Palaeomagnetic insights into the evolution of Neotethyanoceanic crust in the eastern Mediterranean

ANTONY MORRIS1, MARK W. ANDERSON1, JENNIFER INWOOD1 &ALASTAIR H. F. ROBERTSON2

1School of Earth, Ocean and Environmental Sciences, University of Plymouth,Drake Circus, Plymouth PL4 8AA, UK (e-mail: antony.morris@plymouth.ac.uk)

2School of Geosciences, University of Edinburgh, West Mains Road,Edinburgh EH9 3JW, UK

Abstract: A synopsis of palaeomagnetic data from three Late Cretaceous eastern Mediter-ranean Tethyan ophiolites (Troodos, Hatay and Baër–Bassit) and their sedimentary coversequences is presented. These data provide valuable insights into the role of regional- andlocal-scale tectonic rotations in the geodynamic evolution of Neotethyan oceanic crust. Thegeologically earliest phases of tectonic rotation are documented in the Troodos ophiolite,where rotations around both subvertical and subhorizontal axes are readily related to thedevelopment of the spreading fabrics and structures during crustal genesis. Subsequent c. 74°anticlockwise intra-oceanic rotation of a ‘Troodos microplate’ has been quantified throughanalysis of the in situ sedimentary cover of the Troodos ophiolite. Results indicate that bulkanticlockwise rotation began soon after the cessation of spreading and ended by the end ofthe Eocene, with c. 50–60° of microplate rotation being over by the Maastrichtian, the timeat which ophiolite thrust sheets were emplaced onto the Arabian continental margin to theeast of Troodos. Recent results from the emplaced, structurally dismembered Baër–Bassitophiolite indicate extreme anticlockwise rotations of ophiolitic thrust sheets varying on akilometre scale. New data from the post-emplacement sedimentary cover confirm that onlya small component of these rotations is due to post-emplacement tectonism. Baër–Bassitrepresents the leading edge of the emplaced ophiolitic sheet. New data from the morecoherent section preserved in the Hatay ophiolite to the north demonstrate significantanticlockwise rotation. This is equivalent to the rotation of the most northerly part of theBaër–Bassit units to the south, and is of the same sense and magnitude as the pre-Maastrichtian phase of microplate rotation documented in the Troodos. This suggests acommon, intra-oceanic origin for the majority of the Troodos and Hatay rotations, and asignificant component of the more variable rotations observed in Baër–Bassit. Overall,therefore, the data support a model involving: (1) intra-oceanic rotation of a coherent regionof crust within the southern Neotethyan basin; this rotated unit is more areally extensivethan has previously been inferred from consideration of data from the Troodos ophiolitealone; (2) emplacement of part of the rotated unit onto the Arabian platform; (3) subsequentlocalized post-emplacement modification, related to the development of the current plateconfiguration.

Ophiolitic terranes provide unique opportunitiesfor investigating the structure and kinematicsof constructive plate margins, transformfaults and oceanic suture zones. The easternMediterranean–Middle East segment of theTethyan orogenic belt is marked by severalchains of ophiolites, the most prominent of whichare the Troodos (Cyprus), Baër–Bassit (Syria),Hatay (Turkey), Kermanshah and Neyriz (Iran),and Semail (Oman) ophiolites (Fig. 1). These areinterpreted as fragments of ocean lithosphereformed in a southern Neotethyan basin duringthe Late Cretaceous (e.g. Robertson 1998).

The aim of this paper is to present a synopsisof the major palaeomagnetic results obtained

from the Troodos, Hatay and Baër–Bassit ophio-lites (Fig. 2) over the last 20 years. The Troodosophiolite of Cyprus, in particular, has played akey role in developing and testing concepts ofplate tectonics (Gass 1968, 1980; Moores & Vine1971; Robertson & Xenophontos 1993; Rober-tson et al. 1996), because primary sea-floorgeometries are preserved owing to an absence oflarge-scale thrust faulting. Fundamental insightshave been provided through a series of palaeo-magnetic and structural studies (e.g. Clube 1985;Clube & Robertson 1986; Allerton & Vine 1987,1990, 1991; Allerton 1988; Bonhommet et al.1988; MacLeod et al. 1990; Morris et al. 1990,1998; Hurst et al. 1992) that have highlighted

From: ROBERTSON, A. H. F. & MOUNTRAKIS, D. (eds) 2006. Tectonic Development of the Eastern MediterraneanRegion. Geological Society, London, Special Publications, 260, 351–372.0305-8719/06/$15.00 © The Geological Society of London 2006.

352 A. MORRIS ET AL.

Fig. 1. Major ophiolites of the easternmost Mediterranean and Middle East region. The Troodos ophioliteremains in a pre-emplacement setting, in contrast to ophiolites to the east that were tectonically emplaced ontothe Arabian continental margin during the Late Cretaceous.

Fig. 2. Outline tectonic map of the eastern Mediterranean, showing the distribution of Late Cretaceousophiolites (black) and major structural features.

353PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

the role of tectonic rotations in oceanic crustalconstruction and transform tectonism. Theimportance of such rotations is now firmly estab-lished, following their identification in modernspreading systems (e.g. Hurst et al. 1994).

More recently, Morris et al. (2002), Inwoodet al. (2003) and Inwood (2005) have providedthe first palaeomagnetic data from the Baër–Bassit and Hatay ophiolites, which, in contrastto the Troodos ophiolite, were emplaced tecto-nically onto the Arabian continental marginas a series of thrust sheets. These investigationshave discovered ubiquitous large tectonicrotations, only part of which may be attributed topost-emplacement deformation.

We show that the combined palaeomagneticdatabase from all three ophiolites is now suf-ficiently extensive to provide insights into therange of styles of tectonic rotations that haveaffected the Neotethyan oceanic crust duringformation and later deformation, and to discussthe regional-scale significance of these rotations.

Geological summary of the ophiolites

The Troodos ophiolite

The Troodos Complex ophiolite formed duringthe Late Cretaceous (Cenomanian–Turonian,U–Pb age 90–92 Ma; Mukasa & Ludden 1987)in a suprasubduction-zone (SSZ) setting (Pearce1975, 1980). It consists of a complete Penrosepseudostratigraphy disposed in a domal structureas a result of Late Pliocene–Recent uplift, givingrise to a broadly concentric outcrop pattern.Mantle and lower crustal (gabbroic) sequencesare exposed around the central structural andtopographic high. The plutonic section includeslayered cumulates cut by gabbroic intrusions,providing clear evidence for the presence ofsmall, multiple magma chambers beneath theTroodos spreading axis (Robinson & Malpas1990). The contact between the isotropic gabbrosand overlying sheeted dykes is commonly alow-angle extensional detachment fault zone(Varga & Moores 1985), providing evidence foramagmatic stretching during crustal formation.The sheeted dyke complex is exposed over an80 km wide swath and consists of generallynorth–south-striking dykes (present coordinates)that at some localities (e.g. Lemithou) are rotatedto low angles and occasionally cut by later dykes(Dietrich & Spencer 1993). Spreading took placeeither by steady-state processes (Allerton & Vine1987), or by formation of discrete, ephemeral,sea-floor grabens (Varga & Moores 1985). Thebest documented and most distinctive grabenruns through the Solea area to the north of

the plutonic complex, above the ‘KakopetriaDetachment’ fault (Verosub & Moores 1981).This graben is interpreted as a fossil spreadingaxis (the ‘Solea axis’; see Fig. 4). Magnetic fabricand dyke surface lineation data (Staudigel et al.1992; Varga et al. 1998) reveal both upward andlateral magma emplacement through the dykecomplex, supporting a model of magma trans-port along the length of the spreading axis awayfrom isolated magmatic centres beneath axialvolcanoes spaced along the ridge crest. The over-lying extrusive sequence is best exposed along thenorthern and SW margins of the ophiolite, provi-ding classic sections where the interplay betweenmagmatic, tectonic and hydrothermal processesduring construction of the oceanic crust may beestablished (Schmincke et al. 1983; Schmincke &Rautenschlein 1987).

The southern margin of the main outcrop ofthe Troodos ophiolite is marked by the east–west-trending Arakapas Fault Belt, a strike-slipfault system that is interpreted as a fossil oceanictransform fault (Moores & Vine 1971; Simonian& Gass 1978). An anomalous ophiolitic sequenceis exposed to the south of this structure within theLimassol Forest Complex. Here, mantle sequ-ence and lower crustal rocks are exposed athigh structural levels and are cut by numerouseast–west-trending shear zones (Murton 1986;MacLeod 1990). The majority of the LimassolForest Complex is interpreted to have formedwithin a leaky (transtensional) ‘South TroodosTransform Fault Zone’ (STTFZ) whose principaldisplacement zone is represented by the Ara-kapas Fault Belt (MacLeod & Murton 1993). Aprogressive change in orientation of dykes withinthe Sheeted Dyke Complex is observed as theSTTFZ is approached from the north, suggestiveof either primary variation in the orientation ofthe Late Cretaceous stress field adjacent to a sin-istrally slipping transform or post-emplacementclockwise tectonic rotations of dykes resultingfrom dextral shear along the transform. Palaeo-magnetic data (described herein) strongly sup-port this latter model. A small area of normalTroodos-type crust exposed in the SE corner ofthe Limassol Forest Complex is believed to repre-sent a fragment of crust formed at an ‘Anti-Troodos’ spreading axis located to the south ofthe transform domain (MacLeod 1988, 1990).Together these various structures and ophioliticregions provide a unique exposure of the LateCretaceous Neotethyan spreading system that iswell preserved because of a lack of large-scalethrust tectonics during emplacement of theTroodos Complex.

In the Late Cretaceous, shortly after genesisat the Neotethyan spreading axis, the Troodos

354 A. MORRIS ET AL.

oceanic crust became tectonically juxtaposedalong its SW margin (present coordinates) withan allochthonous, highly deformed sequence ofUpper Triassic to middle Cretaceous deep-seasedimentary and Upper Triassic volcanic rocksknown as the Mamonia Complex. This is inter-preted to represent remnants of a passive con-tinental margin and marginal oceanic crust(Robertson & Woodcock 1979; Swarbrick 1979,1980, 1993). The mode of juxtaposition of theTroodos and Mamonia Complex terranes isunder debate, with models invoking eitherstrike-slip-dominated (e.g. Clube & Robertson1986; Swarbrick 1993) or thrust-dominated (e.g.Malpas et al. 1993) emplacement, or a complexmultiphase history of deformation (involvingsuccessive strike-slip, transtensional and con-tractional events; Bailey et al. 2000). Within thesuture zone, high crustal level rocks (extrusiverocks and cross-cutting dykes) are exposed infault-bounded slivers, with faulted contactsmarked by discontinuous, steeply dipping strandsof serpentinite. These slivers are overlain by anin situ sedimentary cover of Campanian umbersand radiolarites (Perapedhi Formation), and bythick, largely undeformed, successions of bento-nitic clays and volcaniclastic sandstones, indistin-guishable from the sedimentary cover foundoverlying the ophiolitic basement of the mainTroodos Complex (Robertson 1977; Clube &Robertson 1986). The highly depleted geoche-mistry of the extrusive rocks (Murton 1990) isdistinctly different from the alkaline within-platebasalt (WPB) to tholeiitic mid-ocean ridge basalt(MORB) compositions of the Triassic extrusiverocks of the Mamonia Complex (Malpas et al.1993). Stratigraphic and petrographic data,therefore, support correlation of these ophioliticoutcrops with the Troodos Complex and itstransform fault-related southern margin. Thisinterpretation is also supported by palaeoma-gnetic data obtained from these slivers by Morriset al. (1998), as discussed below.

