Arguments for Large Sca/E Comp/Ex Thin-Skinned and Thick-Skinned Rotations

Hercynjan pa/aeomagnetism of Europe: arguments for large sca/e comp/ex thin-skinned and thick-skinned rotations

y. BACHTADSE Department of Earth Selences, Oxford University, Parks Road, Oxford, OXI 3PR, UK

ABSTRACT

Palaeomagnetic data for ihe Devonian and Carboniferous of the Eu- ropean Hercynides and from their stable foreland (northern Europe and the BriLish Isles) display a very coherenL pattern in inclinaLion, pointing towards consolidation of Hercynian Europe during Lhe mid Lo late Devo- man. Declination data however, show a significant dispersion and can be used in order to unravel the deformation history of the mountain belt. Differences between the observed palaeomagnetic declinations and those expeeted for the specific areas and the age of magnetizaLion can be corre- lated to the changes in regional sLrike. Linear regression analysis of the declination deviations, shows a very strong correlaLion (close to unity) for [hose areas, where thin-skinned nappe emplacement has been most pre- valent. This suggests the predominantly primary characLer of the strucLu- ral curvature. Data from presumably autochthonous parts of Lhe orogen are more scattered but nevertheless yield a significant, but shallower regression line. A L-test of the slope against zero slope is signifieant aL the 95 0/o confidence level, indicating secondary (oroclinal) bending of an originally deformed fold system for the autochLohonous parts of the rnountain beIL. It can be inferred, that the present shape of Lhe Hercynian moun- tarn belL is the resulL of combined recurrent deep-reaching deformation of the lithosphere and the superimposed effects of thin-skinned thrust rotations, compatible with geodynamic modeis involving the indentation of Hercynian Europe by a microplate or an African promontory during the Hercynian orogeny. 148 J/~ Baehtadse

INTRODUCTION ‘¡he geodynamic evolution of the European Hercynides has been the subject of intense research during the last two decades and raiher anta- gonistic scenarios for the Palaeozoic crustal consolidation of Europe have been proposed by nurnerous authors (e.g. Lorenz and Nicholís, 1984, Behr el al.., 1984, Maite, 1986, Ziegler, 1986 and references Iherein). I-lowe- ver, a unifying model for I-Iercynian geodynamics, which integrates geophysical data and geological observations has not been brought for- ward yet. Ihe Hercynian foid belt of Europe is generally characterized by a central polymetamorphic crystalline beiL, bordered on both sides by ouLward facing foid and thrust belts, consisting of unmetamorphosed or low grade sedirnentary and volcanic assernblages. Although there is evidence for a localized tectonic event during the Devonian (Acadian), the main oroge- níc pulse has been shown to he Carboniferous (i.e. Late Visean to Namu- rian) in age (Behr et al., 1984 and references therein>. The most promi- nent feature of the European Hercynides is the well pronounced curvature of the rnountain belt as defined by the change in the structural trend (Fig. 1). Ihis is rnost spectacular around the Bay of Biscay, where the to- tal bending of the Ibero-Armorican arc in a pre-Mesozoic reconstruction

1 .‘ ( ¡

Fig. 1. Geological sketch map of the European Hercynides in pre-Mesozoic configuration (Van der ‘Voo, 1969). Sho’wn are ihe directiens of Ihe siructural trend (1), ihe norihein ovcrthrust (2), and ihe directions of Devonian and Carbonifereus paleomagnetic declina- iions from autochthonous (3) and allochthonous (4) areas. A: Carpathians; B: Central Eu- ropc (Han Mountains and Franconian Forest); C: Ardenne-Fifel; D: Massif Central; E: Wa- les; E: Armorica; G: Cantabria; H: Galicia-Castilla. Ilerc~’nian palaeomagnetism of Europe: arguments... 149

(Van der Voo, 1969) approximaLes 165’ (Ries and Shackleton, 1976). In Central Europe, Lhe change from the Hercynian (NW-SE) to the Variscan (SW-NE) sLriking structural trend is less dramaLie but still amounts to about 40’. Finally a change in the structural trend of about 80’ can be observed in Lhe Carpathians. Based on palaeomagnetic (Van der Voo, 1979) and general geological evidence (e.g. Ziegler, 1986 and references therein), it is now widely accepted, that the Palaeozoic consolidation of Western Europe was controlled by the convergence of three major plates (Laurentia, Baltica and Gondwana) as well as by the amalgamation and deformation of srnaller, Gondwana-derived, crusLal elements such as the Armorica microplaLe (Van der Voo, 1979) and the Iberian and Austro-alpine allochthonous terranes (Ziegler, 1983). In this paper a brief review of the Devonian to Carboniferous palaeomagnetic data from Central Europe will be given and an attempt will be made Lo elucidate a sLrucLural and geodynamic in- terpretaLion.

