The Ordovician History of the Iapetus Ocean in Britain
Total Page:16
File Type:pdf, Size:1020Kb
Journal of the Geological Society, London, Vol. 148, 1991, pp. 423-425, 3 figs. Printed in Northern Ireland SHORT PAPER al. 1990), from which palaeomagnetic data indicate high southerly latitudes (Perroud & Van der Voo 1985; Perroud The Ordovician history of the Iapetus et al. 1986). Therefore, although several palaeogeographic Ocean in Britain: new palaeomagnetic reconstructions depict S Britain as a peri-Gondwanan block constraints in early Ordovician time (Pickering et al. 1988; Scotese & McKerrow 1990), nopalaeomagnetic data have been T. H. TORSVIK' & A. TRENCH^ available to quantify its palaeo-position. Similarly, palaeo- Department of Earth Sciences, University of magnetic data have been insufficient to discriminate whether Oxford, Oxford, OX1 3PR, UK S Britain (part of EasternAvalonia; Soper et al. 1987; 'Present address: Geological Survey of Norway, McKerrow1988) formed a constituent of the Armorican P.B. 3006 Lade N-7002 Trondheim, Norway plate (Armorican and Iberian Massifs: Van der Voo 1979; 2Present address: Dept of Geology, University of Perroud et al. 1984) during the Ordovician period. Western Australia, Nedlands, Perth, WA6009, To addressthese uncertainties, 15 siteswere sampled within a sequence of late Tremadoc-early Arenig andesitic Australia lavas and volcaniclastic sediments at Treffgarne, Dyfed, SW Wales (Traynor 1988). This sequence dips on average 30" to thenorth. Low-grade metamorphism(Oliver 1988) and New late Tremadoc-edy Arenig palaeomagnetic results from SW tectonismis most likely Acadian (Woodcock et al. 1988), Wales imply that S Britain (part of Eastern Avalonia) occupied a although Hercyniandeformation cannot be excluded southerly latitude of c. 60"s in early Ordovician times. When com- (Hancock et al. 1981). bined with Scottish Ordoviaan pdneomagnetic data,which indicate a 150s latitude, the results indicate that the British sector of the Pdaeomagnetic results. Stepwise-thermaldemagnetization IapetusOcean reached n latitudinalwidth of c. 5OOO km in experiments (161 samples) identify twomagnetization Tremadoc-Arenigtimes, which was reduced to c. 3Wkm by Llanvirn-Llandeilo (mid-Ordovician) times. components. The lower unblocking-temperature com- 'Lbe new data resolve two previous controversies in Palaeozoic ponent (200-500°C) has sub-horizontal to shallow negative palaeogeography.First, the high southerlypalaeolatitude links inclination duesouth. We interpret this component, Avdonia to Gondwana, marginal to WAfrica, thus reconciling termed S, as the Permo-Carboniferous(Hercynian) over- codlicting reconstructions basedupon either pdaeomagnetic or print observedelsewhere in Wales(McClelland-Brown fnundlfaaes evidence done. Second, reliable LIanvirn pdaeomag- 1983; Briden & Mullan1984; McCabe & Channell 1990). netic dataimply that Avdonia hadrifted northwards by Arenig The higher unblocking-temperaturecomponent (>500"C, time,whereas Armorica remained proximal to northernAfrica Fig. la), termed T, is primary,and has the following re- throughout the Ordovician. The combined data therefore establish manence properties. that Avdonia andArmoricn formed separate micro-continents Reversely-magnetized(1) polarity (tilt-corrected when rifting from Gondwana. declination = 298", inclination = 75", alpha 95 = 5.5", ob- served in 10 sites; Fig. lb: Pole N56",E306" & dp/dm = 9/ 10). Palaeomagnetic, biogeographic and palaeoclimatic evidence (2) Clastsfrom anintra-formational conglomerate are in agreement in establishing the existence of the Iapetus (Thomas & Cox 1924)reveal high-unblocking magnetiza- Ocean (Wilson 1966; Harland & Gayer 1972) in Ordovician tions which are directionally-consistent within,but differ times, butestimates of its size havevaried substantially between, individual boulders (Fig. lc).These magnetiza- (Dewey 1969, 1971; McKerrow & Cocks 1976; Smith et al. tions have equivalent unblocking temperaturesto com- 1981; Briden et al. 1984; McKerrow 1988;Torsvik et al. ponent T. Conversely, component S displays consistent 1990a; Scotese & McKerrow 1990). particular,In directions within the boulders. biogeographic considerations have consistently been inter- (3) Laboratory unblocking temperatures coverboth preted to represent wider oceanicseparation than that magnetiteand hematite ranges, but generallydisplay indicated by palaeomagneticdata. This discrepancy is discreteunblocking between 520-550°C (Fig. la), which explicable in that early palaeomagnetic studiesmay not have indicates the dominance of low-Ti titanomagnetite formed employed sufficiently detailed demagnetization experiments during high-temperaturedeuteric oxidation in volcanic to remove Late Palaeozoicsecondary magnetizations of rocks. shallowinclination (seeBriden & Mullan1984, for discussion). Discussion. The new palaeomagnetic data imply that Recently,however, a palaeomagneticresult from the SouthernBritain occupieda latitude of c. 60"s inlate Shelve inlier (McCabe & Channell 1990), as well as a Tremadoc-early Arenigtimes (Fig. 2).Compared with re-compilation of palaeomagnetic datafrom S Britain Ordovician data from the Scottish terranes (c. 15"S, Tor- (Trench & Torsvik 1991), indicates that the British sector of svik et al. 1990a), this implies that the intervening Iapetus the Iapetus Ocean covered approximately 30" of latitude in Oceanreached a latitudinal width of near 5000 km (Fig. mid-Ordovician (Llanvirn-Llandeilo) times (N Britain 15"S, 2) * S Britain 45"s). Prior to the Llanvirn however, no reliable Acomparison with Ordovicianpalaeomagnetic data palaeomagnetic data exist for Southern Britain. from Gondwana (Van der Voo 1988; Bachtadse & Briden Inearly Ordovician times (Tremadoc-Arenig), fauna1 1990),which allow the reconstruction of the west evidence indicatesIapetus to havereached its maximum Gondwananmargin, suggests Eastern Avalonia to have extent(McKerrow & Cocks 1976, 1986), andearly been positioned close to west Africa, possibly in the vicinity Ordovician sedimentary facies suggesta provenance link of Mauritania (Fig. 2) while Armorica was attached to NW with the Armorican Massif (Fortey & Owens 1987; Noblet et Africa (Torsvik et al. 1990a). This scenario attributes late 423 Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/3/423/4892076/gsjgs.148.3.0423.pdf by guest on 23 September 2021 424 T. H. TORSVIK & A.TRENCH N W.UP a I COMPONENT T SITE MEANS N ;;;:I 5rnAIM Fig. 2. Continental reconstruction for early Ordovician (late Tremadoc-early Arenig) times based on palaeomagnetic data Fig. 1. (a) Representative orthogonal projection: Andesite lava. (optional longitude). Southern Britain and other Avalonian Primary (T)magnetization component is indicated. Open (Closed) segments positioned in latitude accordingto the Treffgarne pole. A squares refer to projection in the vertical (horizontal) plane. In-Situ preferred palaeo-orientation has been used to maintain the facing of co-ordinates. Component S not resolved in this example. the Iapetus margin in Britain. In Figs 2 & 3, Western Avalonia (b) Site-mean component (T)directions after structural tilt (WA) is positioned according to data from Eastern Avalonia (EA) correction. Cone of 95% confidence about the total mean direction in a Bullard et al. fit (1965), sinceno structurally reliable early and is dotted. Equal-angle projection. mid-Ordovician data exist for Western Avalonia. Gondwana (c) Clast-mean component (T)directions from an intra- positioned according to Van der Voo (1988) (pole: 34"N,007"E, formational volcanic conglomerate. Cones of 95% confidence are African co-ordinates), Laurentia and Northern Britain combined indicated for each clast. Number of samples from each clast is (Torsvik et al. 19906, pole: 13"S,29"E, European co-ordinates in a shown. In (b) and (c), solid (open) squares represent lower (upper) Bullard et al. fit (1965)), Baltica (Torsvik er al. 19906, pole: 31", hemisphere directions. 086"E), Armorica (A) (Torsvik et al. 199oa, table 7,490 Ma pole: 30"N, 334"E), Siberia according to a mean pole of 30"N, 330"E Precambrian arc rocks in both Avalonia and Armorica (e.g. (Torsvik et al. 1990~).Equal-area polar projection. Scotese & McKerrow 1990) to a peripheral Gondwanan arc. At this time, Baltica was confined to southerly latitudes of 30-60" (Torsvik et al. 19906), whereasLaurentia and Siberia (Mongolianmargin facing north; Scotese & McKerrow1990; Khramov et al. 1981) retainedmore equatorial latitudes (Fig. 2). From this early Ordovician scenario, Eastern Avalonia thenhad rifted away fromGondwana into mid-southerly latitudes by mid-Ordovician (late Llanvirn-early Llandeilo) times (Fig. 3). This movement history is clearly recorded by palaeomagnetic datafrom volcanic rocks of the early Llanvirn Shelve (51"s: McCabe & Channel11990) and the lateLlanvirn Builth inliers (35"s: Briden & Mullan 1984; Trench et al. 1991)respectively. Conversely, the African Gondwanan margin and Armoricaremained in high Fig. 3. Continental reconstruction for mid-Ordovician (late Llanvirn-early Llandeilo) times. Latitudinal positioning of Gondwana and Siberia as Fig. 2. Avalonia from Trench & Torsvik (1991, table 2, 470 Ma pole: 12"N, 23"E), Baltica (Torsvik er al. 19906, combined path X and Y,470 Ma pole: 21"N, 32"E), Armorica (Torsvik er al. 1990a, table 7,470 Ma pole: 33"N, 345"E), Laurentia and Northern Britain (Torsvik er al. 19906, pole: 223, 19"E, European co-ordinates). Equal-area polar projection. Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/3/423/4892076/gsjgs.148.3.0423.pdf by guest on 23 September 2021 S H O R T PAPER