The Hatay (KIzIl Dah) and Baër–Bassitophiolites

In contrast to the Troodos Complex, coevalophiolites to the east have been structurallymodified during Late Cretaceous thrust-dominated emplacement onto the Arabian conti-nental margin during progressive closure of thesouthern Neotethyan basin. The most westerlyof these emplaced units are the Baër–Bassit andHatay ophiolites. These are closely relatedspatially (Fig. 2) and represent parts of a singleunit emplaced during the Maastrichtian. TheHatay ophiolite to the north forms a relatively

intact sheet (Delaloye et al. 1980), whereas theBaër–Bassit ophiolite to the south (Kazmin& Kulakov 1968; Parrot 1980) represents theleading edge of the emplaced sheet and is highlydeformed by thrust faulting (Al-Riyami et al.2000). The radiometric age of these ophiolites ispoorly constrained, with ophiolitic dykes yieldingK–Ar ages in the range 73–99 Ma (Delaloye &Wagner 1984).

The Hatay ophiolite is split into a largesouthwestern massif and a smaller northeasternmassif by a high-angle fault (Tahtaköpru Fault).The ophiolite is separated from the underlyingArabian platform by only a thin mélange andno métamorphic sole is preserved (Robertson2002). The succession in the main KIzIl Dagmassif (Delaloye & Wagner 1984) begins withserpentinized harzburgite tectonite with localintercalations of dunite, wehrlites, lherzolite andfeldspathic peridotites. The ultramafic rocks areseparated from the overlying gabbros by a 50–100 m thick shear zone that extends upwardsinto the base of the layered gabbros (Dilek &Thy 1998). The layered gabbros, in turn, passinto isotropic gabbros, intruded by small bodiesof plagiogranites, leucocratic gabbro and doler-ite. Diabase dykes increase in abundance towardsthe top of the gabbros, which pass upwards intoa sheeted dyke complex. Locally, the gabbro–dyke contact is a low-angle shear zone marked byhydrothermal alteration. In the NE, sheeteddykes are unconformably overlain by Maastrich-tian non-marine to shallow-marine sediments,presumably after erosion of ophiolitic extrusiverocks (Erendil 1984; Pi¦kin et al. 1986).

The succession in the northeasterly massif(NE of the Tahtaköpru Fault) is exposed atseveral localities. At one, serpentinized peridot-ites are tectonically overlain by gabbro, rotateddykes and lavas (Dilek & Thy 1998). Elsewhere,serpentinized peridotites are overlain, above agently dipping normal fault, by massive andpillow lavas that are rarely interbedded with oroverlain by metalliferous sediments (Erendil1984; Robertson 1986). Pillow flows are steeplydipping to subvertical. Further south (southof Antakya), gabbros or serpentinites are inlow-angle faulted contact with overlying pillowlavas. These extrusive rocks include highlymagnesian, boninite-type lavas (‘sakalavite’;Delaloye & Wagner 1984).

The highly dismembered Baër–Bassit ophio-lite is underlain by a well-developed invertedamphibolite or greenschist-facies metamorphicsole (Whitechurch 1977; Al-Riyami et al. 2002),that has a K–Ar age of 86–93 Ma (Thuizatet al. 1981; Delaloye & Wagner 1984). Theophiolitic outcrop is dominated by two massifs,

355PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

Baër in the NE (inland) and Bassit in the NW(near the coast), together with smaller masses ofhighly dismembered ophiolitic rocks furtherSE. The Baër massif is subdivided into severallarge thrust sheets, dominated by harzburgitesoverlain by cumulate ultramafic rocks (Al-Riyami et al. 2000). Layered gabbros (<1 kmthick) are locally cut by dolerite dykes. The Bassitmassif comprises a lower sequence of harzbur-gites and gabbros, which are overthrust by a sliceof mélange and then by thin (<100 m thick)imbricate thrust sheets of gabbro, sheeted dykesand pillow lavas. Geochemical analysis of extru-sive rocks reveals strongly depleted, highlymagnesian boninite types (Al-Riyami et al. 2000).The ophiolite and its metamorphic sole areunderlain by the extensive ‘Baër–Bassit Mélange’(Al-Riyami et al. 2000; Al-Riyami & Robertson2002), which consists of a Late Triassic to mid-Cretaceous deep-water, passive margin succes-sion (Delaune-Mayère 1984; Al-Riyami et al.2000). All three units (Baër–Bassit ophiolite,metamorphic sole and Baër–Bassit Mélange)were thrust onto the Arabian carbonate platformin the mid-Maastrichtian (c. 70 Ma), based onthe biostratigraphic ages of the youngest sequen-ces of the Arabian platform beneath the alloch-thon and the oldest sedimentary rocks of thepost-emplacement cover sequences (Kazmin &Kulakov 1968). Lineations in the metamorphicsole, defined by elongation of amphibole porphy-roblasts, together with fold facing and vergencedirections within the underlying Baër–BassitMélange, indicate that thrust sheets wereemplaced towards the SE (Al-Riyami et al. 2002;Al-Riyami & Robertson 2002).

The disrupted Baër–Bassit allochthon waslater unconformably overlain by a sedimentarysequence of late Maastrichtian to Pliocene age(Boulton et al. 2006). The sedimentary succes-sions are cut by mainly ENE–WSW-trendingstrike-slip faults that extend offshore (Figs 2 and7). This fault system represents part of the exten-sion of the plate boundary zone between theAfrican plate and the Turkish microplate, whichruns eastwards from south of Cyprus as a zone ofdistributed deformation and then comes onshore,passing through the Baër–Bassit region to linkwith the Dead Sea transform fault system to theeast (Al-Riyami et al. 2002; Fig. 2).

The palaeomagnetic database

Sources, data selection and reporting

Discussion is restricted to a synopsis of the majorpalaeomagnetic results obtained to date, and

no attempt is made to summarize the growingliterature on magnetic fabric (anisotropy of mag-netic susceptibility) results from the Troodosophiolite (e.g. Staudigel et al. 1992; Varga et al.1998; Abelson et al. 2001; Borradaile & Lagroix2001). The majority of ophiolite data used in thissynopsis were collated from published sourcesby Morris (2003) for the purpose of assessingthe palaeolatitude of the Neotethyan spreadingaxis. This compilation consisted of results from:(1) 100 palaeomagnetic sampling sites in theTroodos ophiolite, drawn from Clube (1985),Clube et al. (1985) Bonhommet et al. (1988),Allerton (1989a), Morris et al. (1990, 1998) andHurst et al. (1992); (2) 19 sites in the Baër–Bassitophiolite presented by Morris et al. (2002). Addi-tional data from six sites in the Mandria areaof the Troodos ophiolite from MacLeod et al.(1990) are included here. New palaeomagneticdata from 18 sites of the Hatay ophiolite areincluded from Inwood et al. (2003) and Inwood(2005). Finally, to assess the timing of post-crustal genesis rotations in the ophiolites, dataare also drawn from: (1) 26 sites distributedthrough the in situ Upper Cretaceous to Miocenesedimentary cover sequences of the Troodosophiolite, reported by Clube (1985), Abrahamsen& Schönharting (1987) and Morris et al.(1990); (2) 15 sites within the Tertiary post-emplacement sedimentary cover of the Baër–Bassit ophiolite, reported by Morris et al. (2006);(3) 17 sites within the post-emplacement sedimen-tary sequences overlying the Hatay ophiolite,reported by Inwood et al. (2003), Kissel et al.(2003) and Inwood (2005).

The following criteria were used in the selec-tion of data from the source publications: (1)site-level data are based on laboratory cleanedsample remanences with the stability of rema-nences investigated at each site by using eitherdemagnetization of pilot samples or, preferably,full demagnetization of each sample; (2) struc-tural corrections are reported in the sourcepaper or may be recovered from reported in situand tilt-corrected remanence data; (3) data fulfilthe following statistical constraints: number ofsamples included in the site mean, ni5; cone ofconfidence, a95<20°; and Fisherian precisionparameter, k>10.

The reader is referred to the source papers fordetails of sampling procedures and site locations.Stereographic projections of site-mean rema-nence data and the associated cones of confidenceare included therein. Primary tables of data maybe found in the source papers, or have been givenby Morris (2003) in the case of data from theTroodos and Baër–Bassit ophiolitic sites.

356 A. MORRIS ET AL.

Data correction and interpretationtechniques

Standard palaeomagnetic practice involves struc-turally correcting in situ (geographical coordi-nates) remanence data by applying a simpletilt around a strike-parallel axis to restore apalaeohorizontal or palaeovertical surface to thepresent-day horizontal or vertical. Tilt-correctedvectors may then be compared with an appro-priate coeval reference vector, with differencesin declination (azimuth) being interpreted interms of vertical axis rotations, and differences ininclination (dip) being attributed to either palaeo-latitudinal movements or to inclination shal-lowing as a result of compaction (in the case ofsedimentary rocks). This tilt correction approachdecomposes the total deformation at a site intocomponents of rotation around horizontal andvertical axes. Declination errors may be intro-duced artificially if deformation involved tiltingaround inclined axes, if more than one phase oftilting has occurred, or if fold axes are plunging(MacDonald 1980). In the last case, however,declination errors are<10° for fold plunges ofup to 50° if the palaeohorizontal dips at<40°,and are<10° even for vertical beds if the foldplunge is<10°. The most severe source of error inadopting standard tilt corrections in ophioliticterranes, however, arises from restoration ofsheeted dykes to the vertical, as components oftilt around dyke-normal axes are impossible toresolve in the absence of palaeohorizontal mark-ers (Borradaile 2001; Morris & Anderson 2002).Such unresolved tilts may potentially introduceboth declination and inclination anomalies.

An alternative net tectonic rotation approachthat has been widely adopted in these ophiolites(Allerton & Vine 1987; Allerton 1989a; Morriset al. 1990, 1998, 2002; Hurst et al. 1992) is todescribe the deformation at a site in terms of asingle rotation about an inclined axis, whichrestores both the palaeohorizontal and palaeo-vertical to their initial orientation and the sitemean magnetization vector to the appropriatepalaeomagnetic reference direction. This singlerotation may then be decomposed into anynumber of component rotations on the basis ofadditional structural data, and/or net tectonicrotation axes may be interpreted directly in termsof structural history. The key assumptions inthis method (Allerton & Vine 1987) are that: (1)pre-deformational remanences are preserved; (2)an appropriate (coeval) reference magnetizationdirection may be found; (3) dyke margins orpalaeohorizontal surfaces should be restoredto as close to vertical or horizontal as possible;

(4) no internal deformation of sampled units hasoccurred.