‘¡HE PALAEOMAGNETIC REFERENCE FRAMEWORK ‘¡he strucLural significance of palaeomagnetic daLa from deformed a- reas hinges crucially on the quality of the reference daLa base from a stable foreland or craton. It is now widely accepted, that Laurentia, Baltica, the British Isles south of the lapetus and the Central European pre-Va- riscan basement (sLable Europe), have been assembled during the Caledo- nian orogeny (e.g. Soper and Hutton, 1984). Devonian and Carboniferous palaeornagnetic data from the Baltic shield and the British Isles can therefore be used as reference for Lhe geodynamic and structural interpretation of palaeomagnetic data from 11w European Hereynides. ‘¡he various proposed apparent polar wander (APW) paths for stable Europe (Briden et al., 1973, McElhinny, 1973, French, 1976, Duff, 1980) agree raLher well for the Silurian Lo early Carboniferous period. Differen- ces ofup Lo 10’ in the longitudinal position of the Carboniferous poles between various proposed models (Duff, 1980; Briden, 1973, McElhinny, 1973) are however conspicuous, and point towards unresolved problems in age calibration (Perroud, 1986). Following a different une of argument, Edel (1 987a) postulated thaL notably the Devonian and Carboniferous parts of Lhe APW path of stable Europe have been severely affected by Perrno-Carboniferous remagnetizations and Lherefore cannot be used as a reference for the interpretation of pre-Variscan directions. Conse- quently, he proposed to use the APW path for the Russian platform instead. ‘¡he rather tight cluster of Devonian to Permian poles from European Russia (Khramov et al., 1981) however, which are noL significantly different from western European Permian palaeopoles, itidicates 150 U. Baehtadse unresolved pervasive Kiaman (Permo-Carboniferous) remagnetization of the Russian Palaeozoic rocks. Lacks fo accessibility of the original publications withdraws those dala from any further serious evaluation. ‘¡he more vigorous use of modern statistical methods and reliability criteria resulted in several published mean palaeopole positions for stable Europe (Table 1) considered to be more credible and used as reference for this paper.

PALAEOMAGNETIC FRAMEWORK OF THE EUROPEAN HERCY- NIDES Widespread Kiaman remagnetization is a predominan[ feature of the European Palaeozic palaeomagnetic record (Creer, 1 968). Although we are still far from understanding the dynamics of this conLinení, it be- comes more and more obvious, that the remagnetization process is controlled by chemical rather than thermal condiíions (Courtillot et al., 1986). Fluid migration related to Hercynian thrusting can poten- tially provide the chemical conditions for such a large scale event (McCa- be et al., 1983). It comes therefore notas a surprise, that only a minority of the reliable palaeomagnetic pole positions tabulated (Table 2) are thought to refleet primary (e.g. prefolding) directions. Only five primary palaeomagnetic palaeopoles have been reponed for the Devonian so far, namely from the Iberian Meseta (Perroud and Bon- hommet, 1984), Lhe Harz mountains and Franconian Forest (Bachtadse, 1984), Wales (McClelland Brown, 1983) and more recently from the Holy Cross Mountains, Poland (Lewandowski et al., 1987). Except for the Po-

TABLE 1 Siluro-Devonian to Permian reference peles for stable Europe

Pote

Age Lat. Long. A95 N Reference ‘E

Siluro-Devonian —2 321 Duff, 1980 Mid-Devenian 18 332 3 II Pesonen et al., 1987 Late Devonian Early Carbonifereus 30 333 3 8 Duff, 1980 Mid to late Carbonifcrous 37 341 3 6 Frey and Cex, 1987 Late Carboniferous 40 344 3 21 Frey and Cox, 1987

Lal. and Long, are latisude and longilude of Ihe palacopole posilion u degrees SouIh ancí Ezíl, respecíively. N it dic number of dala entries aud .495 it dic radius of Ihe cone of confidence al dic 95% probability le-va. Ilercvnian palaeomagnetisni of Europe: argurnenis... 151

ES .5555 50 55 oSEo, 5555 55555555’55~55 .5.5.5 55-55 555~55555555~ 5 565 55555 5 5555 ~5555 0a2 55= - O 55 5. 55 0.55 555555 5 55 55.55555. 55550 5 55 55 -p 05. — E—:— tOs -“ O5550 55555555 ~~00 5o~- -~“000 5—— .55.55.55.55 E A 550 5555550 0055 555555 0.5.> 55.55550. 53555555 75555 55555555

‘-<“E EE0~ “EQ o 00 (.3000>000000 0.”0 00-j oDE tECE E 0 0002 2

— - $~55 ¡

-¡rl

sg 5 0

5 5 O 5 55

5. -5

YO E 5555 550 - 70 Eso. - . ‘5 5 .5 55 55-55 .55555 0. ~--.55555551 ~0t-~ 6655’ 55 ~o~oE .tE~5ss # ~ UZa~57EoEEE ~50 ~5535555< 05955<0 55555 ~sEóo 5.~. 55 55 ~ GZ25555 ‘055<55 0s.o5555 055 05 152 V. Baehtadse lish result, [he primary character of those magnetizations has been sup- ported by positive fold tests. Ihe Devonian palaeomagnetic inclinations (Table 2) show a raLher good agreement wiLh Ihe inclinations derived from the mid to late Devo- man reference pole for stable Europe (Table 1), putting Armorica next to Laurentia Lo the East and Baltica Lo the North. Any wide Palaeozoic ocean

