 9, WALDERHAUG, H. J. 1996. Continental break-up and Geological Survey of Canada, Paper 1995-E 43-56. collision in the Neoproterozoic and Palaeozoic - A , -- & LANGTON, J. P. 1990. U-Pb zircon ages of tale of Baltica and Laurentia. Earth Science plagiogranite and gabbro from the ophiolitic Reviews, 40, 229-258. Devereaux Formation, Fournier group, northeastern TREMBLAY, A. & PINET, N. 1994. Distribution and New Brunswick. Geological Survey of Canada characteristics of Taconian and Acadian Paper, 89-2, 119-122. deformation, southern Quebec Appalachians. SWINDEN, H. S. & THORPE, R. I. 1984. Variations in style Geological Societ)' of America Bulletin., 106, of volcanism and massive sulphide deposition in 1172-1181.

Early-Middle Ordovician island arc sequences of -- & ST JULIEN, P. 1990. Structural style and evolution the New Brunswick Central Mobile Belt. Economic of a segment of the Dunnage Zone from the Quebec Geology, 79, 1569-1619. Appalachians and its tectonic implications. 1988. Geology and economic potential of the Geological Society of America Bulletin, 102, Pipestone Pond area (12A/1 NE; 12A/8 E), central 1218-1229.

Newfoundland. Newfoundland Department of -- 1992. Tectonic and accretionary history of Taconian Mines, Geological Survey Branch Report, 88-2. oceanic rocks of the Quebec Appalachians. , JENNER, G. A. & SZYBINSKI, Z. A. 1997. Magmatic American Journal of Science, 292, 229-252. and tectonic evolution of the Cambrian-Ordovician --, LAFLI~CHE,M. R., McNtrrr, R. H. & BERGERON, M. Laurentian margin of Iapetus: Geochemical and 1994. Petrogenesis of Cambro-Ordovician isotopic constraints from the Notre Dame Subzone, subduction-related granitic magmas of the Quebec Newfoundland. In: SINHA, K., WHALEN, J. B. & Appalachians, Canada. Chemical Geology, 113, HOGAN, J. (eds) Magmatism in the Appalachian 205-220. Orogen. Geological Society of America Memoir, TRENCH, A., TORSVIK, T. H. & MCKERROW, W. S. 1992. 191,337-365. The palaeogeographic evolution of Southern Britain Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

EVOLUTION OF IAPETUS 241

during early Paleozoic times: a reconciliation of Geological Society of America Bulletin, 103, palaeomagnetic and biogeographic evidence. 1564-1575. Tectonophysics, 201, 75-82. VAN STAAL, C. R. 1987. Tectonic setting of the TRZr W. E., JR 1976. Crossitic amphibole and its Tetagouche Group in northern New Brunswick: possible tectonic significance in the Richmond area, implications for plate tectonic models of the southeast Quebec. Canadian Journal of Earth northern Appalachians. Canadian Journal of Earth Sciences, 13, 711-714. Sciences, 24, 1329-1351.

--, RODGERS, J. & GUIDOTTI, C. V. 1992. Alternative -- 1994. The Brunswick subduction complex in the hypotheses for the Chain Lakes Massif, Maine and Canadian Appalachians: record of the Late Quebec. American Journal of Science, 292, Ordovician to Late Silurian collision between 508-532. Laurentia and the Gander margin of Avalon. TUCKER, R. D. & MCKERROW, W. S. 1995. Early Tectonics, 13, 946-962.

Paleozoic chronology: a review in light of new U- -- & DE Roo, J. A. 1995. Mid-Paleozoic tectonic Pb zircon ages from Newfoundland and Britain. evolution of the Appalachian Central Mobile Belt in Canadian Journal of Earth Sciences, 32, 368-379. Northern New Brunswick, Canada: Collision, & ROBINSON, P. 1990. Age and setting of the extensional collapse and dextral transpression. In: Bronson Hill magmatic arc: a re-evaluation based HIBBARD, J., VAN STAAL, C. R. & CAWOOD, P. (eds) on U-Pb zircon ages in southern New England. Current perspectives in the Appalachian- Geological Society of America Bulletin, 102, Caledonian Orogen. Geological Association of 1404-1419. Canada Special Paper, 41, 95-114. --, O'BmEN, S. J. & O'BRmN, B. H. 1994. Age and --, SULLIVAN R. W. & WHALEN, J. B. 1996a. implications of Early Ordovician plutonism in the Provenance and tectonic history of the Gander type area of the Bay du Nord Group, Dunnage Zone, Margin in the Caledonian/ Appalachian Orogen: southern Newfoundland Appalachians. Canadian implications for the origin and assembly of Journal of Earth Sciences, 31,351-357. Avalonia. In: NANCE, D. & THOMPSON, M. (eds) TURNER, R. D. & TURNER, S. 1974. Thelodonts from the Avalonian and related Peri-Gondwanan Terranes of Upper Silurian of Ringerike, Norway. Norsk the Circum-North Atlantic. Geological Society of Geologisk Tidsskrift, 54, 183-192. America Special Paper, 304, 347-367. UPTON, B. G. J., ASPEN, P. & HUNTER, R. H. 1984. --, HALL, L., SCOFIELD, D. & VALVERDE, P. 1996b. Xenoliths and their implications for the deep Geology Port aux Basques, Newfoundland, scale 1: geology of the Midland Valley of Scotland and 25 000 (part ofNTS 1 l-O/11). Geological Survey of adjacent regions. Transactions of the Royal Society Canada Open File, 3165. of Edinburgh, Earth Sciences, 75, 65-70. --, LIN, S., HALL, L., VALVERDE, P. & GENKIN, M. VALVERDE-VAQUERO,P., DUNNING, G. R. & VAN STAAL, C. 1996c. Geology Rose Blanche, Newfoundland, R. 1998. The Margaree orthogneiss: an Arenig-early scale 1:25 000 (NTS 11-O/10). Geological Survey Llanvirn arc/backarc complex developed on the of Canada Open File, 3219.