Useful insights into the tectonic evolution ofthe ophiolites under consideration here have beenobtained using either one or both of these inter-pretation methods. For the purposes of thissynopsis, however, data are displayed followingsimple tilt correction only, for the followingreasons: (1) this unifies the palaeomagnetic data-base; (2) primary, detailed interpretations areavailable in the source papers, where issuesrelated to choice of methods are discussed indetail; (3) the tectonic significance of tilt cor-rected data is easier to understand intuitively; (4)application of the net tectonic rotation approachrequires use of standard or inclination-only tilttests based on the simple tilt correction approachin order to properly assess the timing of magne-tization acquisition relative to deformation,unless magnetizations are merely assumed to bepre-deformational in age; (5) the tectonic signifi-cance of the data in terms of rotation patternsand styles is largely unaffected by the choice oftechnique. Reference is made to the results of nettectonic rotation analyses where appropriate.

Age of magnetization

Detailed discussion of the palaeomagnetic char-acteristics of the ophiolites is beyond the scopeof this paper. In summary, demagnetizationof natural remanent magnetizations generallyreveals characteristic magnetizations carried by arange of ferrimagnetic phases that can be readilyrelated to assemblages observed within in situoceanic crust and other ophiolites (Dunlop &Özdemir 1997). Of prime concern, however, isconsideration of the age of magnetization withinthe sampled ophiolitic lithologies, as this iscritical to the tectonic interpretation of thepalaeomagnetic data. This is determined withrespect to tectonic disruption of sampled unitsusing field tests of palaeomagnetic stability, themost common of which is the palaeomagneticfold or tilt test (McElhinny 1964; McFadden &Jones 1981). Differential vertical axis rotations,however, invalidate use of area-wide tilt testsbased on full remanence vectors (declination andinclination). An alternative approach that is notaffected by such rotations is to determine theeffect of untilting on the distribution of inclina-tions only. The angle between the inclination andthe palaeohorizontal at a site may be assumedto remain constant during rigid body rotation,regardless of the axis of rotation. Significantimprovement in clustering of inclinations upontilt correction of mean directions from sites

357PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

with different structural orientations thereforesuggests that a pre-tilt magnetization has beenidentified (Enkin & Watson 1996). Figure 3shows the variation of the maximum likelihoodestimates of the Fisher precision parameter,k, with degree of untilting for each of thethree ophiolites. Strongly peaked distributionswith maximum k values at 100% untilting dem-onstrate unequivocally that pre-deformationalremanences are identified within each terrane.The extensive database available for the Troodosophiolite allows separate consideration of theage of magnetization of the extrusive sequencesand sheeted dyke complex. Positive tilt tests areobserved in both cases. Close agreement betweenthe tilt-corrected mean inclinations of both unitssuggest that the sheeted dyke dataset is suffi-ciently large to ensure that any components oftilting around dyke-normal axes (Borradaile2001; Morris & Anderson 2002) at individualsites produces little bias in the overall meaninclination for this unit.

Tectonic significance of the data

On the basis of the tilt test results describedabove, palaeomagnetic data from the ophiolitesare hereafter described and interpreted in tilt-corrected (stratigraphic) coordinates. Stereogra-phic projections of site-mean remanence data aregiven in Figure 4 (Troodos) and Figure 7 (Hatayand Baër–Bassit). Data from the sedimentary

cover sequences are presented in Figure 6(Troodos) and Figure 7 (Hatay and Baër–Bassit).Results are discussed by individual ophiolitebelow, prior to regional synthesis.

Troodos ophiolite

There is a clear distinction within the Troodospalaeomagnetic database (Fig. 4) betweenregions with magnetization vectors that clusteraround westerly declinations and regions wheredeclinations are widely variable from WSW,through northerly to easterly declinations. Theformer regions (stereonets 1–4 with unshaded a95

ellipses in Fig. 4) are the northern margin of themain ophiolite (Clube 1985), the Solea region(Allerton & Vine 1987; Allerton 1989a; Hurstet al. 1992), the Akamas peninsula (Clube 1985;Morris et al. 1998) and the western margin of theophiolite (Morris et al. 1998). These regions areremote from the STTFZ and its inferred westerlyalong-strike extension, and provide the evidencefor the ophiolite-wide bulk anticlockwise rotationof the Troodos microplate first identified fromnatural remanent magnetization data by Vine &Moores (1969) and Moores & Vine (1971). Themean direction of magnetization of these 29 sitesis declination (D)=272.5°, inclination (I )=38.4°,a95=6.5°, k=17.7. This westerly-directed magne-tization is also observed at sites (Morris et al.1990) within the SE part of the Eastern LimassolForest Complex to the immediate south of theSTTFZ (stereonet 5, Fig. 4), providing palaeo-magnetic support for the presence of a fragment

Fig. 3. Variation in the Fisher precision parameter with progressive untilting of palaeomagnetic data from siteswithin the Troodos, Hatay and Baër–Bassit ophiolites. These data indicate positive inclination-only tilt tests(Enkin & Watson 1996) in all three cases, with peaked distributions centred on 100% untilting. Thisdemonstrates unequivocally that pre-deformational magnetizations are recorded by the sampled units, and thatpalaeomagnetic data should therefore be interpreted in tilt-corrected coordinates.

358 A. MORRIS ET AL.

of an Anti-Troodos plate in this area, as origi-nally deduced by MacLeod (1988, 1990) fromfield geological observations.

Palaeomagnetic data from the Solea area pro-vide important insights into tectonic processes atslow- to intermediate-spreading axes. The Soleagraben is a 15–20 km wide asymmetrical struc-ture defined principally by variations in dykeattitude in the Sheeted Dyke Complex (Varga &Moores 1985), which in this area is separatedfrom the underlying plutonic complex by thelow-angle Kakopetria detachment fault zone. Anoceanic environment for formation of the grabenis demonstrated by the horizontal disposition ofthe overlying sedimentary sequences (MacLeodet al. 1990). To the east of the inferred spreadingaxis, dykes are generally steeply dipping tosubvertical and trend NNW except for several

small areas where dykes dip more gently (Hurstet al. 1992). The wider western flank is distinctlydifferent, with north–south-trending dykes dip-ping at low angles of 25–45° to the east. The low-angle orientations of these dykes were attributedby Verosub & Moores (1981) to listric normalfaulting associated with the Kakopetria detach-ment, above a magma chamber at the activespreading axis. In this model, plate separationwas at least partly accommodated by tectonicthinning of the upper crust during periods ofamagmatic stretching. Palaeomagnetic analysisof these dykes (Allerton & Vine 1987; Allerton1989a; Hurst et al. 1992) confirms that theywere originally intruded in (sub)vertical orienta-tions. Tilt-corrected vectors (stereographic pro-jection 3, Fig. 4) generally cluster around themean inclination observed at other localities far

Fig. 4. Outline geological map of Cyprus, showing the location of major structural features and the locations ofareas that have been investigated palaeomagnetically. Lower hemisphere stereographic projections showtilt-corrected site-level palaeomagnetic directions and associated a95 cones of confidence from each area. Shadeda95 ellipses indicate data obtained from ophiolitic crust that is inferred to have experienced significant transformtectonism, whereas unshaded a95 ellipses indicate data obtained from localities outside the transform tectonizedzone. Black star in stereographic projection 1 indicates the direction of the present-day geomagnetic field.STTFZ, South Troodos Transform Fault Zone; AFB, Arakapas Fault Belt. Data sources: Clube (1985);Bonhommet et al. (1988); Allerton (1989a); Morris et al. (1990, 1998); Hurst et al. (1992).

359PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

from the STTFZ, in contrast to in situ incli-nations (not shown) that vary widely from −25°to 75°. The data are most informatively analysedusing the net tectonic rotation approach(Allerton & Vine 1987). This reveals rotationaxes for dykes in the western portion that aresubhorizontal and subparallel to the originaldyke strikes (Allerton & Vine 1987; Allerton1989a; Hurst et al. 1992), consistent with rotation(tilting) of dykes by faults above the detachmentsurface. For the eastern Solea graben, markedby steeper and more variable dyke orientations,structural evidence does not support the existenceof a throughgoing detachment fault. Net tectonicrotation analysis here indicates that some dykesappear to have been tilted towards the Soleaaxis, but the majority show minor to significantrotation about vertical axes. This was attributedby Hurst et al. (1992) to local variations inthe amount of extension, related to late-stageepisodic intrusions.

Extensive palaeomagnetic data (Clube 1985;Bonhommet et al. 1988; Allerton 1989a;MacLeod et al. 1990; Morris et al. 1990) from thesouthern half of the main Troodos ophiolite andthe Limassol Forest Complex have been used toaddress the debate on the sense of displacementalong the STTFZ and its relationship with theSolea axis. A progressive change in dyke trend isobserved within the Sheeted Dyke Complex asthe transform zone is approached from the north,from a predominant north–south orientation intoeventual alignment with the transform lineament(Fig. 5a). This has been interpreted as the resultof either dyke injection into a sigmoidal stressfield, implying that dykes are in their originalorientations relative to the sinistrally slippingtransform (Fig. 5b), or clockwise vertical-axisfault block rotations in response to dextralslip (Simonian & Gass 1978; Fig. 5c). Discrimi-nation between these alternative models can beachieved palaeomagnetically, as systematic tec-tonic rotations would result in systematic varia-tions in magnetization vectors away from thewesterly directed vectors observed outside thetransform-influenced zone. Data from the regionto the north and NE of the Arakapas FaultBelt (Bonhommet et al. 1988; Allerton 1989a;stereonet 6, Fig. 4) show a broad spread ofdirections that are clearly rotated in a clockwisesense relative to this westerly vector, supportinga dextral shear sense along the transformbetween sinistrally offset spreading axes. Themost unequivocal analysis was presented byBonhommet et al. (1988), who showed that mag-netizations in sampled dykes cluster tightly with

westerly declinations after correction of dykesback to the predominant north–south trend,thereby ruling out initial NE–SW trends. Furtherpalaeomagnetic support for dextral shear alongthe transform zone is provided by data from theSSTFZ itself (Morris et al. 1990; stereonet 7,Fig. 4), the majority of which are also rotatedclockwise away from the general Troodos vector.Net tectonic rotation analysis of data from setsof cross-cutting dykes (Allerton & Vine 1990;Morris et al. 1990) demonstrates that clockwiseblock rotations were actively occurring duringcrustal genesis, rather than resulting from laterreactivation of the fault zone.

Analysis of upper crustal rocks of Troodos-type exposed in fault-bounded slivers along thesuture zone with the Mamonia Complex in SWCyprus (stereonet 10, Fig 4) reveals significantrotations in a generally clockwise sense awayfrom the Troodos vector (Morris et al. 1998). Inparticular, differences in remanence directionsbetween cross-cutting units are observed atseveral localities, interpreted elsewhere as a char-acteristic of synmagmatic rotation during trans-form tectonism. Net tectonic rotation analysesallow decomposition of the total rotation at thesesites into early and late components. Early rota-tions are consistently clockwise, in agreementwith studies of rotations associated with theSTTFZ further to the east. These data, therefore,suggest that transform-tectonized crust is preser-ved in SW Cyprus, an interpretation supportedby stratigraphic and geochemical similaritiesbetween these units and the main Troodosophiolite and its transform fault-related southernmargin.

The overwhelming palaeomagnetic evidencefor clockwise fault block rotations associatedwith dextral slip along the Southern Troodostransform contrasts with sinistral kinematic datareported by Murton (1986) within the westernLimassol Forest Complex (i.e. within the trans-form zone). This apparent paradox was resolvedby MacLeod & Murton (1995), who proposed amodel in which sinistral shear developed locallyat block boundaries within an overall dextralshear zone.

