evidence for non-coaxiality of ferrimagnetic and paramagnetic fabrics, developed during magma flow...

Tectonophysics xxx (2014) xxx–xxx

TECTO-126271; No of Pages 10

Contents lists available at ScienceDirect

Tectonophysics

j ourna l homepage: www.e lsev ie r .com/ locate / tecto

Evidence for non-coaxiality of ferrimagnetic and paramagnetic fabrics,developed during magma flow and cooling in a thick mafic dyke

P.F. Silva a,⁎, F.O. Marques b, M. Machek c, B. Henry d, A.M. Hirt e, Z. Roxerová c, P. Madureira f, S. Vratislav g

a ISEL and IDL (Univ. Lisbon), Lisbon, Portugalb University of Lisbon, Lisbon, Portugalc Institute of Geophysics AS CR, v.v.i, Boční II/1401, 14131, Prague 4, Czech Republicd Paléomagnétisme, Institut de Physique du Globe de Paris, Sorbonne Paris Cité, Université Paris Diderot, UMR 7154 CNRS, 94107 Saint-Maur cedex, Francee Laboratory of Natural Magnetism, Institute of Geophysics, ETH-Zürich, CH-8092 Zurich, Switzerlandf EMEPC, 2770-047 Paço d'Arcos and University of Évora, D. Geociências and C. Geofísica de Évora, 7000-671 Évora, Portugalg Laboratory of Neutron Diffraction, Department of Solid State Engineering, Faculty of Nuclear Sciences and Physical Engineering, CTU in Prague, Czech Republic

⁎ Corresponding author. Tel.: +351 918986811, +351E-mail address: [email protected] (P.F. Silva).

http://dx.doi.org/10.1016/j.tecto.2014.04.0170040-1951/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Silva, P.F., et al., Evand cooling in a thick mafic dy..., Tectonoph

a b s t r a c t
a r t i c l e i n f o
Article history:Received 3 October 2013Received in revised form 26 February 2014Accepted 8 April 2014Available online xxxx

Keywords:ParamagneticFerrimagneticFabricsMicrostructuresDyke emplacementStress field

A detailed analysis of fabrics of the chilled margin of a thick dolerite dyke (Foum Zguid dyke, SouthernMorocco)was performed in order to better understand the development of sub-fabrics during dyke emplacementand cooling. AMS data were complemented with measurements of paramagnetic and ferrimagnetic fabrics(measured with high field torque magnetometer), neutron texture and microstructural analyses. The ferrimag-netic and AMS fabrics are similar, indicating that the ferrimagnetic minerals dominate the AMS signal. The para-magnetic fabric is different from the previous ones. Based on the crystallization timing of the differentmineralogical phases, the paramagnetic fabric appears related to the upward flow, while the ferrimagnetic fabricrather reflects the late-stage of dyke emplacement and cooling stresses.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

The high sensitivity and rapid measurements made anisotropy ofmagnetic susceptibility (AMS) one of the most applied and powerfultools to assess the petrofabric, even for low anisotropy rocks (see,Borradaile and Henry, 1997; Borradaile and Jackson, 2010; Hrouda,1982; Tarling and Hrouda, 1993). Despite these advantages, it wassoon recognized that AMS interpretation is not always straightforward.The superposition of magnetic fabrics, related to magnetic carriers withnormal and inverse fabric, orwith distinct preferred orientations and/orshapes, is one factor that can result in awhole-rockAMS fabric that doesnot reflect the true preferred orientation of minerals (Borradaile andHenry, 1997; Chadima et al., 2009; Fanjat et al., 2012; Ferré, 2002; Hirtand Almqvist, 2012; Lamali et al., 2013; Potter and Stephenson, 1988;Rochette, 1988; Rochette et al., 1999; Silva et al., 2008; Tarling andHrouda, 1993).

To decompose composite magnetic fabrics complementary experi-mental methods (e.g., anisotropy of magnetic remanence and high-fieldtorque magnetometry) and analytical and computational solutions have

217500000x28328.

idence for non-coaxiality of feysics (2014), http://dx.doi.org

been developed to separate sub-fabrics (e.g., Banerjee and Stacey, 1967;Callot and Guichet, 2003; Ferré et al., 2004; Jelinek, 1996; Kratinováet al., 2006, 2010; Henry, 1983, 1997; Henry and Daly, 1983; Hroudaand Jelinek, 1990; Martín-Hernandez and Garcia-Hernandez, 2010;Martín-Hernandez and Hirt, 2001, 2004; McCabe et al., 1985; Moreiraet al., 1999; Roperch and Taylor, 1986; Schmidt et al., 2007; Stephenson,1980; Stephenson et al., 1986).

More recently, rockmagnetic fabrics studies start to be complementedbyquantitativemicrostructural and crystallographic preferred orientationanalyses in order to better understand the rheological behaviour of rocks(e.g., Bascou et al., 2005; Boiron et al., 2013; Cifelli et al., 2009; Macheket al., this volume; Závada et al., 2007).