(cg. mid Lo late Devonian - early Carboniferous, see Johnson, 1974) must therefore he indentified South of the Central Crystalline belt (Moldanu- bian) of Lhe Hercynian orogen. Althought a detailed review of the Palaeozoic palaeomagnetism of Gondwana is beyond Lhe seope of this account, several recent studies of Palaeozoic rocks from Gondwana will be discussed briefiy since their interpretation has a direct bearing on the understanding of the tectonic evolution of Hercynian Europe. Unfortunately, Palaeozoic data from Gond- wana published during the last years is rather controversial. Earlier re- constructions of the palaeo-geodynamics of the circum-AtlanLic conti- nents for Lhe late Devonian (Van der Voo, 1982, Van der Voo, 1983), no- sition the leading edge of Africa at about 37’ South, allowing for an ocean of maximal 3,700 km in width between Africa and sothern Europe. ‘¡he position of Africa was exclusively based on the result for the Msissi norite (Hailwood, 1974), which has been cited as late Devonian in age. A re-investigaLion of the Msissi key pole by Salmon and co-workers (1987) however, demontrated a Jurassic age of the Msissi norite. FurLheremore Kent et al. (1984) reported a pole position for the early to mid Devonian Oneiguira Formation in Mauritania, which plots off southern Africa, and which would imply, that any ocean separating Africa and Europe must have been vanished below palaeomagnetic resolution in the Devonian, e.g. before the main pulse ofdic Hercynian orogeny. Saradeth el al. (1987) further more reponed a palacopole position from the Devonian Sabaloka ring complex, Sudan, very similar to the one from the Gneiguira Forma- 1 ion. Conflicting palaeomagnetic results have also been reponed from Aus- tralia. A very well defined palaeopole for the late Devonian from the cra- tonic Canning Basin (Hurley and Van der Voo, 1986), puts the pole into Central Africa (vcry close to the original Msissi position), while Schmidt et al. (1987), on the basis of palaeomagnetic data from the eastern Aus- tralian Lachían fold belt, proposed a Devonian pole posiLion east of the southern tip of Africa. There is howevcr reason fon the suspicion that Lhe Gneiguira pole is based on a Carboniferous remagnetization of the sedimenLs sLudied (May and KenL, 1987) and thaI Lhe palaeomagnetic direction from the Austra- han Lachían fold belt has been deflected by tectonic rotations of eastern Australia (Hurley and Van der Voo, 1987). Studies on Palaeozoic sedi- menís from South Africa (Bachtadse et al., 1987) as well as from sout- Hercynian palaeomagnetism of Europe: arguments... 153 hern Iberia (Perroud and Bonhommet, 1984) yielded furLher support for a Central African position of Lhe palaeopole during Devonian times and are in support of Van der Voo’s (1983) reconstruction. However much more high quality palaeomagnetic data fon the Devonian of Africa is ur- gently needed before the pre-Hereynian palaeogeography of Europe and Africa and Hercynian geodynamies can be finally assessed.

PALAEOMAGNETIC DECLINATTONS

While observed palacomagneLie inclinations from a wide vaniety of rocks and areas within the Hercynian mountain chain show a remarkably good agreement with the inclinatíons expecLed for the particular area and the given age of magnetization (Table 2), Lhere exisLs a rather large varia- Lion in palaeomagnetic declinations along the belL (Eldredge eL al., 1985). Ihose changes in declination appear Lo be systematic anomalies are in following the trend of the northern Hercynian front (Fig. 1). As declination anomalies are indicative for tectonic rotations about vertical axis, a detailed analysis of palaeomagneLic data can reveal furLher insight inLo Lhe strucíural evolution of the mountain beIl (Van den Voo and Channel, 1980). Any correlation between the change of strike along Ihe Hercynian Froní and the change in declination can be tested rather sLraightforwar- dly using graphic (Fig. 2) and maihematical (linear rcgression) methods (Schwartz and Van der Voo, 1983). The deviation of the locally observed strucLural trend (Sa) can be determined by subtracLion from an arbiLranily chosen reference value (S5). Accordingly, changes in declination will be revealed by comparing the locally detenmined observed palaeomagnetic declination (D0) Lo Lhe reference declination (De), which has been compuLed for the locality and Ihe age assigned Lo the observed declinaLion using the relevant palaeopole position (Table 2) as reference. Linear regression on Lhose two variables (Sr~So, Dr~Do) reveals Lhe nature of the data set. I-Iigh correlation coefficienLs and regression slopes nean unity (m = 1) sug- gesL, that Ihe palaeomagnctic declinations have seen thc same amount of roLation as Lhe structural grain (Fig. 2). Significant deviation from unity poinís either Lowards secondary Lightcning (m<) or secondary opening of an originally bent mountain beIL (m>). Regression siopes, which are not significanLly differení from the abscissa (m=0), suggest that tectonie deformation is likely Lo pre-daLe Lhe acquisition of the magnetization. Using Lhe dala set for Lhe mid-Devonian to late Carboniferous from Lhe European Hercynides, Eldredge eL al. (1985), concluded that there is a good correlaLion between changes in regional strike and anomalies in declination. The inLermediate siope of thc reponed regression line strongly suggested that the cunvature of the Hercynian belt is in facL a secondary feature, an onocline in Lhe sense of Caney (1958). A more detailed inspee- 154 y Bachtadse

DrDo m>1 m~i 1~• y

m’1 ~2~

—e

SrSo ,2 ~—1/ • — t// e— -5.---• x —5.-

Fig. 2. PIot of declinatien anomalies as a function of strike anomalies. Slopes of the regression une near unity (m=I) indicate a primary origin of the curvature. Siopes signifs- cantly different from unity are indicative for secondary tightening (m<1) or opening (ni> 1) of aB originally curved structure.