Detailed palaeomagnetic and structuralanalyses (MacLeod et al. 1990) have identified alimit to the zone of transform-related rotationsalong the STTFZ (within area 8, Fig. 4; Fig. 5d).The changeover from rotated to unrotated dykesoccurs across a complex zone about 2–5 kmwide to the west of the village of Mandria.This zone was interpreted by MacLeod et al.(1990) as a fossilized intersection between

360 A. MORRIS ET AL.

the Solea axis and the SSTFZ. They notedthat the radius of curvature of dyke swingremains approximately constant across the entireexposed width of the Troodos massif to theeast of the ridge–transform intersection (seeFig. 5a). MacLeod et al. (1990) concluded thattransform-related rotations occurred within theactive inside corner of the intersection itselfrather than being accommodated progressivelywith increasing strike-slip displacement along the

transform. This supports a theoretical model(Allerton 1989b) for distortions within weakcrust at ridge–transform intersections. Distrib-uted rotational deformation is, therefore, consid-ered to be largely confined to the inside corner ofthe intersection itself (MacLeod et al. 1990), andsubsequent strain is taken up by strike-slipfaulting concentrated almost exclusively in theprincipal transform displacement zone within thetransform valley (Arakapas Fault Belt).

Fig. 5. (a) Simplified geological map of the southern margin of the main Troodos ophiolite and the LimassolForest Complex (after Simonian & Gass 1978), showing the progressive change in dyke trend into nearparallelism to the Southern Troodos transform zone over a distance of 10–15 km (from Morris et al. 1990).(b) and (c) show possible alternative settings in which deviations in dyke trend could occur close to the SouthernTroodos Transform Fault Zone (from Morris et al. 1990): (b) dyke injection into a sigmoidal stress fieldoperating across a sinistrally slipping transform between dextrally offset spreading axes; (c) rotation of faultblocks related to dextral slip along the active transform domain between sinistrally offset spreading axes.(d) Geological map of the Mandria area (modified from MacLeod et al. 1990), with palaeomagnetic sites shown(in situ declinations indicated by arrows). Dykes in the west have an average north–south trend in contrast tohighly rotated dykes to the east. The boundary between these domains represents a fossil ridge–transformintersection.

361PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

The debate over the tectonic interpretationof the STTFZ has recently been reopened byCann et al. (2001). They noted the presence inthe Limassol Forest area of outcrops of deepermantle or crustal lithologies in low-angle exten-sional tectonic contact with a range of shallowercrustal lithologies, as mapped by B. J. Murtonand C. J. MacLeod (in Gass et al. 1991, 1994).In the eastern Limassol Forest Complex theseextensional structures were attributed byMacLeod (1988, 1990) to a sustained period ofpost-volcanic extensional reactivation of theSTTFZ in the Late Cretaceous, related to theinitiation of palaeorotation of the ‘Troodosmicroplate’. By contrast, Cann et al. (2001)compared these outcrop patterns and structureswith the characteristics of extensional oceaniccore complexes developed at inside corners ofridge–transform intersections in modern slow-spreading systems (e.g. those associated withthe Atlantis Transform Fault, Mid-AtlanticRidge at 30°N; Blackman et al. 1998). In thisalternative interpretation (Fig. 6a), the LimassolForest Complex represents a remnant of a stripof inside-corner crust formed to the present-daysouth of a sinistrally slipping active transform,with the zone of curved dyke trajectories(described above) lying in crust formed at theoutside corner. This model, however, providesno geologically realistic explanation for the pro-gressive change in dyke strike as the STTFZ isapproached from the north (Fig. 5a), a patternthat has been unequivocally shown to result fromdifferential vertical-axis tectonic rotations. It isalso inconsistent with the dominance of clockwiserotations within the transform zone itself (thatdemonstrably were synchronous with activemagmatism), and finally cannot be reconciledwith the evidence for the presence of a ridge–transform intersection in the Mandria area(Fig. 5d), which indicates that inside-corner crustlies to the present-day east of the Solea axis(Fig. 6b).

The large anticlockwise rotation of the‘Troodos microplate’ is a regionally significantevent within the Neotethys ocean. Initial esti-mates of the size of the rotated unit (Clube 1985;Clube & Robertson 1986) suggested that rotationwas restricted to an oceanic microplate of appro-ximately the same area as the Troodos ophioliteitself, although this now requires re-evaluation inthe light of new data from the emplaced ophio-lites exposed further east (summarized below).The timing of the rotation may be determinedby palaeomagnetic analysis of the continuousupper Cretaceous to Recent in situ sedimentarycover to the ophiolite, on the basis that magneticdeclinations within the sediments record rotation

of the underlying ophiolitic basement. Severalattempts at such analyses have been made, theearliest of which (Shelton & Gass 1980) washindered by difficulties in measuring the veryweak magnetizations of the predominantlycarbonate lithologies and a lack of appropriatedemagnetization experiments. Subsequent inves-tigations by Clube (1985) and Abrahamsen &Schönharting (1987) yielded data distributedthrough the succession, and further constraintson the early rotation history were provided byMorris et al. (1990). Site-level data from thesestudies that meet the quality criteria outlinedabove are shown in Figure 7a. Hydrothermalsediments (umbers) of the Perapedhi Formationshare a common direction with the underlyingextrusive sequence. The overlying Campanianradiolarian mudstones (Perapedhi Formation)and Maastrichtian–Oligocene chalks (LefkaraFormation) show a general progression fromWNW to northerly declinations upwardsthrough the stratigraphy. Data for some timeintervals show significant scatter, most notablyin the inclination of the Maastrichtian sites andin both inclination and declination of the Pale-ocene sites. This may reflect the influence ofcompaction-related inclination shallowing and/or potential contamination of site-level rema-nence vectors by residual normal polarityoverprints. This latter effect is the most likelyexplanation of the inclination difference betweenthe two Maastrichtian sites with the shallowestinclinations (representing inverted reversedpolarity sites) and the remaining three (normalpolarity) sites of this age. Late Oligocene toMiocene sites exhibit exclusively northerly decli-nations (within error). Figure 7b shows the varia-tion of rotation angles through time, derivedfrom a comparison of mean remanence data fortime periods represented by three or more sitemean directions in Figure 7a with expected direc-tions calculated from the African master appar-ent polar wander path of Besse & Courtillot(2002). These data clearly indicate the progres-sive and prolonged nature of the rotation of theTroodos microplate during the Campanian–Eocene interval. However, interpretation interms of variations in rotation rate through timeis made difficult by palaeomagnetic uncertaintiesrelating to limited site numbers and lack of moreprecise dating of sampled units.

Hatay and Baër–Bassit ophiolites

Morris & Anderson (2002), Morris et al. (2002),Inwood et al. (2003) and Inwood (2005) haverecently provided the first palaeomagnetic datafrom the emplaced ophiolites directly to theeast of Troodos. Layered gabbros, and massiveand pillowed lava flows of the Hatay ophiolite

362 A. MORRIS ET AL.

(Fig. 8a) share a common tilt-corrected inclina-tion with declinations strung out along a partialsmall circle through this inclination. The meandirection of D=274°, I=34° is indistinguishablefrom that commonly reported for non-transformtectonized parts of the Troodos ophiolite(D=273°, I=38° (this paper); D=276°, I=32°(Vine & Moores 1969); or D=274°, I=36°(Clube & Robertson 1986)), and indicates amean anticlockwise rotation of 73° when com-pared with an expected direction derived from theapparent polar wander path for Africa (Besse &Courtillot 2002). The distribution of vectors iscomparable with that observed in the Solea area

of Troodos (stereonet 3, Fig. 4). The spread ofdeclinations may result from relatively minorvertical-axis rotations during crustal accretionand/or the cumulative effects of minor tilting orrotation during phases of post-emplacementfaulting (Inwood 2005). Finally, a contributionto the spread of declinations from transform-tectonism cannot be excluded (although cross-cutting units, characteristic of such deformationwithin the STTFZ in Cyprus, were not observedin the field).

Data for the post-emplacement sedimentarycover of the Hatay ophiolite, reported by Kisselet al. (2003) and Inwood (2005), include sites of

Fig. 6. (a) Summary of the Cann et al. (2001) model for the origin of the Limassol Forest Complex. This modelinvokes a sinistrally slipping transform between dextrally offset spreading axes to account for an inferredinside-corner setting for the development of extensional structures observed in the Limassol Forest area(modified from Cann et al. 2001). (b) The geometry of the alternative model that is consistent with field andpalaeomagnetic evidence for clockwise transform-related rotations and their distribution relative to the Soleaspreading axis (as discussed in the text). In this model, extensional structures in the Limassol Forest Complexresult from post-volcanic reactivation of the STTFZ during the early stages of microplate rotation, rather thanthe development of synspreading oceanic core complexes. RTI, ridge–transform intersection; STTFZ, SouthTroodos Transform Fault Zone.

363PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

both normal and reverse polarity (Fig. 8c) thatpass reversal and tilt tests (Inwood 2005) suggest-ing that pre-deformational remanences are pre-served. These data, including those for the oldest(Palaeocene) sequences, indicate only minoranticlockwise rotation (Fig. 8c). This contrastswith the large anticlockwise rotations observed inthe underlying ophiolite, and confirms that most(c. 62°) of the rotation of the ophiolite occurred inan intra-oceanic setting and/or during tectonicemplacement in the Maastrichtian.

Results from the highly dismembered Baër–Bassit ophiolite (Morris et al. 2002), representingthe leading edge of the emplaced ophiolite sheet(Robertson 2002), indicate extreme relativerotations between sampled localities (Fig. 8b),with rotations varying on a kilometre scale. Theeastern, Baër massif is dominated by mantlesequence rocks and data are available from onlyone locality, in the sheeted dyke complex. This isdominated by tilting with only moderate compo-nents of anticlockwise rotation about a verticalaxis (NW directions, Fig. 8b). Within the Bassitmassif, anticlockwise net tectonic rotationsincrease from c. 90° in the north (westerly direc-tions, Fig 8b) to in excess of 200° in the south (SE

directions, Fig. 8b), and rotation axes determinedusing the net tectonic rotation approach aresteeply plunging to subvertical. The possibilitythat the largest rotations occurred in a clockwisesense cannot be excluded, but anticlockwisesolutions were preferred by Morris et al. (2002),as this results in a systematic pattern of rotationsacross the Bassit massif. Data from another eightsites at a fifth locality (in the Bassit massif) wheresubvertical remanence directions resulted fromrotation around shallow- to moderately plungingdyke-normal axes are not shown in Figure 8b butwere discussed in detail by Morris & Anderson(2002).

The increase in rotation angles southwardsthrough the Bassit sheet is most readily explainedby increasing proximity to a major strike-slipfault zone that traverses the southern half of theophiolite, representing the expression of thepresent-day plate boundary system in this region(Al-Riyami et al. 2000). Palaeomagnetic data forthe Palaeogene post-emplacement sedimentarycover sequences within and adjacent to this faultzone (Morris et al. 2006) reveal widely differentrotation angles and senses in different faultblocks (Fig. 8d), confirming that significant

Fig. 7. (a) Lower hemisphere stereographic projection of tilt-corrected site-level palaeomagnetic data obtainedfrom the in situ sedimentary cover of the Troodos ophiolite. Data from Campanian radiolarites andMaastrichtian and Paleocene carbonates include both normal and reversely magnetized sites, with the latterinverted to the lower hemisphere in this plot. Eocene to Miocene data are all of normal polarity. Data sources:Clube (1985); Abrahamsen & Schönharting (1987); Morris et al. (1990). (b) Variation of rotation angle throughtime derived by comparing mean data for stratigraphic intervals with three or more palaeomagnetic site meandirections with expected directions calculated from the African apparent polar wander path of Besse &Courtillot (2002). These data clearly demonstrate the progressive anticlockwise rotation of the underlyingophiolite during the Late Cretaceous and Palaeogene.