The aim of this paper is a better understanding of intrusive process-es. This requires recognition of how a dyke's petrofabric records differ-ent strain regimes, due to the evolution of stress fields and thermalgradients duringmagma emplacement and cooling. Different rockmin-erals crystallize at different times and likely under variable stress fields.Recognizing petrofabrics of individual minerals that are related to dif-ferent episodes of mineral crystallization during magma emplacementis therefore a key approach. To this aim, we employ a combination ofmicrostructural and textural approaches (neutron diffraction—ND, andimage analyses to obtain crystallographic preferred orientation—CPO)with analyses of magnetic fabrics in low and high fields (AMS for

rrimagnetic and paramagnetic fabrics, developed during magma flow/10.1016/j.tecto.2014.04.017

http://dx.doi.org/10.1016/j.tecto.2014.04.017

mailto:[email protected]


http://www.sciencedirect.com/science/journal/00401951


2 P.F. Silva et al. / Tectonophysics xxx (2014) xxx–xxx

whole-rock, anisotropy of ferrimagnetic susceptibility—AFMS andanisotropy of paramagnetic susceptibility APMS).

The thick Foum Zguid doleritic dyke—FZD (e.g., Leblanc, 1974) insouthern Morocco was selected for this case study (Fig. 1). The dykeemplacement at great depthwas associatedwith a forceful magma injec-tion that affected their sedimentary host-rocks both mechanically andthermodynamically: folding andflattening of the host sedimentary layers,mineralogical and textural transformations due to Fe-metasomatism andthermally induced recrystallizations (Silva et al., 2010). Previous FZDstudies (Silva et al., 2004, 2006a,b, 2010) were based on detailed low-field AMS, partial anisotropy of anhysteretic remanence (pAARM), paleo-magnetism and rock magnetism. The comparison between AMS andpAARM fabrics has often revealed the composite character of the AMSfabric, related to superimposition of normal and inverse fabrics. Thepresence of an inverse fabric is common in the inner domains of thedyke due to single-domain particles that result from lamella exsolutionprocesses of pristine Ti–iron oxides. Near the margins, where Ti-iron ox-ides are euhedral grains without exsolution textures, AMS and pAARMfabrics are coaxial with a normal fabric. The magnetic foliation mostlystrikes parallel to the vertical dyke plane, but with dip variations along

FZ7

−10˚ −6˚ −2˚26˚

30˚

34˚

38˚

42˚

Study area

Morocco

Algeria

SpainPortugal

AtlanticOcean

a

Dyke

N220o

S

K1

K3 K2

K1

K3K3 K3K1

K1

K2K2K2

Flow

?

Chilled margin

Core

c

40 m 1 m

Fig. 1. (a) Satellite images (available from Google) with geographical location of the study statFoum Zguid dyke (topographic high) with indication of its thickness and contacts with sedimeobserved for dyke and sedimentary host rocks along the distance to the contact (adapted fromdiscontinuities observed during field work.

Please cite this article as: Silva, P.F., et al., Evidence for non-coaxiality of feand cooling in a thick mafic dy..., Tectonophysics (2014), http://dx.doi.org

cross-sectionswithin the dyke: dip towards the inner part in the chilledmargin and towards the outside in inner domains. Magma flowinterpreted from the imbrication between magnetic foliation and dykeplane (Geoffroy et al., 2002) should be downward at the chilledmarginsand upward in the inner domain. However, downward flowwould con-tradict the observed upward deflection of bedding in the host sedimen-tary rocks (cf. Silva et al., 2010). A similar contradictory AMS patternbetween the margin and the central part has been also observed inthe great dolerite Messejana–Plasencia dyke (Silva et al., 2008).

2. Sampling and methods

A total of 38 igneous samples were selected from the samplingstation FZ7 (e.g., Silva et al., 2010). These samples were quasi-continuously collected along twodyke cross-sections from theNWmar-gin of the dyke: i) cross-section A1, between the NWmargin and 20 mfrom it, and ii) cross-section A2 is spaced 10 m from A1, and limited tothe first meter of the contact with host sedimentary rocks. The aim ofthis sampling strategy is to better evaluate the evolution of the fabricsalong those profiles located near the margins.

b

Foum Zguid dyke

host-rockshost-rocks 100 m

NWSE

K3

K1 K2K1

K3 K2

K1

K2K3

Sedimentary host-rocks

50 m

ion (white circle). White dashed line represents the main igneous body; (b) photo of thentary host rocks (white dashed line); (c) nonscaled sketch of the AMS ellipsoids patternsSilva et al., 2010); white dashed lines at the sedimentary host rocks exemplify the planar



3P.F. Silva et al. / Tectonophysics xxx (2014) xxx–xxx

To complement previous AMS results, APMS and AFMS weredetermined using high-field torsion magnetometery, MT (ETH-Zürich,Switzerland), and following the experimental protocol defined byMartín-Hernandez and Hirt (2001, 2004). MT results from the energy ofmagnetization (E) observed for an angular difference (θ) between thesample induced magnetization and the applied field (B): MT ¼ −dE

�dθ .