tion of the same data and of the tectonic setting of the individual directions in particular (Bachtadse and Van der Voo, 1986) however revealed that abouL 20 % of the data points originally analyzed, oniginaLed from thin-skinned data set into allochLhonous arcas such as the Cantabrian Arc and Wales. Dividing the data set into allochthonous and autochthonous groups and perfonming linear regression on both scts independently, yielded two signifucantly different regression lines. While the data from thin-skinned arcas such as Cantabria and Wales showed an extremely good correlation between changes in sLrike and declination (close to unity), Ihis was much less pronounced fon data from presumably autochthonous pants of the mountain bclt, yielding a rather shallow slope of the Hercynian palaeomagnetism of Europe: arguments... ‘55 regression line. Consequently, Bachtadse and Van den Voo (1986) arrived at the conclusion, that the bulk of Hercynian deformation had been ta- ken up by thin-skinned thrusting in the outer pan of the mountain belt, while the final collision of Africa and Europe during the Carboniferous led to further tightening of the already bent inner (auLochthonous) regions of the mountain belt. Additional dala, which has emerged duning the last couple of years such as from the allochtchonous Ardennc-Eifel Massif (Nowaczyk and Bleil, 1985, Edel and Coulon, 1985), and the Cantabrian arc (I-IirL et aL, 1987) as well as fnom the presumably autochthonous Massif Central, France (Edel, 1984, Edel, 1 987b), ‘¡he Armorican Massif (Perroud et aL, 1 976a,b) and Ihe holy Cross Mountains, Poland (Lewandowski et al., ¡987), resulted in widen geographic coverage of Lhe palaeomagnetic data and allows a more stringent evaluation of the large scale structural setting of the Hencynidcs. In order to minimize the dcviations in declination, normal directions tabulated (Table 3), have been reversed. Palaeomagnetic and structural data from the Iberian Peninsula have been adjusted for the Mesozoic opening of the Bay of Biscay by adding 35’ (Van der Voo, 1969) to stnike and declination. ‘¡he reference declination for each study has been computed using the reference palacopole positions given in Table 1, and thc age of (re-) magneLization as assigned by the authors of the original publications (Table 2). A strike of 210’ hastren used for reference (Sr). Vaniations in declination (D~-D0) have been plotted as a function of thc change in general strike (Sr-So; Fig. 3) The errors asso- ciated with the obsenved declinations (‘¡able 3 and error bars in Fig. 3) have been calculated according to Beck (1980). Following BachLadse and Van der Voo (1987) thc structural sctting of each palacomagnetie datum had been determined and a statistical analysis had been carried out on [he autochthonous (28 entries) and allochthonous (11 entries) sub sets independently. Linear negression of the data from the allochthonous outer zones of the Hencynides, namely the Ardenne, South Wales and, mosí pronounced, Cantabria (see review by Mattc, 1987), yield a very well defined regnession line (conrelation coefficient r = 0.933, line «A» in Fig. 3). A t-test

(e.g. Lownie and Hirt, 1986) of the slope (m = 0.695) ofLhe regression une

against zeno slope results in t = 7.785 which is significant at the 95 % confidence level. The same test against uniL slope (pure primary curvature)

gives t = 1.918, which is marginally greater than thc reference value (1.833) at the 95 % level, and thenefore indicates, that the slope of the regression line is significantly diffenent fnom unity. ‘¡he rather steep slope of ¡inc «A» neverthelcss points towards the predominantly pnimary curvature of the structural tnend observed in the thin-skinned zones of the orogen. Li- nean regnession of the «autochthonous» data set (line «B» in Fig. 3), re- 156 PI Rachtadse

TABLE 3 Deviatien of observed declina[ions and s[rike from reference directions and reference strike