364 A. MORRIS ET AL.

rotational deformation accompanied the devel-opment of these neotectonic structures. Theweakly magnetized Paleocene carbonates aregently folded, and an inclination-only tilt testsuggests that remanences were acquired duringfolding. Data from Neogene sequences (Fig. 8e)exposed to the NE are inherently more difficultto interpret, as the distinction between northerlydirected viscous overprints and unrotatedNeogene magnetizations is impossible in thesesubhorizontal sequences.

Overall, the data obtained to date from theHatay and Baër–Bassit ophiolites and their post-emplacement sedimentary cover sequences aremost consistent with large-scale, bulk rotationof the ophiolitic sheet, either in an intraoceanicsetting or during emplacement. The northern,Hatay, part of the emplaced sheet experienced

only moderate post-emplacement tectonic rota-tion (Inwood et al. 2003; Kissel et al. 2003;Inwood 2005), in contrast to the more highlydeformed southern leading edge in Baër–Bassit,where subsequent amplification of the amountof rotation resulted from neotectonic modifica-tion during development of the modern plateboundary configuration.

Discussion of regional implications

Palaeogeographical implications

Palaeolatitudes of c. 21–24°N derived frominclination-only statistical analysis of the ophio-lite palaeomagnetic database are consistent witha Late Cretaceous position for the Neotethyan

Fig. 8. Outline geological map of the Hatay (KIzIl Dag) and Baër–Bassit ophiolites, showing the distribution oflithologies and major structures. Lower hemisphere stereographic projections of tilt-corrected site-levelpalaeomagnetic data from: (a) and (b) the Hatay and Baër–Bassit ophiolites, respectively (data from Inwoodet al. (2003), Inwood (2005) and Morris et al. (2003)); in (b), reversed polarity data (14 out of 19 sites) have beeninverted to allow ready comparison with (a); (c) the post-emplacement sedimentary cover of the Hatay ophiolite(data from Inwood et al. (2003), Kissel et al. (2003) and Inwood (2005)); (d) and (e) the Palaeogene andNeogene post-emplacement sedimentary cover of the Baër–Bassit ophiolite (data from Morris et al. (2006)). Aninclination-only tilt test performed on data from the Palaeogene sequences indicates maximum clustering ofinclinations at 50% untilting. Black star in stereographic projection (a) indicates the direction of the present-daygeomagnetic field.

365PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

spreading axis between the Arabian and Eurasianmargins (Morris 2003). These data, together witha well-defined palaeolatitude of c. 26°N for theeastern Pontides (Channell et al. 1996) andpalaeolatitudes of the Arabian and Eurasianmargins derived from appropriate apparent polarwander paths (e.g. Besse & Courtillot 1991,2002), provide constraints on potential tectonicreconstructions of the eastern MediterraneanTethys. Two model reconstructions that involvegenesis of the ophiolites in a southerly basin (e.g.Robertson & Dixon 1984; Robertson et al. 1991;Robertson 1998) and that satisfy the palaeo-latitudinal data are shown in Figure 9, using thedata from the Baër–Bassit ophiolite as an exam-ple. The solution giving the widest southerlyNeotethys is obtained by placing the Arabiancontinental margin at its southernmost limit.This produces a 1000 km wide Neotethyanstrand to the south of the subduction zone associ-ated with the Baër–Bassit spreading axis. Majorarc magmatism would result from subduction ofthis lithosphere during continued plate conver-gence. Arc-related products are sparse, althoughin SE Turkey the Ba¦kil arc (Akta¦ & Robertson1984; Yazgan & Chessex 1991; Rizaoglu et al.2006) 200 km to the NE of Baër–Bassit providesevidence for Andean-type arc magmatism alongthe southern Tauride margin during the lateCretaceous (Robertson 2002). The problem of

the scarcity of arc products in the wider region ispotentially avoided in the ‘narrowest solution’(and near intermediate solutions), which impliesan oceanic tract c. 300 km wide to the south ofthe subduction zone (Fig. 9). In both solutions,the northern Neotethyan strand was essentiallyconsumed by subduction beneath the Pontides.This is consistent with geological evidencesuggesting that northward subduction beneaththe Pontides was active from the Late Mesozoiconwards (Ustaömer & Robertson 1993, 1994),providing sufficient time for the northernNeotethys to have been subducted by the LateCretaceous. The ‘narrowest solution’ also impliesa restricted width of the southern Neotethyanstrand following ophiolite emplacement (Fig. 9).In this respect, intermediate solutions that placethe Eurasian margin further to the north (therebypermitting a wider oceanic tract between theTaurides and the Neotethyan spreading axis)are more consistent with geological evidence indi-cating that a southern Tethyan basin persistedto the north of the Arabian margin well into theTertiary after partial basin closure associatedwith ophiolite emplacement (Akta¦ & Robertson1984; YIlmaz 1993; Robertson et al. 1996).

Alternative models (e.g. Ricou 1971; Ricouet al. 1979, 1984; Dercourt et al. 1986, 1993)involve formation of the ophiolites in a northerlyNeotethyan basin by spreading at a ‘normal’

Fig. 9. End-member tectonic cross-sections along a north–south transect (from the Arabian margin, through theBaër–Bassit spreading axis and eastern Pontides to the Eurasian margin) that satisfy the palaeomagneticconstraints on the palaeolatitude of the Eurasian and Arabian continental margins, the Baër–Bassit spreadingaxis and the eastern Pontide volcanic arc. Upper and lower sections respectively illustrate the solution thatyields the widest and narrowest south Neotethyan ocean (Eur, Eurasian margin; EP, eastern Pontides; Tau,Taurides; BB, Baër–Bassit segment of the Neotethyan spreading axis; BS, Black Sea; Arab, Arabian margin).

366 A. MORRIS ET AL.

oceanic ridge, with subsequent, Late Cretaceouslarge-scale thrusting (hundreds of kilometres)to the south of emplaced ophiolites over micro-continental fragments to reach their presentpositions. These cannot be discounted on apurely palaeomagnetic basis in the absence ofreliable data from the eastern Taurides (Morris2003). Such models are not supported, however,by a number of key geological observations(described in detail elsewhere, e.g. Robertsonet al. 1996), including the presence of an essen-tially unbroken Campanian to Lower Tertiarysedimentary sequence overlying the Troodosophiolite, and also the continuous upperPalaeozoic to Neogene sedimentary sequencesexposed in parts of the Tauride Mountains(notably in the Isparta Angle, SW Turkey; e.g.Robertson et al. 2003).

Age implications

Available radiometric ages for the Troodos,Hatay and Baër–Bassit ophiolites, and for theBaër–Bassit metamorphic sole, are shown inFigure 10, in relation to the geomagnetic polaritytime scale for the Late Cretaceous and thebiostratigraphical constraints on the timing oftectonic emplacement of Hatay and Baër–Bassitophiolites. Pre-deformational characteristicremanences in the Troodos ophiolite are ubiqui-tously of normal polarity, consistent with the U–Pb age of 90–92 Ma (Mukasa & Ludden 1987)indicating formation during the Cretaceous LongNormal Period (chron C34N; Fig. 10). Reversepolarity overprints are observed in the TroodosSheeted Dyke Complex (Gee et al. 1993), but arenot ubiquitous. Where they are present, normalpolarity characteristic magnetizations are alsoisolated. These overprints were demonstrablyacquired prior to tilting of the dykes (Gee et al.1993), and are attributed to low-temperaturealteration during reversed polarity chron C33R(early Campanian; Gee et al. 1993). Polarities ofcharacteristic magnetizations in the Hatayophiolite are dominantly normal, in contrast todominantly reversed polarities observed in theBaër–Bassit ophiolite. This suggests acquisitionof pre-deformational remanences during at leasttwo polarity chrons and hence significant varia-tions in age within the emplaced ophiolite sheet.Radiometric ages for the Hatay and Baër–Bassitophiolites are at present too poorly constrainedto resolve these age differences, with dykes yield-ing K–Ar ages in the range 73–99 Ma (Delaloye& Wagner 1984) spanning polarity chrons C33N,C33R and part of C34N (Fig. 10). Whetherreversed polarity remanences in Baër–Bassitcould be acquired during chron C33R, however,depends critically on the mode of formation ofthe Baër–Bassit metamorphic sole, which has aK–Ar age of 86–93 Ma (Thuizat et al. 1981;Delaloye & Wagner 1984), i.e. within chronC34N (Fig. 10). Assuming that this age is reli-able, alternative interpretations are as follows:(1) The metamorphic sole may have formed dur-ing initial detachment of the oceanic crust nearthe Neotethyan spreading axis (Whitechurch1977; Coleman 1981; Boudier et al. 1982), andhence represents the latest age of formation of theophiolite. The observed reversed polarities wouldthen require a substantial (possibly>30 Ma) agedifference between Baër–Bassit and Troodos,with the magnetization of the former beingacquired during the Early Cretaceous or within apoorly documented reverse polarity event withinchron C34N (Hailwood 1989). (2) The sole mayhave formed during the initiation of subduction,

Fig. 10. Summary of available radiometric ageconstraints for the Troodos, Hatay and Baër–Bassitophiolites and the Baër–Bassit metamorphic sole(together with biostratigraphic constraints on thetiming of emplacement of the Hatay and Baër–Bassitthrust sheets), in relation to the geomagnetic polaritytime scale for the Late Cretaceous (Cande & Kent1995). Sources are referred to in the text.

367PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

before SSZ spreading of the Baër–Bassit crust(e.g. Casey & Dewey 1984). This would reconcilethe available radiometric and magnetic polarityage constraints, and would require SSZ spreadingto have continued over a c. 10 Ma period betweenthe initiation of subduction in the Turonianand the start of microplate rotation in the Cam-panian. More reliable, higher resolution radio-metric dates for the Hatay and Baër–Bassitophiolite and metamorphic sole are clearlyrequired to resolve this debate.

Implications for intra-oceanic microplaterotation

Palaeomagnetic data from the Troodos ophioliteand its sedimentary cover are near universallyinterpreted in terms of intra-oceanic rotationof a ‘Troodos microplate’. Data from the sedi-mentary cover of the Troodos ophiolite indicatethat a large component (50–60°; Fig. 7) of intra-oceanic anticlockwise rotation had occurred bythe Maastrichtian, i.e. by the time of emplace-ment of the Hatay and Baër–Bassit ophiolitesheet onto the Arabian margin. This rotationangle is comparable with the mean rotationobserved in the Hatay ophiolite (Inwood et al.2003; Inwood 2005; Fig. 8a), after removing thepost-emplacement rotation recorded by its post-emplacement sedimentary cover (Inwood et al.2003; Kissel et al. 2003; Inwood 2005; Fig. 8c),and also represents a large component of themore extreme rotations observed in Baër–Bassit.