It was mostly measured in incremental fields between 500 and1300 mT, with steps of 100mT. For each applied field, MTwas measuredduring sample full rotations with angular steps of 15–20° about the threeperpendicular sample axes.

Before any attempt to perform magnetic fabric interpretation, it iscrucial to determine how the different types of minerals contribute, interms of intensity and angular symmetry, to the range of fields used intheMTmeasurements. In high fields (N500mT), the ferrimagnetic com-ponent is saturated, which means that its contribution to the torque isconstant. Antiferromagnetic and paramagnetic fractions, however, arelinearly proportional to the applied field (B and B2, respectively). For amixture of ferrimagnetic and paramagnetic components, the totaltorque, above the saturation field of ferrimagnetic carriers, will belinearly related to B2, obeying a linear equation,

MT ¼ mferri þ npara � B2

Paramagnetic and ferrimagnetic contributions were determinedthrough the slope (npara) and the intersection with the ordinate axis(mferri), respectively (cf. Martín-Hernandez and Hirt, 2001).

In order to compare the shape and intensities of anisotropy of AMS,APMS and AFMS ellipsoids, the used shape parameter U (Jelinek, 1981)is defined as:

U ¼ 2k2−k1−k3k1−k3

and the intensity parameter k′ (Jelinek, 1984), defined as:

k0 ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffik1−kð Þ2 þ k2−kð Þ2 þ k3−kð Þ2

3

s

Both these parameters can be obtained from the eigenvalues of thedeviatoric paramagnetic and ferrimagnetic tensors, as well as from themagnitudes of the principal axes of the AMS tensor (for more details,see Hirt and Almqvist, 2012). Analyses of the rock magnetic fabricswere complemented by petrofabric microstructure and texture evalua-tions on three samples (21b and 26 from site FZ7-NW/A1, and 32 fromsite FZ7-NW/A2). Neutron diffraction (ND) measurements were per-formed on large samples (cylinders of ~20 mm diameter by ~10 mmof length) in order to assess the volumetric distribution of CPOs (moredetails in Annex A). For crystal orientation, poles to (010) prism planesand [001] directions for plagioclase, and poles to (100) prismplanes and[001] directions for orthopyroxene represent the largest plane and thelongest axis of euhedral plagioclase and orthopyroxene crystals,respectively.

The crystal alignment domains (CAD)were evaluated along sectionsthat correspond to vertical planes perpendicular to the dykewall. It wasmanually traced as lines in the ArcGis environment from SEM images atleast from the 10 × 20mm area of the thin section. The orientation andlength of traced CADswere obtained using the PolyLXMatlab™ toolboxfor quantitative microstructure analysis (Lexa et al., 2005, http://petrol.natur.cuni.cz/~ondro/polylx:home).

3. APMS and AFMS fabrics

The analyses of magnetic torque per volume as a function of angleshow several shapes that result from the superposition of componentswith distinct angular symmetry, i.e., with 2θ and 4θ symmetries.


The 2θ component clearly dominates the magnetic torque for sam-ples located near the margin (less than 1 m from the contact; Fig. 2a,b). It is also observed that the shape of the curves does not cross exactlyalong the axis with null torque value. Towards the middle of the dyke(between1 and 20m from the contactwith thehost-rocks), the 4θ com-ponent increases, thus becoming clearly visible (Fig. 2c). The presenceof this component is also accompanied by asymmetric peaks and anasymmetry of magnetic torque curves plotted as a function of angle rel-ative to the axis with null torque value. This leads to a visible offset ofthe crossing points between the curves achieved for each plane. Thedominant and symmetrical 2θ angular symmetry most probably arisesfrom magnetite shape anisotropy (Martín-Hernández and Hirt, 2001),because the amplitude of the torque as a function of field is close to sat-uration. The 4θ signal with asymmetric peaks suggests the presence ofhematite (Martín-Hernández and Hirt, 2004) or of some ferrimagneticsingle crystals with biaxial anisotropy (Martín-Hernández et al., 2006).

The evolutions of the total magnetic torque for fields above 500 mTare also distinct for samples near or further away from the contact withthe host-rocks:

- For samples located nearest the margin, where the 4θ component isminimal, the total magnetic torque above 500 mT is linearly relatedto B2 (Fig. 2a, b) indicating a contribution from paramagnetic min-erals (e.g., Banerjee and Stacey, 1967; Hrouda and Jelinek, 1990).The MT, however, is dominated by ferrimagnetic carriers (e.g., mag-netite; see Table 1). Depending on the samples, the paramagneticcontribution varies from 6 to 26% of the deviatoric susceptibility.One sample with a weak paramagnetic contribution has poorly-defined paramagnetic tensor, and was thus not used in furtheranalysis. The remaining samples have more than 8% paramagneticcontribution, mostly defining triaxial APMS ellipsoids for bothsample sites (see Table 1 and Fig. 3a).