regional DecI/Inc Localiíy sírucí. 5~-5~ D,-0 Erend 0 hl0 dD observed expeeled 180 179/±36 191/±10 +30 +12 --26 11 Bucaco ‘C” 180 188/+05 198/+03 +30 +10 — 8 6 Bucaco ‘D”2 135 148/±34 203/+38 ±75 +55 +4 9 San Pedro’Emiliano’2 125 137/±13 19 1/±09 +85 +56 —4 2 Alba’2 180 190/±40 199/+20 ±30 +9 —20 Atienza’2 160 1941+19 199/+20 +50 —5 +1 12 Cabe de Penas’2 265 178/+19 1991+19 —55 —18 — Meulin de Chateaupanne ovp 270 2 17/±25 194/+08 —60 —23 —17 7 St. Malo dykes 270 2061±14 1941±07 —60 —12 —7 4 Laval syncline 307 220/—06 190/+02 —97 —30 +8 12 Cambro-Ord. red beds 259 207/+06 191/+06 —49 —16 — 14 Cap Frehel 272 195/±02 190/+0 1 —62 —5 —1 12 Zene Bocaine 290 203/±08 191/ 00 —80 —12 —8 13 Montmartin 270 206/—03 191/ 00 —60 —15 ±3 12 Carteret ‘B’ 272 216/±28 190/+02 —62 —26 —26 14 Rezel ‘E’ 255 203/ 00 191/+01 —45 —12 — 1 7 Crozon ‘B’ 240 217/±29 189/+03 —30 —28 —26 lo Thouars overprint 270 219/±20 192/+04 —60 —27 —16 18 Flamanville granite 270 203/±14 193/±04 —60 —10 —10 15 Pleurive 275 213/±17 19 1/—01 —65 —22 —16 12 Tregastel-Ploumanac’h granite 275 200/+09 194/—04 —65 — 6 +13 7 Jersey dolerites 275 199/±16 198/ 00 —65 — 1 —16 9 N. Britanny 275 212/+10 1 93/±05 —65 —19 —5 Mill Haven sedimenis ‘P-C’2 280 251/±10 188/—OS —70 —63 +15 8 Freshwater ‘C” 280 222/—lI 188/—04 —70 —34 —7 7 Freshwater ‘D’2 280 278/+20 203/±32 —70 —75 12 16 Aigurande Plateau 300 2471±04 197/±09 —90 —50 +5 13 Limeusin 300 249/+07 196/+1- —90 —5- +3 6 Montmarault 300 205/±20 197/±09 —90 —8 —II 17 Morvao 250 258/±02 195/+08 —40 —63 +6 12 Hohes Venn2 220 191/—12 1971—06 —lO +6 +6 6 Ardennes2 230 236/±02 197/—06 —20 —36 —8 II Brabant2 220 2041—07 196/— —lO —8 —2 9 Franconian Forest ‘U 240 203/—02 200/—07 —30 —3 —5 8 Franconian Forest ‘D’ 240 186/+30 217/±28 —30 +31 —2 9 l-Iarz mts ‘C’ 240 183/—04 200/—08 —30 +17 —4 18 Harz mts ‘D’ 240 189/+24 2 18/+24 —30 +29 20 HoIy Cross 290 231/ 00 2 10/—02 —80 —12 —2 6

Devonian and Carboniferous paleoo,agnetic direclions (Dcc. aod Inc.) for l-Iercyoian Europe. S, it íbe reference slrike <210’), 5, it Ihe regional sírike D (‘r) ~dic reference declinalion (inclinalion), D, (1,) Ibe observed nagneíic declination. dD it Ihe radius ol’ Ihe cirele of confidence aboul Ehe observed declination al Ihe 95% probabilily leEd (Beck, 1 980). Magnelic aod síruclural daía correcíed for Ihe opcoing of Ihe Bay of Bitcay.

2 Dala from presumably allochtbooous arcas of Ihe orogen. Hereynian palaeomagnetism of Europe: arguments... 157

0r0o 03

y i 40 y » .5- y

‘5— -.5

—14

y —40 y. •1 y y y

Fig. 3. Declination deviations (Dr~Do plolted as a function of s[rike deviation (S~-S 0) for the European Hercynides.5Openr’ reference(closed)strike;symbols:S~: observeddata from(regional)allochthonousstrike;(autochtDr: refe- honoes)rence paleomagneticparts of the orogen.declination; D 0: obsenved paleomagne[ic declination. Error bars on the declination anomalies calcula[ed according to Beck (1980). Lines «A» and «B» are [he best ¡‘it for Ihe dala sets from allochlhonous aB autochthonous regions of the mountain belt, respectively.

sults in a correlation coefficienL r = 0.551, which again is significant at the 95 % confidence level. Testing the slope of «B» againsL the slope of

«A» gives t = 3.617, which exceeds the reference value and shows, that both data sub-seLs are independent and jusLifles Lhe original decision to evaluaLe both daLa sets individually. FurLhermore a t-Lest for Lhe raiher shallow slope (m = 0.334) of line «B» against zero slope (no oroclinal bending at alí) results in t = 2.477, which is grealer than the reference value

(t = 1.708) underlining Lhe significance of Lhe regression line and therefore being indicative for [he secondary character of the stnuctural curvaLure associaíed with the autochthonous areas of Ihe orogen (see also Fig. 2). Ihus the analysis of palaeomagneLic directional data from the presumably autochthonous areas of the Hercynides suggest, that the bending of Lhe orogen is only in part a secondary feature. Furthermore the rather steep siope of Ihe regression Une for thc «allochthonous» dala set strongly in- 158 PI Bachtadse dicates, that the pronunced curvature of the thin-skinned zones of Lhe Hercynides is a predominantly pnimary feature.

CONCLUSION ‘¡he allochthony of a substantial number of areas within the Hercyni- des reduces the value of the I-lercynian palaeomagnetic data base, for reconstruction of the predeformational palaeogeography of the CenLral Eu- ropean crust. ‘¡he analysis of directional data can nevertheless provide in- sights into the stnuctural evolution of a mountain belt. Combined primany (thin-skinned) and secondary (thick-skinned) structural bending along the European Hercynides, as reponed in this paper, is compatible with several geodynamic models (BachLadse and Van der Voo, 1986). StnucLural data from the Ibero-Armonican arc (MaLte, 1986 and references therein) as well as palaeogeognaphical reconstrucLions (Ziegler, 1986) seem to support a Himalaya— type (Molnan and Tappo- nier, 1978) setting of the geodynamic evoluLion of the Hercynides during the Carboniferous (Lefort and Van der Voo, 1981). Although more palaeomagnetic data from arcas south of the mountain belt is needed in or- den to substantiate this geodynamic model, northwestward impingemenL of the hypothetical Ebro-Pynenean microplaLe (Ziegler, 1983, 1986) oran African promontory (Matte and Ribeiro, 1975) into Europe duning Lhe Carbonifenous provides a viable setting to explain intensive thin-skinned thrusting in the Cantrabnian zone as well as along the norLhern edge of Lhe Hercynides. Parallel [o thrusting in the extennal parts of the orogen, the internal zones were subjected to furLher tightening of an already exisLing fold system. ‘¡hus deep reaching crustal deformation in combination with extensive thin-skinned thrust development caused by continenLal inden- Lation is a rather simplistie but neventheless plausible model for the Can- bonifenous evoluLion of the Hercynides.