Hence, these data are most readily explainedby coherent rotation of a Neotethyan oceanicmicroplate that was more areally extensive thaninferred from the Troodos data alone (Inwoodet al. 2003; Inwood 2005).

The mechanism of microplate rotation isdifficult to identify with certainty, particularlyat the level of identifying accommodating (boun-ding) structures. A common feature of existingmodels (e.g. Clube et al. 1985; Clube &Robertson 1986; Robertson 1990) is that rotationis related to oblique convergence across thesouthern Neotethyan subduction zone, resultingfrom NE motion of Arabia relative to Eurasiathroughout the Late Cretaceous and Early Ter-tiary (Dewey et al. 1989). After correcting for theeffects of opening of the Red Sea, the Africanapparent polar wander path (Besse & Courtillot2002) places the northernmost Arabian conti-nental margin at c. 16°N during the Turonian,several 100 km to the south of the southernNeotethys spreading axis at 21–24°N (Fig. 11a).Subsequent motion of Arabia to the NE (Fig. 11band c) may then have generated a sinistral senseof motion across the subduction zone separatingthe SSZ oceanic crust from Arabia. Within thisoverall plate-scale framework, impingement ofthe Arabian continental margin with the sub-duction trench has been invoked as a potentiallymajor contributor to the initiation and pro-gression of microplate rotation (e.g. Clube &Robertson 1986; Robertson 1990). Althoughfurther higher resolution palaeomagnetic and

Fig. 11. Schematic illustration of the motion of the Arabian continental margin (Ar) through the LateCretaceous. Palaeolatitudinal constraints are derived from the African apparent polar wander path of Besse &Courtillot (2002), after correcting for the effects of Red Sea opening (using the Euler pole of Savostin et al.(1986)), Large grey arrow indicates the relative motion vector of Africa–Arabia relative to a fixed Eurasia (fromDewey et al. 1989). Thick black arrows illustrate the amount of palaeorotation of the Troodos microplatebetween time frames. The palaeolatitude of the microplate is accurately constrained only in (a), and cannot bedetermined reliably for subsequent time periods because of the potential effects of sedimentary compaction onpalaeomagnetic inclination.

368 A. MORRIS ET AL.

biostratigraphic data from the Troodos sedimen-tary cover are required to reduce uncertainties inthe progressive history of microplate rotation(Fig. 7b), rotation was clearly under way for atleast 10–15 Ma (and possibly up to 20 Ma) priorto ophiolite emplacement in the Maastrichtian.The timing of approach of the Arabian conti-nental margin to the trench cannot be accuratelydetermined, but if this impingement acted as atrigger for initiation of rotation then the palaeo-magnetic data suggest that rotation progressedfor a substantial period of time before eventualemplacement of part of the rotated unit onto thecontinental margin. Finally, we note that anti-clockwise intra-oceanic rotation of the easternMediterranean ophiolites considered here con-trasts with the clockwise rotation of the Oman(Semail) ophiolite further to the east (e.g. Weiler2000). This suggests that rotation in both caseswas controlled by plate-scale geometry of theArabian margin during regional convergence.Palaeomagnetic data are now required from theemplaced ophiolites to the east of Hatay andBaër–Bassit (i.e. the Kermanshah and Neyrizophiolites of Iran; Fig. 1) to further constrain thepattern and hence the mechanism of Neotethyanintra-oceanic rotations.

Conclusions

(1) The Late Cretaceous Troodos, Hatayand Baër–Bassit ophiolites of the easternMediterranean Tethyan orogenic belt pre-serve remanent magnetizations of pre-formational age. Palaeomagnetic data fromthese units, in conjunction with data fromthe overlying in situ (Troodos) and post-emplacement (Hatay and Baër–Bassit) sedi-mentary cover sequences, therefore provideinsights into the styles and timing of rota-tional deformation that have affected theNeotethyan oceanic crust.

(2) Within the Troodos ophiolite, localized rota-tions are demonstrably related to processesoperating during construction of oceaniccrust at a Neotethyan spreading axis in closeproximity to an oceanic transform faultzone. Rotations around (sub)horizontal,dyke-parallel axes are associated withcrustal extension during sea-floor spreading.Rotations around (sub)vertical axes domi-nate in areas adjacent to the South TroodosTransform Fault Zone, reflecting rotation ofkilometre-scale fault blocks in response todextral shearing within weak crust at theridge–transform intersection.

(3) The tectonically emplaced Hatay and Baër–Bassit ophiolites record large, and locallyextreme, anticlockwise rotations. Withinthe Baër–Bassit ophiolite, the magnitudeof observed rotations increases generallysouthwards. Analysis of the overlying post-emplacement sedimentary successions sug-gests that this amplification of rotationreflects the development of a strike-slip faultsystem related to the initiation of thepresent-day plate configuration.

(4) The Hatay ophiolite and the most northerlylocality in the Baër–Bassit ophiolite share asimilar anticlockwise rotation. After remov-ing the post-emplacement component ofrotation documented in the Hatay post-emplacement sedimentary cover, this rota-tion angle is directly comparable with thepre-Maastrichtian rotation of the ‘Troodosmicroplate’. At the regional scale, therefore,these data are best explained by intra-oceanic anticlockwise rotation of a moreareally extensive oceanic microplate thanhas been considered previously (Inwoodet al. 2003; Inwood 2005).

We would like to thank K. Al-Riyami for field assis-tance in the Baër–Bassit ophiolite, and U.-C. Ünlügençfor administrative and logistical support duringfieldwork in the Hatay ophiolite. Inclination-only tilttests were performed using software by R. Enkin.

References

ABELSON, M., BAER, G. & AGNON, A. 2001. Evidencefrom gabbro of the Troodos ophiolite for lateralmagma transport along a slow-spreading mid-ocean ridge. Nature, 409, 72–75.

ABRAHAMSEN, N. & SCHÖNHARTING, G. 1987. Palaeo-magnetic timing of the rotation and translation ofCyprus. Earth and Planetary Science Letters, 81,409–418.

AKTA¦, G. & ROBERTSON, A. H. F. 1984. The MadenComplex, S. E. Turkey: evolution of a Neotethyancontinental margin. In: DIXON, J. E. & ROBERTSON,A. H. F. (eds) The Geological Evolution of the East-ern Mediterranean. Geological Society, London,Special Publications, 17, 375–402.

ALLERTON, S. 1988. Palaeomagnetic and structuralstudies of the Troodos ophiolite, Cyprus. PhD thesis,University of East Anglia, Norwich.

ALLERTON, S. 1989a. Fault block rotations inophiolites: results of palaeomagnetic studies in theTroodos Complex, Cyprus. In: KISSEL, C. & LAJ, C.(eds) Palaeomagnetic Rotations and ContinentalDeformation. NATO ASI Series C, 254, 393–410.

ALLERTON, S. 1989b. Distortions, rotations and crustalthinning at ridge–transform intersections. Nature,340, 626–628.

369PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

ALLERTON, S. & VINE, F. J. 1987. Spreadingstructure of the Troodos ophiolite, Cyprus: somepalaeomagnetic constraints. Geology, 16, 593–597.

ALLERTON, S. & VINE, F. J. 1990. Palaeomagnetic andstructural studies of the southeastern part of theTroodos complex. In: MALPAS, J., MOORES, E. M.,PANAYIOTOU, A. & XENOPHONTOS, C. (eds) Ophio-lites: Oceanic Crustal Analogues. Geological SurveyDepartment, Nicosia, 99–111.

ALLERTON, S. & VINE, F. J. 1991. Spreading evolutionof the Troodos ophiolite, Cyprus. Geology, 19,637–640.

AL-RIYAMI, K. & ROBERTSON, A. H. F. 2002. Meso-zoic sedimentary and magmatic evolution of theArabian continental margin, northern Syria:evidence from the Baër–Bassit Mélange. GeologicalMagazine, 139, 395–420.

AL-RIYAMI, K., ROBERTSON, A. H. F., XENOPHONTOS,C., DANELIAN, T. & DIXON, J. E. 2000. Tectonicevolution of the Mesozoic Arabian passivecontinental margin and related ophiolite in Baër–Bassit region (NW Syria). In: PANAYIDES, I.,XENOPHONTOS, C. & MALPAS, S. (eds) Proceedingsof the Third International Conference on theGeology of the Eastern Mediterranean. GeologicalSurvey Department, Nicosia, 61–82.

AL-RIYAMI, K., ROBERTSON, A. H. F., DIXON, J. E. &XENOPHONTOS, C. 2002. Origin and emplacementof the Late Cretaceous Baer–Bassit ophiolite andits metamorphic sole in NW Syria. Lithos, 65,225–260.

BAILEY, W. R., HOLDSWORTH, R. E. & SWARBRICK, R.E. 2000. Kinematic history of a reactivated oceanicsuture: the Mamonia Complex Suture Zone, SWCyprus. Journal of the Geological Society, London,157, 1107–1126.

BESSE, J. & COURTILLOT, V. 1991. Revised andsynthetic apparent polar wander paths of theAfrican, Eurasian, North American and Indianplates and true polar wander since 200 Ma. Journalof Geophysical Research, 96, 4029–4050.

BESSE, J. & COURTILLOT, V. 2002. Apparent and truepolar wander and the geometry of the geomagneticfield over the last 200 Myr. Journal of GeophysicalResearch, 107, doi: 10.1029/2000JB000050.

BLACKMAN, D. K., CANN, J. R., JANSSEN, B. & SMITH,D. K. 1998. Origin of extensional core complexes:evidence from the Mid-Atlantic Ridge at theAtlantis Fracture Zone. Journal of GeophysicalResearch, 103, 21315–21333.

BONHOMMET, N., ROPERCH, P. & CALZA, F. 1988.Paleomagnetic arguments for block rotationsalong the Arakapas Fault (Cyprus). Geology, 16,422–425.

BORRADAILE, G. J. 2001. Paleomagnetic vectors andtilted dikes. Tectonophysics, 333, 417–426.

BORRADAILE, G. J. & LAGROIX, F. 2001. Magneticfabrics reveal upper mantle flow fabrics in theTroodos Ophiolite Complex, Cyprus. Journal ofStructural Geology, 23, 1299–1317.

BOUDIER, F., NICOLAS, A. & BOUCHEZ, J. L. 1982. Ki-nematics of oceanic thrusting and subduction frombasal sections of ophiolites. Nature, 296, 825–828.

BOULTON, S. J., ROBERTSON, A. H. F. & ÜNLÜGENÇ,U. C. 2006. Tectonic and sedimentary evolution of

the Cenozoic Hatay Graben, Southern Turkey: atwo-phase model for graben formation. In:ROBERTSON, A. H. F. & MOUNTRAKIS, D. (eds)Tectonic Development of the Eastern MediterraneanRegion. Geological Society, London, SpecialPublications, 260, 613–634.

CANDE, S. C. & KENT, D. V. 1995. Revised calibrationof the geomagnetic polarity timescale for the lateCretaceous and Cenozoic. Journal of GeophysicalResearch, 100, 6093–6095.

CANN, J. R., PRICHARD, H. M., MALPAS, J. G. &XENOPHONTOS, C. 2001. Oceanic inside cornerdetachments of the Limassol Forest area, Troodosophiolite, Cyprus. Journal of the GeologicalSociety, London, 158, 757–767.