- For distances greater than 1 m from the contact with the host rocks,where the 4θ component is present, there is a linear dependency be-tween MT and B values (Fig. 2c). This suggests the presence of ahigh-coercivity phase, like hematite. The presence of hematite orany other high-coercivity mineral was not demonstrated in anyrock magnetic analyses conducted by Silva et al. (2006a, 2006b,2010). The thermomagnetic analyses showed a maximum Curietemperature of 570 °C for all dyke samples, and the hysteresis curves(Fig. 6c and d from Silva et al., 2010) have closed loops after para-magnetic correction. In addition, paleomagnetic analyses conductedby Silva et al. (2006a, 2006b) show 530 °C as maximum unblockingtemperature. Therefore, none of these results suggest the presenceof hematite. Supplementary experiments of acquisition of isother-mal remanent magnetization (IRM analyses) were carried out onseven samples collected between the margin and 20 m from it. Allsamples show similar IRM curves, saturating for applied fieldsbelow ~300 mT (Fig. 4). Therefore, the presence of hematite or anyother high coercivity magnetic phase (e.g., pyrrothite) is not sup-ported by any of the rock magnetic analyses conducted for the FZDsamples.

In order to understand the origin of the 4θ signal, the protocol ofMartín-Hernández and Hirt (2004) was applied to separate the para-magnetic, ferrimagnetic and hematite fabrics (Fig. 5a and c). Resultsfrom such analyses indicate that:

1. Samples nearest the margin (distances b1 m): i) the paramagneticcontribution is present, with a minor influence of the 4θ componentfor one of the planes; and ii) the ferrimagnetic component isnegligible.

2. For distances greater than 1 m: i) the paramagnetic contributionstarts to be significantly affected by a 4θ component, and ii) theferrimagnetic component is negligible.

These results indicate a very significant contribution of a carrier ofthe 4θ component for the magnetic torque in samples further away


http://petrol.natur.cuni.cz/~ondro/polylx:home

http://petrol.natur.cuni.cz/~ondro/polylx:home


0 60 120 180 240 300 360-10

-8

-6

-4

-2

0

2

4

6

8

10

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.10

2

4

6

8

10

12

Angle (°) Applied Field - B2 (T2)

0 40 80 120 160 200 240 280 320 360-12

-10

-8

-6

-4

-2

0

2

4

6

8

10

12

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.10

2

4

6

8

10

12

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.10

2

4

6

8

10

12

Am

plitu

de (

Jm-3

)T

orqu

e (J

m-3

)T

orqu

e (J

m-3

)

2

Axis 1 Axis 2 Axis 3

a

b

c

-12

-10

-8

-6

-4

-2

0

2

4

6

8

10

12

0 40 80 120 160 200 240 280 320 360

Tor

que

(Jm

-3)

Tor

que

(Jm

-3)

Tor

que

(Jm

-3)

Angle (°) Applied Field - B2 (T2)

Angle (°) Applied field B(T)

Fig. 2.Magnetic torque amplitude per unit of volume as a function of the sample rotation for distinct applied fields (left) andmaximumamplitude of the torque (after applying fast Fouriertransform) as a function of the appliedmagneticfield B and/orB2 (right). a) Example of a sample controlled by ferrimagnetic 2θ-termwith a paramagnetic contribution between 8 and 26%of the total torque (selected for this study); b) the same as in a) but with minor contribution of the paramagnetic fraction (b8%; sample rejected); c) sample that shows superposition oftwo distinct components with 2θ and 6θ symmetries (sample rejected).


Please cite this article as: Silva, P.F., et al., Evidence for non-coaxiality of ferrimagnetic and paramagnetic fabrics, developed during magma flowand cooling in a thick mafic dy..., Tectonophysics (2014), http://dx.doi.org/10.1016/j.tecto.2014.04.017


Table 1Resume of themagnetic fabric properties of the samples located near the NWmargin of the FoumZguid dyke (station FZ7). Directions (declination/inclination) of the principal axes of theparamagnetic (P1 ≥ P2 ≥ P3), ferrimagnetic (F1 ≥ F2 ≥ F3) and AMS (K1 ≥ K2 ≥ K3) ellipsoids. U and k′—parameters defining the shape and intensity of anisotropy for the three kindsof ellipsoids (Jelinek, 1981, 1984); %—percentage of the paramagnetic and ferrimagnetic components to the total magnetic torque.