Acknowledgements ‘¡he manuscript benefited from valuable suggestions and discussions with Dr. Trond H. ‘¡orsvik.

REFERENCES

BACHTADSE, V., 1984, Palaeomagnetische Untersuchungen im Hercynikum Mitte- leuropas, Ph. D. thesis: Johannes-Gutenberg- Universitael, Mainz, 143 p. BACHTADSE, V., and VAN DER VOO, R., 1986, Paleomagnetic evidence for crustal and thin-skinrted rotations in ffie European Hercynides: Geophys. Res. Lctt.. y. ¡3. 161-164. I-Iercynian palaeomagnetism ofEurope: argun-ients... 159

BACHTADSE, y., VAN DER VOO, R., and HAELBICH, 1. W., 1987, Paleomagne- tism of the Western Cape Foid Belt, South Africa, and its bearing on the Paleo- zoíc apparent polar wander path for Gondwana: Eanth Planet. Sci. Len., y. 84, 487-499. BECK, ME., Jr., 1980, Paleomagnetie record ofpla[e-margin [ectonic processes along the western edge of North America: J. Geophys. Res., y. 85, 7115-7131. BEHR, H.-J., ENGEL, W., FRANKE, W., GIESE, P., and WEBER., K., 1984, The Variscan Belt in Central Europe: main Structures, geodynamic implications, open questions: Tectonophysics, y. 109, 15-40. BONHOMMET, N., COBBOLD, P. R., and PERROUD, H., 1981, Paleomagne[ism and cross-folding in a key area of the Asturian arc (Spain): J. Geophys. Res., y. 86, 1873-1887. BRIDEN, J. C., MORRIS, W. A., and PIPER, J. D. A., 1973, Paleomagnetic studies in ihe Bnitish Caledonides VI, Regional and global implicalloas: Geophys. J. Soc. y. 34, 107-134. CAREY, 5. W., 1958, A tectonie approach [o continental drift, In: Continen[al drif[ a symposium, edited by 5. W. Carey, 177-355, Tasmania University, Hobart. COURTILLOT, y., CHAMBON, P., BRUN, J. P. and ROCHETTE, P., 1986, A mag- neto[ectonic study of the Hercynian Montagne Noire (France): Tectonies, y. 5, 733-751. CREER, K. M., 1968, Paleozoic paleomagnestism: Nature, y. 219, 246-250. DE BOUVIER, M. C., BONHOMMET, N., and VAN DER VOO, It., 1979, Paleo- magnetism and K-4OAr dating of dolerite dykes from the Armonican Massif, Fran- ce (abstr.): FOS, Trans. Am. Geophys. Union, y. 60, 220. DUFF, B. A., 1979, The palaeomagnetism of Cambro-Ordovician red beds, Ihe Erqui spillite series, and the Tregastel-Ploumanac’h grani[e complex, Armorican Massif (France and the Channel lslands), Geophys. J. astr. Soc., y. 59 345-365. DUFF, B. A., 1980, The paleomagnetism of Jersey volcanies and dykes, and [he Lo- wer Paleozoic apparen[ polar wander pa[h for Europe: Geophys. J. R. astr. Soc., y. 60, 355-375. EDEL, J. B., 1984, Etude paléomagné[ique du Nord du Massif Central. In: Program- me Géologie Profonde de la France, Theme 2.: Doc. BRGM, y. 81-2, 45-60. EDEL, J. E., 1 987a, Paleoposition of [he western Europe Hercynides during the Late Carboniferous deduced from paleomagnetie data: consequences for siable «Euro- pe»: Tectonophysics, y. 139, 31-41. EDEL, J. E., 1987b, Paleomagnetie evolu[ion of the Cen[ral Massif (France) during the Carboniferous: Ean[h Planet. Sci. Le[., y. 82, 180-192. EDEL, J. B., and COULON, M., 1984, Late Hercynian remagnetization of Tournaí-

sian series from the Laval syncline - Armorican Massif, France: Earth Píanet. Sci. Lett., y. 68, 343-350. EDEL, J. B., and COULON, M., 1985, A paleomagne[ic profile across [he Ardenne and the Brabant. Evidence for late Variscan clockwise rolations deduced from late-Carboniferous-Permian overprinting (abs[ract): Terra cognita, y. 5, 140. ELDREDGE, 5., BACHTADSE, V., and VAN DER VOO, R., 1985, Paleomagnetism and the orocline hypothesis: Tectonophysics, y. 119, 153-179. FREI, L. 5. and COX, A., 1987, Relative displacemen[ between Eurasia and Non[h America prior to the formation of oceanie crust in [he North Atlan[ic: Tectonophy- síes, y. 142, 111-136. 160 J/~ Bachtadse