peri-Gondwanan margin of SW Newfoundland. -- & WILLIAMS, H. 1991. Dunnage Zone-Gander Zone Geological Society of America Annual meeting relationships in the Canadian Appalachians. Program with Abstracts, 30, 82. Geological SocieD' of America Annual meeting VAN DER PLUIJM, B. A. 1986, Geology of eastern New Program with Abstracts, 23, 143. World Island, Newfoundland: an accretionary --, RAVENHURST,C., WINCHESTER, J. A., RODDICK, J. C. terrane in the northeastern Appalachians. & LANGTON, J. P. 1990. Post Taconic blueschist Geological Socie~ of America Bulletin, 97, suture in the northern Appalachians of northern 932-945. New Brunswick, Canada. Geology, 18, 1073-1077. , VAN DER VOO, R. & TORSVIK, T. H. 1995. --, WINCHESTER, J. A. & BEDARD, J. H. 1991. Convergence and subduction at the Ordovician Geochemical variations in Ordovician volcanic margin of Laurentia. In: HIBBARD, J. P., VAN STAAL, rocks of the northern Miramichi Highlands and their C. R. & CAWOOD, P. A. (eds) Current Perspectives tectonic significance. Canadian Journal of Earth in the Appalachian-Caledonian Orogen. Geological Sciences, 28, 1031-1049. Association of Canada Special Paper, 41, 127-136. WASOWSKI, J. J. & JACOBI, R. D. 1985. Geochemistry and -- & VAN STAAL, C. R. 1988. Characteristics and tectonic significance of the mafic volcanic blocks in evolution of the central mobile belt, Canadian the Dunnage mdlange, north central Newfoundland. Appalachians. Journal of Geology, 96, 535-547. Canadian Journal of Earth Sciences, 22, VAN DER Woo, R. 1988. Paleozoic paleogeography of 1248-1256. North America, Gondwana and intervening WELLENSIEK, M. R., VAN DER PLUIJM, B. m., VAN DER Woo, displaced terranes: comparisons of paleomagnetism R. & JOHNSON, R. J. 1990. Tectonic history of the with paleoclimatological and biogeographical Lunksoos composite terrane in the Maine patterns. Geological Society of America Bulletin, Appalachians. Tectonics, 9, 719-734. 100, 311-324. WELLINGS, S. A. 1998. Structure and metamorphism of , JOHNSON, R. J. E., VAN DER PLUIJM, B. A. & syn-tectonic mafic intrusions from Connemara, KNUTSON, L. C. 1991. Palaeogeography of some western Ireland. Journal of the Geological SocieO', vestiges of Iapetus: paleomagnetism of the London. 155, 25-38. Ordovician Robert's Arm, Summerford, and WEST, D. P., LUDMAN, A. & LUX, D. R. 1992. Silurian age Chanceport Groups, Central Newfoundland. for the Pocomoonshine Gabbro-Diorite, south- Downloaded from http://sp.lyellcollection.org/ by guest on September 26, 2021

242 c.R. VAN STAAL ET AL.