FZ7 Sample Paramagnetic Ferrimagnetic AMS

P1 (°) P2 (°) P3 (°) k′ (10−6) U % F1 (°) F2 (°) F3 (°) k′ (10−6) U % K1 (°) K2 (°) K3 (°) k′ (10−6) U

NW/A1 FZ7-20 338/69 237/4 146/21 0.74 −0.28 12 106/64 199/2 290/26 5.56 0.24 88 134/64 26/8 293/24 95.2 0.50FZ7-21A 42/2 304/74 133/16 0.31 0.40 11 133/69 24/7 291/20 3.58 0.35 89 161/42 44/27 292/36 53.6 0.55FZ7-21B 350/55 210/28 110/19 0.37 0.22 9 144/64 32/10 297/24 3.58 0.24 91 137/53 30/13 291/34 62.7 0.56FZ7-24B 25/43 241/40 134/19 0.29 −0.55 26 337/67 82/6 175/22 3.10 −0.41 74 115/56 225/13 323/31 72.8 −0.34FZ7-26 216/3 319/74 126/15 0.53 0.52 13 56/17 175/58 317/27 3.65 0.84 87 79/41 199/30 312/35 79.2 0.93FZ7-27A 35/69 200/20 292/5 0.41 0.32 10 111/49 207/6 302/41 3.89 −0.87 90 109/46 215/15 318/41 111.1 −0.22

Mean directions 22/46 229/42 127/15 102/66 211/2 303/25 122/54 213/2 304/34NW/A2 FZ7-29A 320/59 210/12 113/29 2.46 0.57 26 197/65 34/25 301/7 7.01 0.39 74 194/69 32/20 300/6 128.8 0.51

FZ7-29B 299/82 48/3 138/8 1.31 0.33 17 190/70 41/17 308/9 6.44 0.18 83 180/76 37/11 305/8 119.0 0.45FZ7-31A 226/55 14/31 113/15 0.57 −0.05 8 189/48 14/42 282/2 5.78 0.24 92 79/87 201/2 291/3 141.3 0.41FZ7-31B 220/42 26/47 124/7 1.53 −0.09 22 182/38 21/50 279/9 5.49 0.07 78 108/84 198/0 288/6 118.3 0.41FZ7-32A 251/38 38/47 148/17 1.58 0.01 16 195/48 11/42 103/2 6.62 0.39 84 52/77 198/10 289/7 195.0 0.50FZ7-34 234/36 26/51 134/14 2.22 0.22 16 198/18 55/68 292/12 1.15 0.56 84 205/59 28/31 298/2 405.6 0.80FZ7-35B 250/53 36/32 137/16 1.53 0.07 19 192/52 45/34 304/16 9.40 0.63 81 91/72 217/11 310/14 151.5 0.84

Mean directions 285/55 31/31 130/15 192/49 30/40 298/8 159/82 27/5 297/7


from the contactwith the host rocks, which should have a coercivity notsaturated in the maximum field, and a negligible effect of the ferrimag-netic magnetite. This is contradictory to the data obtained from rockmagnetic experiments and paleomagnetic analyses. The amplitude ofthe torque signal is also saturated, which allows the separation of theferrimagnetic fabrics using the method of Martín-Hernández and Hirt(2001) for all of the analysed samples (Fig. 5b and d). Maybe, an expla-nation for the angular symmetry has to be searched in some ferrimag-netic single crystals with biaxial anisotropy in the measuring plane ordue to the presence of a superposition of two fabrics normal to one an-other (Flanders and Schuele, 1964; Martín-Hernández et al., 2006;Porath andRaleigh, 1967). However, thepresence of ferrimagnetic com-ponents does not explain the linear field-dependence of the magnetictorque for high-fields as observed for samples further away from thecontact. This is a subject that requires further research but is out of thescope of the present study.

Because we cannot reconcile these observations, we omit the sam-ples that have a strong high-order contribution to the torque signal.Therefore, only 13 samples located nearest to the contact could beused for further analysis. These samples are located in the dyke the do-main where magma flow shows an unexpected downward magmaflow, as inferred from the AMS results. The 13 samples belong to twosampling sites, FZ7-NW/A1 and FZ7-NW/A2. High-field torsion magne-tometer experiments reveal coherent but non-coaxial APMS and AFMS(see Table 1 and Fig. 3a). The AFMS fabric is very similar to the observedAMS fabric, which has a magnetic foliation that strikes parallel to thedyke azimuth and dips towards the middle part of the dyke, but moresteeply in FZ7-NW/A2. The AFMS lineation at both sites stronglyplunges towards the middle part of the dyke, but less than observedfor AMS. APMS, which was significant at both sites, defines a very simi-larmagnetic foliation striking parallel to the dyke trend, but, contrary tothe AMFS and AMS fabrics, dips towards the dyke wall.

The plot of U versus k′ (Jelinek, 1984) is presented in Fig. 3b. Thethree types of rock magnetic shape fabrics display similar results, withpredominance of oblate ellipsoids.

4. Microstructure

SEM images of the three samples reveal an igneous texture charac-terized by a fine-grained subophitic texture of subhedral plagioclase,subhedral to anhedral clino and orthopyroxene, and rare phenocrystsof olivine. The texture is seemingly isotropic,with locally developed dis-continuous crystal alignment domains (CAD) of plagioclase and pyrox-ene in the rock groundmass. Such domains, where crystals are oriented


at a low angle to the domain margin (Schelley, 1985), have beeninterpreted as shear zones (Smith, 1998; Smith et al., 1993a,b, 1994),and usually occur in conjugate sets. The orientation of domains ofaligned crystals is sporadically followed also by individualmatrix grains.The rockmicrostructure is also characterized by the occurrence of inter-granular microcracks, commonly filled by barite.