FRENCH, R. B., 1976, Lower Paleozoic paleomagnetism of [he North American cra- Ion, Ph. D. thesis: University of Michigan, Ann Arbor, 159 Pp. HAILWOOD, E. A., 1974, Paleomagnetism of the Msissi Norite (Morocco) and Ihe Palaeozoic reconstruction of Gondwanaland: Earíh Planet Sci. Lett., y. 23, 3 76-3 86. HIRT, A. M., LOWRIE, W., JULIVERT, M., and ARBOLEVA, M. L., 1987, Preli- minary paleomagnetic results from the Asturian Arc: a test for the origin of curvature (abstr.): International Workshop Paleomagnetism in Ihe Iberian Peninsula, Platja d’Aro, Costa Brava (Spain), Servei Geológie de Catalunya, y., 10-12. HURLEY, N. F. and VAN DER VOO, R., 1987, Paleomagnetism of Upper Devo- nian Devonian reefal limestones, Canning Basin, western Australia: Geol. Soc. Am. Bulí., y. 98, 138-146. IRVING, E., and IRVING, G. A., 1982, Apparent polar wander paths Carboniferous through Cenozoic and the assembly of Gondwana: Geophysical Surveys, y. 5, 14 1-188. iONES, M., VAN DER VOO, R., and BONHOMMET, N., 1979, Late Devonian to Early Carboniferous palaeomagneíic poles from Ihe Armorican Massit France: Geophys. J. R. astr. Soc., y. 58, 287-308. JOHNSON. G. A. L., 1974, Crustal margins and plate tectonics during thc Carboni- ferous, C. R. lth mt. Congr. Carboniferous Stratigraphy and Geology, y. 3, 26 1-265. KFNT, DV., DíA, O., and SOUGY, J. MA., 1984, Paleomagnetism ofLower- Midd- le Devonian and Upper Proíerozoic-Cambrian(?) rocks from Mejeria (Mauritania, West Africa): AGU Geodynamics Series, y. 12, American Geophysical Union, Washington, D.C., 99-115. KENT, D. Y., and MAY. 5. It., 1987, Polar wander and paleomagnetic reference pole controversies: Rey. Geophys, y. 25, 96 1-970. KHRAMOV, A. N., PETROVA, O. N., and PECHERSKY, D. M., 1981, Paleomag- netism of the Soviet Union, Geodynamics Series, y. 2, American Geophysical Union, Washington, D. C., 177-194. LEFORT, J.-. P., and VAN DER VOO, R., 1981, A kinematic model for [he collision and complete suturing between Gondwanaland and Laurussia in the Carbonife- rous: J. Geol., y. 89, 537-550. LEWANDOWSKI, M., JELENSKA, M., and HALVORSEN, E., 1987, Paleomagne- tism of [he Lower Devonian sandstones from Holy Cross Mountains, Central Po- land: Pan 1: Tectonophysics, y. 139, 21-29. LORENZ, Y., and NICHOLLS, 1. A., 1984, Plate and intraplate processes of Hercy- nian Europe during the Late Paleozoic: Tectonophysics, y. 107, 25-56. LOWRIE, W., and HIRT, A. M., 1986, Paleomagnetism in arcuate mountain beíts, in F.-C. Wezel (cd.), The Origin of Arcs: Elsevier, Amsterdam, 141-158. MAITE, P., 1986, Teclonics and plate tectonics model for the Variscan belt of Eu- rope: Tectonophysics, y. 126, 329-374. MATIE, P. L., and RIBEIRO, A., 1975, Forme e[ orientation de l’ellipsoíde de dé- formation dans la virgation hercynienne de Galice. Rela[ions avec le plissement et hypothése sur la genése de Parc Ibero-Armoricain: C.R. Acad. Sci. Paris, Serie D. y. 280, 2825-2828. MCCABE. C., VAN DER VOO, R.. PEACOR, D. It., SCOTESE. C. It.. and FREE- Hercynian palaeomagnet¡sm of Europe: argumenis... ¡61