eastern Maine and its regional tectonic implications. BI~LAND, J. (eds) Major Structural Zones and Faults American Journal of Science, 292, 253-273. of the Northern Appalachians. Geological WHALEN, J. B. 1989. The Topsails igneous suite, western Association of Canada Special paper, 24, 43--66. Newfoundland: an Early Silurian subduction-related --, COLMAN-SADD, S. P. & SWINDEN, H. S. magmatic suite.'? Canadian Journal of Earth 1988.Tectonic-stratigraphic subdivisions of central Sciences, 26, 2421-2434. Newfoundland. Geological Survey of Canada , CURRIE, K. L. & VAN BREEMEN, O. 1987. Episodic Paper, 88-1B, 91-98. Ordovician-Silurian plutonism in the Topsails --, CURRIE, K. L. & PIASECKI, M. A. J. 1993. The Dog terranne, western Newfoundland. Transactions of Bay Line: a major Silurian tectonic boundary in the Royal Socie~ of Edinburgh, Earth Sciences, 78, northeast Newfoundland. Canadian Journal of 17-28. Earth Sciences, 30, 2481-2494. , JENNER, G. A., LONGSTAFFE, E J. 8z GARII~PY, C. WILLIAMS, P. E, ELLIOTT, C. G. & LAFRANCE, B. 1988. 1997a. Implications of granitoid geochemical and Structural geology and melanges of eastern notre isotopic (Nd, O, Pb) data from the Cambro- Dame Bay, Newfoundland. Geological Association Ordovician Notre Dame arc for the evolution of the of Canada, Guidebook Trip B2, 60pp. Central Mobile Belt, Newfoundland Appalachians. WILLIAMS, S. H., BOYCE, W. D. & COLMAN-SADD, S. P. In: SINHA, K., WHALEN, J. B. & HOGAN, J. (eds) 1992. A new Lower Ordovician (Arenig) faunule Magmatism in the Appalachian Orogen. Geological from the Coy Pond complex, central Newfoundland, Society of America Memoir, 191, 367-395. and a refined understanding of the closure of the , VAN STAAL, C. R., LONGSTAFFE, E J., GARIEPY, C. & Iapetus Ocean. Canadian Journal of Earth Sciences, JENNER, G. A. 1997b. Insights into tectonostrati- 29, 2046-2057. graphic zone identification in southwestern --., HARPER, D. A. T., NEUMAN, R. B., BOYCE, W. D. & Newfoundland based on isotopic (Nd, O, Pb) and MAC NIOCAILL, C. 1995. Lower Paleozoic fossils geochemical data. Atlantic Geology, 33, from Newfoundland and their importance in 231-241. understanding the history of the Iapetus Ocean. In: WHITEHEAD, J., REYNOLDS, P. H. & SPRAY, J. G. 1996. HIBBARD, J., VAN STAAL, C. R. & CAWOOD, P. (eds) 40Ar/39Ar age constraints on taconian and acadian Current perspectives in the Appalachian- events in the Quebec Appalachians. Geology, 24, Caledonian Orogen. Geological Association of 359-362. Canada Special Paper, 41, 115-126. WILLIAMS, D., ARMSTRONG, H. A. & HARPER, D. A. 1988. WILLIAMS, T. M., HENNEY, P. J., STONE, P. & LINTERN, B. The age of the South Connemara Group, Ireland and C. 1996. Rare earth element geochemistry of Lower its relationship to the southern Uplands Zone of Palaeozoic turbidites in the British trans-Iapetus Scotland and Ireland. Scottish Journal of Geology, zone: provenance patterns and basin evolution. 24, 279-287. Scottish Journal of Geology, 32, 1-8. WILLIAMS, H. 1978. Tectonic-Lithofacies map of the WILLIAMSON, B. E • KARABINOS,P. 1993. Constraints on Appalachian Orogen. Memorial Universit3' of ages of Taconian and Acadian deformation from Newfoundland, St. John ~, Newfoundland, Map no. zircon evaporation ages of felsic plutons from 1. western Massachusetts. Geological Society of -- 1995. Temporal and spatial divisions. In: WILLIAMS, America, Annual meeting program with Abstracts, H. (ed.) Geology of the Appalachian-Caledonian 25, 90. Orogen in Canada and Greenland. Geological WINCHESTER, J. A. & VAN STAAL, C. R. 1994. The Survey of Canada, Geology of Canada, 6, 21-44. chemistry and tectonic setting of Ordovician

-- & HISCOTr, R. N. 1987. Definition of the Iapetus volcanic rocks in northern Maine and their rift-drift transition in western Newfoundland. relationships with contemporary volcanic rocks in Geology, 15, 1044 1047. northern New Brunswick. American Journal of

-- & PAYNE, J. G. 1975. The Twillingate granite and Science, 294, 641-662.

nearby volcanic groups: an island arc complex in -- & -- 1995. Volcanic and sedimentary terrane northeast Newfoundland. Canadian Journal of correlation between the Dunnage and Gander zones Earth Sciences, 12, 982-995. of the Canadian Appalachians and the British PIASECKI, M. A. J. 1990. The Cold Spring Caledonides reviewed. In: HIBBARD,J., VAN STAAL, Melange and a possible model for Dunnage-Gander C. R. & CAWOOD, P. (eds) Current perspectives in zone interaction in central Newfoundland. the Appalachian-Caledonian Orogen. Geological Canadian Journal of Earth Sciences, 27, 1126- Association of Canada Special Paper, 41, 95-114. 1134. YARDLEY, B. W. D. & SENIOR, A. 1982. Basic magmatism -- & ST JULIEN, P. 1982. The Baie Verte-Brompton in Connemara, Ireland: Evidence for a volcanic arc? Line: Early Paleozoic continent ocean interface in Journal of the Geological Socie~, London, 139, the Canadian Appalachians. In: ST JULIEN, P. & 67-70.