CAD data can help explaining the twomain orientations achieved byrock magnetic fabric experiments (Fig. 6). The two main preferred ori-entations of CAD are oblique to the vertical dyke margin, one plungingtowards the dyke wall (to NW) and the other towards the middle partof the dyke (to SE). In one sample (FZ7-21) two other directions arealso observed, parallel and perpendicular to the dyke, respectively.The plunging towards the dyke wall is in close agreement with the atti-tude of the APMS foliation, but, despite some similarities, that is not thecase with the dip toward the dyke middle part of the dyke and AFMSand AMS foliations.

Themedian lengthof crystal alignment domainswas estimated to be1.4 mm in samples FZ7-21b and FZ7-26, and to be 2.1 mm in sampleFZ7-32. The estimated length of crystal alignment domains is howeverinfluenced by the subjectivity of the manual tracing of the zones.

5. Crystallographic preferred orientation (CPO)—neutron texture

In all three samples, the different CPO diagrams are characterized bythe presence of several distinguishable concentrations (Fig. 7). A goodcorrelation exists between the poles to (010) prism planes for plagio-clase and the [001] directions for orthopyroxene. An inverse correlationappears between poles to (100) prism planes and [001] directions fororthopyroxene. No clear relationship could be observed between dia-grams for the three studied samples, except a dominant plane thatsteeply dips towards NNE to NE, which is highlighted by the isodensitylines.

6. Discussion

6.1. Data for the paramagnetic minerals

The interstitial rock matrix mainly comprises magnetite, ilmen-ite, hornblende and biotite. The main minerals are plagioclase,forming either (micro-) phenocrysts or developing a tightly packedgroundmass network, clinopyroxene (diopside-hedenbergite, some-times as microphenocrysts) and pigeonite that may include formerlydeveloped olivine grains.



90270

180

N

90270

180

N

FZ7-NW/A1

FZ7-NW/A2

270

180

90

N

270

180

90

N

270

180

90

N

AMS Ferrimagnetic Paramagnetic

90270

180

N

a

b

AMS

Ferrimagnetic

FZ7-NW/A2FZ7-NW/A1

Paramagnetic

1E-7 1E-6 1E-5 1E-4 1E-3

-1.00

-0.75

-0.50

-0.25

0.00

0.25

0.50

0.75

1.00

Sha

pe p

aram

eter

- U

k' (SI)

Fig. 3. a) Stereographic projections (equal area, lower hemisphere) in geographical coordinates presenting the AMS, AFMS and APMS data: Squares, triangles and circles correspond tomaximum, intermediate and minimum principal axes, respectively. Full black symbols correspond to the mean directions with their 95% confidence ellipse. Black NE–SW diameter inall plots represents the dyke. b) Diagram showing shape parameter U versus k’ for the three different magnetic fabrics.


Plagioclase and pyroxene, the main paramagnetic minerals of thisdolerite dyke, are the first crystals to nucleate (after olivine), athigher-temperatures, typically around 800–1000 °C (Lutgens et al.,2011). The main microstructural feature related to the flow in the mar-ginal part of the FZD dyke corresponds to the oriented crystal magneticalignment—CAD, which can be interpreted as microscale shear zones(Smith, 1998; Smith et al., 1993a,b, 1994). The development of suchshear zones indicates dilatancy textures resulting from shear thickeningrheology of a highly crystallized magma (Park and Means, 1996; Smith,1997, 2002), and that the crystalline mush was actively deformedduring – or at the final stages – of magma emplacement.

The CPO data show several distinguishable concentrations of polesto 010 planes in plagioclase and (100) planes in orthopyroxene. Theseconcentrations are not clearly correlated with the magnetic fabric.This can be the result of development of CAD and of “quasiviscous”flow/deformation of magma ground mass facilitated by slip along the(010) planes of plagioclase and the (100) planes of pyroxene (Závada


et al., 2007), as evidenced by deformation experiments with fine micaflakes (Tullis, 1976 ).

The CPO diagrams show several clusters and are different from onesample to the other. That suggests aweak corresponding fabric. Howev-er, the CAD analysis clearly indicates an anisotropic distribution. Thesame observation can be made comparing these different data withthe APMS fabric. The latter does not clearly correlate with CPO data,but is consistent with the CAD results. This difference reflects the pres-ence of different populations of paramagnetic minerals, probably relat-ed to different crystallization times. The paramagnetic data should betherefore composite, with different dominance according to the usedmethod. Each CPO diagram represents the contribution of a mineral, in-dependently of crystal size. There is, however, a good agreement be-tween orthopyroxene and plagioclase data. The CAD is related todomains with relatively homogeneous orientation of crystals. TheAPMS is related to pyroxene, and perhaps partially to the low amountof biotite. The susceptibility of plagioclase is in fact veryweak compared



10 100 10000.00

0.25

0.50

0.75

1.00IR

M/IR

MM

AX

Applied field (mT)

Fig. 4. Acquisition of Isothermal Remanent Magnetization – IRM for seven samples until amaximum applied field of 2.75 T. Samples were collected between 0.3 and 20 m from thecontact with the host-rocks.


to that of pyroxene and biotite (Borradaile and Werner, 1994;Borradaile et al., 1987; Rochette et al., 1992). The agreement betweenAPMS and CAD therefore underlines that this fabric is mainly relatedto these “homogeneous” domains. The relative scattering on CPO dia-grams should be therefore mainly due to other grains.