MAN, R., 1983, Diagenetic magnetite carnes ancient yet secondary remanence in sorne Paleozoic carbonates: Geology, y. 11, 221-223. MCCLELLAND BROWN, E., 1983, Palaeomagnetic studies of fold developmen[ and propagation in the Pembrokeshire Oíd Red Sands[one: Tectonophysics, y. 98, 13 1-149. MCELHINNY, M. W., ¡973, Paleomagnetism and Plate Tectonics: Academie Press, New York, 358 Pp. MOLNAR, P. and TAPPONIER, P., 1978, Ac[ive [ectonics of Tibet: J. geophys. Res., y. 83, 5361-5375. NOWACZYK, N., and BLEIL, U., 1985, Palaeomagnetic reconnaissance study of Pa- leozoic rocks from [he Hohes Venn area (West Germany): N. Jb. Geol. Palaeont. Abh., y. 171, 351-362. PERROUD, H., 1986, Paleomagnetic evidence for tectonic rotations in the Variscan Mountain BeI[: Tec[onics, y. 5, 205-214. PERROUD, H., and BONHOMMET, N., 1981, Palaeomagnetism of the Ibero-Ar- monican arc and the Hercynian orogeny in Western Europe: Nature, y. 292, 445-448. PERROUD, H., and BONHOMMET, N., 1984, A Devonian paleornagnetic pole for Armorica: Geophys. J. R. as[r. Soc., y. 77, 839-845. PERROUD, H. and VAN DER VOO, R., 1984, Paleomagnetism of the Late Ordo- vícian Thouars massif, Vendée Province, France: J. Geophys. Res., y. 90, 4611-4625. PERROUD, H., BONHOMMET, N., and ROBARDET, M., 1982, Comment on «A paleomagnetic study of Cambrian redbeds from Carteret, Normandy, France» by W. A. Morris: J. Geophys. Res., y. 69, 573-578. PERROUD, H., BONHOMMET, N., and VAN DER VOO, R., 1983, Palaeomagne- tism of the Ordovician dolerites of [he Crozon Peninsula (France): Geophys. J. astr. Soc., y. 72, 307-319. PERROUD, H., ROBARDET, M., VAN DER VOO, R., BONHOMMET, N., and PA- RIS, F., 1984, Revision of the age of magnetiza[ion of the Montmartin red beds, Normandy, France: Geophys. J. R. astr. Soc., y. 80, 541-549. PERROUD, H., VAN DER VOO, R., and BONHOMMET, N., 1984, Paleozoic evolution of the Armorica plate on the basis ofpaleomagnetie data: y. 12, 579-582. PERROUD, H., BONHOMMET, N., and THEBAULT, J. P., 1986(a), Palaeomagne- tism of the Ordovician Moulin de Chateaupanne formation, Vendé, western Fran- ce: Geophys. 1. It. astr. Soc., y. 85, 573-582. PERROUD, H., AUVRAY, B., BONHOMMET, N., MAC, and VAN DER VOO, R., 1986, Palaeomagnetism and K-Ar dating of Lower Carboniferous dolerite dykes from northern Britanny: Geophys. J. R. asir. Soc., y. 87, 143-154. PESONEN, L. J., TORSVIK, T. H., ELMINO, 5.-A., and BYLUND., G., 1988. Crus-

tal evolution of Fennoscandia - Palaeomagnetic constraints: Tectonophysics, in press. RIES, A. C. and SHACKLETON, R. M., Patterns of strain variation in arcuate fold belts: PhiI. Trans. It. Soc. London, Ser. A., y. 283, 281-288. SALMON, E., MONTIGNY, R., EDEL, J. B., PIQUE, A., THUIZAT, R., and WESTPHAL, M., 1986, The Msissi Norite revisited: KIAr da[ing, Pe[rography and Paleomagne[ism: GRL, y. 13, 74t-743. SARADETH, 5., SOFFEL, H., and SCHULT, A., 1987, Paleomagnetic and rockmag- 162 PI Bachtadse

netic s[udies in SW-Egypt and N-Sudan: Contributions to the African polar wander path: Berliner Geowiss. Abh., (A), y. 75. SCHMIDT, P. W., EMBLETON, E. J. J., and PALMER, H. C., 1987, Pre- and post- folding magnetizations from the early Devonian Snowy River Volcanics and Bu- chan Caves Limes[one, Victoria: Geophys. J. R. as[r. Soc., y. 91, 155-177. SCHWARTZ, 5. Y., and VAN DER VOO, R., 1983, Paleomagnetic evolution of the orocline hypothesis in [he central and southern Appalachians: Geophys. Res. Lett., y. 10, 505-508. SOPER, N. J., and HUTTON, D. H. W., 1984, Late Caledonian sinistral displace- ments in Britain: implications for a three plate collision model: Tectonics, y. 3, 78 1-794. VAN DER VOO, It., 1967, The rotation of Spain: paleomagne[ic evidence from [he Spanish Meseta: Palaeogeogr., Palaeoclimatol., Palaeoecol., y. 3, 417-432. VAN DER VOO, It., 1969, Paleomagnetic evidence for the rota[ion ofthe Iberian Pc- ninsula: Tectonophysics, y. 55, 5-56. VAN DER VOO, It., 1979, Paleozoic assembly of Pangea: A new plate tectonic model for the Taconic, Caledonian and Hercynian orogenies (abstract): Eos Trans. AGU, y. 60, 241. VAN DER VOO, It., 1982, Pre Mesozoic Paleomagnetism and Plate Tectonics: Ann. Rey. Earth Planet. Sci., y. 10, 191-220. VAN DER VOO, It., 1983, Paleomagnetic constrainis on the assembly ofthe Oíd Red Continen[: Tectonophysics, y. 91, 271-283. VAN DER VOO, It., and CHANNELL, J. E. T., 1980, Paleomagnetism in orogenic belts: Rey. Geophys. Space Phys., y. 18, 455-481. VAN DER VOO, It., and KLOOTWIJK, C. T., 1972, Paleomagnetic reconnaissance study of the Flamanville grani[e with special reference to thc anisotropy of its sus- ceptibility: Geol. Mijnbouw, y. 51, 609-617. ZIEGLER, P. A., ¡983, Caledonian and Hercynian crustal consolidation of Western and Central Europe - A working hypothesis: Geol. Mijnbouw, y. 63, 93-108. ZIEGLER, P. A., 1986, A Geodynamic model for [he Paleozoic crustal consolidation of Wes[ern and Central Europe: Tec[onophysics, y. 126, 303-328.

Recé red 14 Jan. /988. A ccepted 30 April ¡988.