CASEY, J. F. & DEWEy, J. F. 1984. Initiation of subduc-tion zones along transforms and accreting plateboundaries, triple junction evolution, and fore-arcspreading centres—implications for ophioliticgeology and obduction. In: GASS, I. G., LIPPARD,S. J. & SHELTON, A. W. (eds) Ophiolites and OceanicLithosphere. Geological Society, London, SpecialPublications, 13, 269–290.

CHANNELL, J. E. T., TÜYSÜZ, O., BEKTAS,, O. &³ENGÖR, A. M. C. 1996. Jurassic–Cretaceouspaleomagnetism and paleogeography of thePontides (Turkey). Tectonics, 15, 201–212.

CLUBE, T. M. M. 1985. The palaeorotation of theTroodos microplate. PhD thesis, University ofEdinburgh.

CLUBE, T. M. M., CREER, K. M. & ROBERTSON, A. H.F. 1985. The palaeorotation of the Troodosmicroplate. Nature, 317, 522–525.

CLUBE, T. M. M. & ROBERTSON, A. H. F. 1986. Thepalaeorotation of the Troodos microplate, Cyprus,in the Late Mesozoic–Early Cenozoic plate tectonicframework of the Eastern Mediterranean. Surveysin Geophysics, 8, 375–434.

COLEMAN, R. G. 1981. Tectonic setting for ophioliteobduction in Oman. Journal of GeophysicalResearch, 86, 2497–2508.

DELALOYE, M. & WAGNER, J.-J. 1984. Ophiolites andvolcanic activity near the western edge of theArabian plate. In: DIXON, J. E. & ROBERTSON,A. H. F. (eds) The Geological Evolution of the East-ern Mediterranean. Geological Society, London,Special Publications, 17, 25–233.

DELALOYE, M., PISKIN, Ö., SELÇUK, H., VUAGNAT, M.& WAGNER, J. -J. 1980. Geological section throughthe Hatay ophiolite along the Mediterranean coast,southern Turkey. Ofioliti, 5, 205–216.

DELAUNE-MAYÈRE, M. 1984. Evolution of a Mesozoicpassive continental margin: Baër–Bassit (NWSyria). In: DIXON, J. E. & ROBERTSON, A. H. F.(eds) The Geological Evolution of the EasternMediterranean. Geological Society, London,Special Publications, 17, 10–11.

DERCOURT, J., ZONENSHAIN, L. P., RICOU, L.-E., et al.1986. Geological evolution of the Tethys belt fromthe Atlantic to the Pamirs since the Lias.Tectonophysics, 123, 241–315.

DERCOURT, J., RICOU, L. E. & VRIELYNCK, B. (eds)1993. Atlas Tethys Palaeoenvironmental Maps.Beicip-Franlab, Rueil-Malmaison, France.

370 A. MORRIS ET AL.

DEWEY, J. F., HELMAN, M. L., TURCO, E., HUTTON,D. H. W. & KNOTT, S. D. 1989. Kinematics ofthe western Mediterranean. In: COWARD, M. P.,DIETRICH, D. & PARK, R. G. (eds) Alpine Tectonics.Geological Society, London, Special Publications,45, 265–283.

DIETRICH, D. & SPENCER, S. 1993. Spreading-inducedfaulting and fracturing of oceanic crust:examples from the Sheeted Dyke Complex of theTroodos ophiolite, Cyprus. In: PRICHARD, H. M.,ALABASTER, T., HARRIS, N. B. & NEARY, C. R.(eds) Magmatic Processes and Plate Tectonics.Geological Society, London, Special Publications,76, 121–139.

DILEK, Y. & THY, P. 1998. Structure, petrology andseafloor spreading tectonics of the KIzIldagophiolite, Turkey. In: MILLS, R. A. & HARRISON,K. (eds) Modern Ocean Floor Processes and theGeological Record. Geological Society, London,Special Publications, 148, 43–69.

DUNLOP, D. J. & ÖZDEMIR, Ö. 1997. Rock Magnetism:Fundamentals and Frontiers. Cambridge UniversityPress, Cambridge.

ENKIN, R. J. & WATSON, G. S. 1996. Statistical analysisof palaeomagnetic inclination data. GeophysicalJournal International, 126, 495–504.

ERENDIL, M. 1984. Petrology and structure of theupper crustal units of the KIzIl Dag ophiolite. In:TEKELI, O. & GÖNCÜO bGLÜ, C. M. (eds) Geology ofthe Taurus Belt. Proceedings, Mineral Researchand Exploration Institute, Ankara, 269–284.

GASS, I. G. 1968. Is the Troodos massif of Cyprus afragment of Mesozoic ocean floor? Nature, 220,39–42.

GASS, I. G. 1980. The Troodos massif: its role inthe unravelling of the ophiolite problem andits significance in the understanding of constructivemargin processes. In: PANAYIOTOU, A. (ed.) Ophio-lites: Proceedings of the International Symposium,Cyprus, 1979. Geological Survey Department,Nicosia, 23–35.

GASS, I. G., MACLEOD, C. J., MURTON, B. J.,PANAYIOUTOU, A., SIMONIAN, K. O. &XENOPHONTOS, C. 1991. Geological map of theSouth Troodos Transform Fault Zone at 1:25 000:Sheets 1 (west) and 2 (east). Geological SurveyDepartment, Nicosia.

GASS, I. G., MACLEOD, C. J., MURTON, B. J.,PANAYIOUTOU, A., SIMONIAN, K. O. &XENOPHONTOS, C. 1994. The Geology of the SouthTroodos Transform Fault Zone. Geological SurveyDepartment, Nicosia, Memoirs, 9.

GEE, J., VARGA, R., GALLET, Y. & STAUDIGEL, H.1993. Reversed-polarity overprint in dikes from theTroodos ophiolite: implications for the timing ofalteration and extension. Geology, 21, 849–852.

HAILWOOD, E. A. 1989. Magnetostratigraphy. Geo-logical Society, London, Special Report, 19.

HURST, S. D., VEROSUB, K. L. & MOORES, E. M. 1992.Paleomagnetic constraints on the formation of theSolea Graben, Troodos Ophiolite, Cyprus.Tectonophysics, 208, 431–445.

HURST, S. D., KARSON, J. A. & VEROSUB, K. L. 1994.Palaeomagnetism of tilted dikes in fast spread

oceanic crust exposed in the Hess Deep Rift:implications for spreading and rift propagation.Tectonics, 13, 789–802.

INWOOD, J. 2005. The tectonic evolution of the Hatay(KIzIl Dag) ophiolite, southern Turkey. PhDthesis, University of Plymouth.

INWOOD, J., MORRIS, A., ANDERSON, M. W.,ROBERTSON, A. H. F. & ÜNLÜGENÇ, U. 2003. Firstpalaeomagnetic results from the Hatay (KIzIl Dag)ophiolite of Turkey and their implication for thetectonic evolution of the eastern MediterraneanNeotethys. Geophysical Research Abstracts, 5,09595.

KAZMIN, V. G. & KULAKOV, V. V. 1968. The Geolo-gical Map of Syria, Scale 1:50 000 (Sheet Al-Latheqiyeh). Explanatory note. Technoexport,Nedra, Moscow.

KISSEL, C., LAJ, C., POISSON, A. & GÖRÜR, N. 2003.Paleomagnetic reconstruction of the Cenozoicevolution of the eastern Mediterranean.Tectonophysics, 362, 199–217.

MACDONALD, W. D. 1980. Net tectonic rotation,apparent tectonic rotation, and the structural tiltcorrection in palaeomagnetic studies. Journal ofGeophysical Research, 85, 3659–3669.

MACLEOD, C. J. 1988. The tectonic evolution of theeastern Limassol Forest Complex, Cyprus. PhDthesis, Open University, Milton Keynes.

MACLEOD, C. J. 1990. Role of the Southern TroodosTransform Fault in the rotation of the Cyprusmicroplate: evidence from the Eastern LimassolForest. In: MALPAS, J., MOORES, E. M.,PANAYIOTOU, A. & XENOPHONTOS, C. (eds)Ophiolites: Oceanic Crustal Analogues. GeologicalSurvey Department, Nicosia, 75–85.

MACLEOD, C. J. & MURTON, B. J. 1993. Structureand tectonic evolution of the Southern TroodosTransform Fault Zone, Cyprus. In: PRICHARD, H.M., ALABASTER, T., HARRIS, N. B. W. & NEARY, C.R. (eds) Magmatic Processes and Plate Tectonics.Geological Society, London, Special Publications,76, 141–176.

MACLEOD, C. J. & MURTON, B. J. 1995. On the sense ofslip of the Southern Troodos Transform-FaultZone, Cyprus. Geology, 23, 257–260.

MACLEOD, C. J., ALLERTON, S., GASS, I. G. &XENOPHONTOS, C. 1990. Structure of a fossil ridge–transform intersection in the Troodos ophiolite.Nature, 348, 717–720.

MALPAS, J., CALON, T. & SQUIRES, G. 1993. Thedevelopment of a Late Cretaceous microplatesuture zone in SW Cyprus. In: PRICHARD,H. M., ALABASTER, T., HARRIS, N. B. W. &NEARY, C. R. (eds) Magmatic Processes and PlateTectonics. Geological Society, London, SpecialPublications, 76, 177–195.

MCELHINNY, M. W. 1964. Statistical significance ofthe fold test in palaeomagnetism. GeophysicalJournal of the Royal Astronomical Society, 8,338–340.

MCFADDEN, P. L. & JONES, D. L. 1981. The fold test inpalaeomagnetism. Geophysical Journal of the RoyalAstronomical Society, 67, 53–58.

MOORES, E. M. & VINE, F. J. 1971. The TroodosMassif, Cyprus and other ophiolites as oceanic

371PALAEOMAGNETIC INSIGHTS INTO NEOTETHYS

crust: evaluation and implications. PhilosophicalTransactions of the Royal Society of London, SeriesA, 268, 433–466.

MORRIS, A. 2003. The Late Cretaceous palaeolatitudeof the Neotethyan spreading axis in the easternMediterranean region. Tectonophysics, 377,157–178.

MORRIS, A. & ANDERSON, M. W. 2002. Palaeo-magnetic results from the Baër–Bassit ophiolite ofnorthern Syria and their implication for fold tests insheeted dyke terrains. Physics and Chemistry of theEarth, 27, 1215–1222.

MORRIS, A., CREER, K. M. & ROBERTSON, A. H. F.1990. Palaeomagnetic evidence for clockwise rota-tions related to dextral shear along the SouthernTroodos Transform Fault, Cyprus. Earth andPlanetary Science Letters, 99, 250–262.

MORRIS, A., ANDERSON, M. W. & ROBERTSON, A. H.F. 1998. Multiple tectonic rotations and transformtectonism in an intra-oceanic suture zone, SWCyprus. Tectonophysics, 299, 229–253.

MORRIS, A., ANDERSON, M. W., ROBERTSON, A. H. F.& AL-RIYAMI, K. 2002. Extreme tectonic rotationswithin an eastern Mediterranean ophiolite (Baër–Bassit, Syria). Earth and Planetary Science Letters,202, 247–261.

MORRIS, A., INWOOD, J., ANDERSON, M. W. &ROBERTSON, A. H. F. 2006. First palaeomagneticresults from Tertiary carbonates of NW Syria, andtheir implications for the timing of extreme tectonicrotations in the Baër–Bassit ophiolite. Earth andPlanetary Science Letters (in press).