0 180 360−20

0

20

0 180 360−20

0

20

0 180 360−20

0

20

0 180 360−20

0

20

0 180 360−1

0

1

0 180 360−1

0

1

y - z plane x - z plane

0 180 360

−20

0

20y - z plane

0 180 360

−20

0

20

0 180 360−20

0

20

0 180 360−20

0

20

0 180 360−10

0

10

0 180 360−10

0

10

x - z plane

a

c

Fig. 5. Samples from themargin of the dyke – a) each line of graphics shows the angular symmetbottom, the expected paramagnetic, “hematite” and ferrimagnetic components, respectively. b)as a and b, respectively, but for the inner domains of the dyke.


6.2. APMS and AFMS

The magnetic foliations of paramagnetic and ferrimagnetic min-erals dip towards the dyke wall or towards the middle part of thedyke, respectively. The magnetic lineations show a slightly differentorientation, which may indicate that the APMS and AFMS were notacquired at the same time. On the other hand, the similarity betweenAMS and AFMS indicates that ferrimagnetic minerals dominate theAMS fabric.

6.3. Magma flow and dyke emplacement

Magnetite is essentially disseminated in the rock matrix, accordingto its late nucleation and crystallization later than plagioclase and py-roxene. Therefore APMS was acquired before AFMS, both with orienta-tions related to that of the dyke but distinct between them. If magmaflow is interpreted based on the imbrication betweenmagnetic foliationand dyke margin, an upward magma flow can be inferred from APMS.That agreeswith the AMS in inner dyke domains. At the chilledmargins,AMS and AFMS however indicate a seemingly downward flow, an AMSpattern also observed at the chilled margins of the great doleriteMessejana–Plasencia dyke (Silva et al., 2008). As pointed out in a previ-ous work (Silva et al., 2010), the presence of a downward flow seemsunlikely; in fact, field observations show upward deflection of the bed-ding in the host rocks, consistent with upward push and flow. Besides,downward flow is prevented by magma pressure that drives theupward forceful injection.

0 180 360−20

0

20

0 180 360−20

0

20

0 180 360−1

0

1

x - y plane

0 0.2 0.4 0.6 0.8 1−4

−3

−2

−1

0

1

2

3

4

B [T]

0 180 360

−20

0

20

0 180 360−20

0

20

0 180 360−10

0

10

x - y plane

−8

−6

−4

−2

0

2

4

6

8

0 0.2 0.4 0.6 0.8 1B [T]

Tor

que

[Jm

−3 ]

Tor

que

[Jm

−3 ]

sin(θ)sin(2θ)sin(4θ)cos(θ)cos(2θ)cos(4θ)

b

d

ry for three distinct planes. Thefirst, second and third lines of figures, represent from top toFourier coefficients of torque curve as a function of applied field; Figures c and d, the same



Magnetic foliationsAMS Paramagnetic Ferrimagnetic

(c) Fz7-32

0.0020.0040.0060.0080.01

5mm

Up

Down

NW SE

0.002 0.004 0.006 0.008

(b) Fz7-26

5mm

Up

Down

NW SE

(a) Fz7-21b

0.002 0.004 0.006 0.008

5mm

Up

Down

NW SE

NW contact

14 cm

40 cm

21 cm

Fig. 6. Crystal alignment domains: On a vertical plane perpendicular to the dyke trend, amap of traced crystal alignment domains and corresponding rose diagrams comparedwith magnetic fabrics orientation of samples FZ7-21b (a), FZ7-26 (b) and FZ7-32 (c).


The microstructural analysis indicates that the two observed para-magnetic sub-fabrics most likely developed during magma flow. Thisis not necessarily the case of the ferrimagnetic fabric (as determinedby AMS and AFMS fabrics), whichmainly results from late stage crystal-lization. The fast cooling close to the margin of the dyke leads to an in-crease in the viscosity of magma, which at later stages of coolingshows a transition to brittle behaviour. With hardening of the rock atthe chilled margin, and under a simple shear regime promoted bymagma flow in inner dyke domains, the mixture of crystals andmelt is able to produce stable planes in the matrix that dip towardsthe shear sense (e.g., Batchelor, 1967; Holtzman et al., 2005),which is in close agreement with the attitude observed for the ferri-magnetic phase. It is interesting to notice that the planar discontinu-ities, which can be observed in the host-rocks at the margin of thedyke, also dip symmetrically towards the dyke (Silva et al., 2010),i.e., with an attitude similar to that of the AMS and AFMS magneticfoliations determined in the dyke chilled margins. This similaritysuggests that chilled margins could have reacted as host rocks aftertheir fast cooling and hardening, while magma was still flowing inthe inner domains of the dyke. Therefore, we should be aware thatthe ferrimagnetic phase is not recording the magma flow itself, buta later phase of near solidus deformation related to dyke emplacementand cooling.