MUKASA, S. B. & LUDDEN, J. N. 1987, Uranium–leadisotopic ages of plagiogranites from the Troodosophiolite, Cyprus, and their tectonic significance.Geology, 15, 825–828.

MURTON, B. J. 1986. Anomalous oceanic lithosphereformed in a leaky transform fault: evidence fromthe western Limassol Forest Complex, Cyprus.Journal of the Geological Society, London, 143,845–854.

MURTON, B. J. 1990. Was the Southern TroodosTransform Fault a victim of microplate rotation?In: MALPAS, J., MOORES, E. M., PANAYIOTOU,A. & XENOPHONTOS, C. (eds) Ophiolites: OceanicCrustal Analogues. Geological Survey Department,Nicosia, 87–98.

PARROT, J. F. 1980. The Baer–Bassit (NorthwesternSyria) ophiolitic area. Ofioliti, 2, 279–295.

PEARCE, J. A. 1975. Basalt geochemistry used to inves-tigate past tectonic environments in Cyprus.Tectonophysics, 25, 41–67.

PEARCE, J. A. 1980. Geochemical evidence for thegenesis and setting of lavas from Tethyanophiolites. In: PANAYIOTOU, A. (ed.) Ophiolites:Proceedings of the International Symposium,Cyprus, 1979. Geological Survey Department,Nicosia, 261–272.

PI¦KIN, O., DELALOYE, M. & WAGNER, J. J. 1986.Guide to the Hatay geology. Ofioliti, 11, 87–104.

RICOU, L.-E. 1971. Le croissant ophiolitique péri-arabe, une ceinture de nappes mises en place auCretacé supérieur. Revuede Géographie Physiqueet de Géologie Dynamique, 13, 327–349.

RICOU, L.-E., MARCOUX, J. & POISSON, A. 1979.L’allochthonie des Bey DaglarI orientaux, Recon-struction palinspastique des Taurides occidentales.Bulletin de la Societé Géologique de France, 21,125–134.

RICOU, L.-E., MARCOUX, J. & WHITECHURCH, H.1984. The Mesozoic organisation of the Taurides:one or several ocean basins. In: DIXON, J. E. &ROBERTSON, A. H. F. (eds) The Geological Evo-lution of the Eastern Mediterranean. GeologicalSociety, London, Special Publications, 17,349–360.

RIZAO bGLU, T., PARLAK, O., HÖCK, V. & dIcSLER, F. 2006.Nature and significance of Late Cretaceousophiolitic rocks and their relation to the Baskilgranitic intrusions of the ElazIg region, SE Turkey.In: ROBERTSON, A. H. F. & MOUNTRAKIS, D. (eds)Tectonic Development of the Eastern MediterraneanRegion. Geological Society, London, Special Publi-cations, 260, 327–349.

ROBERTSON, A. H. F. 1977. The Kannaviou Forma-tion, Cyprus: volcaniclastic sedimentation of aprobable Late Cretaceous volcanic arc. Journal ofthe Geological Society, London, 134, 269–292.

ROBERTSON, A. H. F. 1986. Geochemistry and tectonicimplications of metalliferous and volcaniclasticsedimentary rocks associated with Late Cretaceousophiolitic extrusives in the Hatay area, SouthernTurkey. Ofioliti, 11, 121–140.

ROBERTSON, A. H. F. 1990. Tectonic evolution ofCyprus. In: MALPAS, J., MOORES, E. M.,PANAYIOTOU, A. & XENOPHONTOS, C. (eds) Ophio-lites: Oceanic Crustal Analogues. Geological SurveyDepartment, Nicosia, 235–252.

ROBERTSON, A. H. F. 1998. Mesozoic–Tertiary tec-tonic evolution of the easternmost Mediterraneanarea: integration of marine and land evidence. In:ROBERTSON, A. H. F., EMEIS, K.-C., RICHTER, C. &CAMERLENGHI, A. (eds) Proceedings of the OceanDrilling Program, Scientific Results, 160, OceanDrilling Program, College Station, TX, 723–782.

ROBERTSON, A. H. F. 2002. Overview of the genesisand emplacement of Mesozoic ophiolites in theEastern Mediterranean Tethyan region. Lithos, 65,1–67.

ROBERTSON, A. H. F. & DIXON, J. E. 1984. Introduc-tion: aspects of the geological evolution of the East-ern Mediterranean. In: DIXON, J. E. & ROBERTSON,A. H. F. (eds) The Geological Evolution of the East-ern Mediterranean. Geological Society, London,Special Publications, 17, 1–74.

ROBERTSON, A. H. F. & WOODCOCK, N. H. 1979. TheMamonia Complex, southwest Cyprus: the evolu-tion and emplacement of a Mesozoic continentalmargin. Geological Society of America Bulletin, 90,651–665.

ROBERTSON, A. H. F. & XENOPHONTOS, C. 1993.Development of concepts concerning the Troodosophiolite and adjacent units in Cyprus. In:PRICHARD, H. M., ALABASTER, T., HARRIS, N. B.W. & NEARY, C. R. (eds) Magmatic Processes andPlate Tectonics. Geological Society, London,Special Publications, 76, 85–119.

ROBERTSON, A. H. F., CLIFT, P. D., DEGNAN, P. J.& JONES, G. 1991. Palaeogeographic and palaeo-tectonic evolution of the Eastern Mediterranean

372 A. MORRIS ET AL.

Neotethys. Palaeogeography, Palaeoclimatology,Palaeoecology, 87, 289–343.

ROBERTSON, A. H. F., DIXON, J. E., BROWN, S., et al.1996. Alternative tectonic models for the LatePalaeozoic–Early Tertiary development of Tethysin the Eastern Mediterranean region. In: MORRIS,A. & TARLING, D. H. (eds) Palaeomagnetism andTectonics of the Mediterranean Region. GeologicalSociety, London, Special Publications, 105,239–263.

ROBERTSON, A. H. F., POISSON, A. & AKINCI, O. 2003.Developments in research concerning Mesozoic–Tertiary Tethys and neotectonics in the IspartaAngle, SW Turkey. Geological Journal, 38,195–234.

ROBINSON, P. T. & MALPAS, J. 1990. The Troodosophiolite of Cyprus: new perspectives on its originand emplacement. In: MALPAS, J., MOORES, E. M.,PANAYIOTOU, A. & XENOPHONTOS, C. (eds) Ophio-lites: Oceanic Crustal Analogues. GeologicalSurvey Department, Nicosia, 13–36.

SAVOSTIN, L. A., SIBUET, J.-C., ZONENSHAIN, L. P.,LE PICHON, X. & ROULET, M.-J. 1986. Kinematicevolution of the Tethys Belt from the AtlanticOcean to the Pamirs since the Triassic.Tectonophysics, 123, 1–35.

SCHMINCKE, H.-U. & RAUTENSCHLEIN, M. 1987.Troodos extrusive series (Akaki River Canyon) andthe Sheeted Diabase. In: XENOPHONTOS, C. &MALPAS, J. G. (eds) Field Excursion Guidebook toAccompany the Troodos 87 Ophiolites and OceanicLithosphere Symposium. Geological SurveyDepartment, Nicosia, 36–91.

SCHMINCKE, H.-U., RAUTENSCHLEIN, M. ROBINSON,P. T. & MEGEHAN, J. M. 1983. Troodos extrusiveseries of Cyprus: a comparison with oceanic crust.Geology, 11, 405–409.

SHELTON, A. W. & GASS, I. G. 1980. Rotation ofthe Troodos microplate. In: PANAYIOTOU, A.(ed.) Ophiolites: Proceedings of the InternationalSymposium, Cyprus, 1979. Geological SurveyDepartment, Nicosia, 61–65.

SIMONIAN, K. O. & GASS, I. G. 1978. Arakapas faultbelt, Cyprus: a fossil transform belt. GeologicalSociety of America Bulletin, 89, 1220–1230.

STAUDIGEL, H., GEE, J., TAUXE, L. & VARGA, R. J.1992. Shallow intrusive directions of sheeted dikesin the Troodos ophiolite: anisotropy of magneticsusceptibility and structural data. Geology, 20,841–844.

SWARBRICK, R. E. 1979. The sedimentology and struc-ture of SW Cyprus and its relationship to theTroodos Complex. PhD thesis, University ofCambridge.

SWARBRICK, R. E. 1980. The Mamonia Complex ofSW Cyprus: a Mesozoic continental margin andits relationship to the Troodos Complex. In:

PANAYIOTOU, A. (ed.) Ophiolites: Proceedingsof the International Symposium, Cyprus, 1979.Geological Survey Department, Nicosia, 86–92.

SWARBRICK, R. E. 1993. Sinistral strike-slip andtranspressional tectonics in an ancient oceanicsetting: the Mamonia Complex, southwest Cyprus.Journal of the Geological Society, London, 150,381–392.

THUIZAT, R., WHITECHURCH, H., MONTIGNY, R. &JUTEAU, T. 1981. K–Ar dating of some intrao-phiolite metamorphic soles from the East Mediter-ranean: new evidence for oceanic thrustingbefore obduction. Earth and Planetary ScienceLetters, 52, 302–310.

USTAÖMER, T. & ROBERTSON, A. H. F. 1993. A LatePalaeozoic–Early Mesozoic marginal basin alongthe active southern continental margin of Eurasia:evidence from the Central Pontides (Turkey) andadjacent regions. Geological Journal, 28, 219–238.

USTAÖMER, T. & ROBERTSON, A. H. F. 1994. LatePalaeozoic marginal basin and subduction–accretion: the Palaeotethyan Küre Complex,Central Pontides, northern Turkey. Journal of theGeological Society, London, 151, 291–305.

VARGA, R. J. & MOORES, E. M. 1985. Spreading struc-ture of the Troodos ophiolite, Cyprus. Geology, 13,846–850.

VARGA, R. J., GEE, J. S., STAUDIGEL, H. & TAUXE,L. 1998. Dike surface lineations as magma flowindicators within the sheeted dike complex ofthe Troodos ophiolite, Cyprus. Journal ofGeophysical Research, 103, 5241–5256.

VEROSUB, K. L. & MOORES, E. M. 1981. Tectonic rota-tions in extensional regimes and their paleomag-netic consequences for oceanic basalts. Journal ofGeophysical Research, 86, 6335–6349.

VINE, F. J. & MOORES, E. M. 1969. Palaeomagneticresults from the Troodos Igneous Massif, Cyprus.Transactions of the American Geophysical Union,50, 131.

WEILER, P. D. 2000. Differential rotations in the OmanOphiolite: paleomagnetic evidence from the south-ern massifs. Marine Geophysical Researches, 21,195–210.

WHITECHURCH, H. 1977. Les roches métamorphiquesinfrapéridotitiques du Baër–Bassit (NW Syrien),témoins de l’écaillage intraocéanique téthysien.Étude pétroloqique et structurale. PhD thesis,Nancy University, France.

YAZGAN, E. & CHESSEX, R. 1991. Geology and tectonicevolution of Southeastern Taurus in the regionof Malatya. Turkish Association of PetroleumGeologists Bulletin, 3, 1–42.

YILMAZ, Y. 1993. New evidence and model on theevolution of the southeast Anatolian orogen.Geological Society of America Bulletin, 105,251–271.