7. Conclusions

The use of combined microstructural and magnetic approachesallows deciphering the mechanical processes at chilled margins of thedeep-seated and thick mafic Foum Zguid dyke, related with magmaflow, emplacement and cooling. Namely:

- High-field torque magnetometry analyses evidenced non-coaxialitybetween paramagnetic and ferrimagnetic fabrics

- Comparison of CPO and CAD data, retrieved from pyroxenesand plagioclases, reveals the presence of sub-fabrics for theparamagnetic minerals, most likely developed during magmaflow.

- The ferrimagnetic fabric reflects late-stage cooling stresses as-sociated with hardening of the rock at the chilled margins andnot directly related to the flow petrofabric.

All these data highlight the composite character of a fabric. They alsoimprove our knowledge of the evolution of the strain/stress field duringmagma flow, dyke emplacement and cooling.

Acknowledgements

The authours would like to acknowlegment FCT throught projectPEST/Oe/cte/LA-2013-2014. We are very grateful to Fátima Martin-Hernandez and Martin Chadima for their thorough review and usefulcomments.

Annex A

The volume integrated CPO diagrams from the large samples wereacquired by neutron diffraction (ND) measurements. The latter wereperformed on the KSN-2 neutron diffractometer located at the horizon-tal channel of the research reactor LVR-15 in the Nuclear Research Insti-tute, plc. Řež, Czech Republic. The neutron diffractometer KSN-2 isequipped with the texture goniometer TG-1 and HUBER goniometercradle with automatic data collection for transmission and reflectiongeometry. The primary data were measured either in the form of polefigures or diffraction patterns and then they were used to calculate thecoefficients of expansion C (μ,γ,l) for expanding the orientation distri-bution function (ODF) into a number of generalized spherical functions.The software package used for the experimental data treatments andODF calculation included TODFND, GSAS, Material Studio and MAUD.The instrument coordinate system was defined (I, J, K) and a set ofright-handed goniometer angles (Ω, Χ, Φ) after the description givenby Von Dreele (1997). The set of the measured patterns consists ofabout the 50 sample diffraction vectors for texture analysis of the sam-ples. The crystalline structure of samples was refined within the spacegroups: labradorite (phase I, triclinic space group C-1) and enstatite(phase IV, orthorhombic space group Pbca) by means of the GSASsoftware package (VonDreele, 1997). Firstly, the complete set of instru-mental characteristic (scale factors, zero-point, and profile parameters)and structure parameters (including set of the atom coordinates, occu-pation factors, and overall thermal coefficients) were based on the indi-vidual powder samples. These background coefficient and averagelattice parameters were used in the final Rietveld refinements byGSAS. The orientation distribution function (ODF—spherical harmoniccoefficients Cl

m,n) were determined by the Rietveld method (VonDreele, 1997; ResMat (http://www.resmat.com/)). Final refinementswere done with “none” sample symmetry and maximum harmonicorder L = 8. By means of harmonic analysis using the GSAS code weredetermined the pole figures with (hkl): (1–10), (001), (020), (110),(11–1), (100) for phase I and with (hkl): (200), (210), (020), (111),(001), (100) for phase IV, respectively.


http://www.resmat.com/


Fig. 7. Crystallographic preferred orientation (CPO) of plagioclase and orthopyroxene determined by neutron diffractionmeasurements. Pole figures are in lower hemisphere stereographicprojection in geographic coordinates. Contours andmaximum values represent levels of multiples of uniform distribution. CPO of samples FZ7-21b (a), FZ7-26 (b) and FZ7-32 (c) are com-pared with the AMS (green symbols), AFMS (blue symbols) and APMS (red symbols) data.


References

Banerjee, S.K., Stacey, F.D., 1967. The high-field torque-meter method of measuring mag-netic anisotropy in rocks. In: Collinson, D.W., Creer, K.M., Runcorn, S.K. (Eds.),Methods in Palaeomagnetism. Elsevier, Amsterdam, New York, pp. 470–476.

Bascou, J., Camps, P., Dautria, J.M., 2005. Magnetic versus crystallographic fabrics in a ba-saltic lava flow. J. Volcanol. Geotherm. Res. 145 (1–2), 119–135.

Batchelor, G.K., 1967. An Introduction to Fluid Mechanics. Cambridge Univ. Press,Cambridge, U. K (615 pp.).

Boiron, T., Bascou, J., Camps, P., Ferré, E.C., Maurice, C., Guy, B., Gerbe, M.C., Launeau, P.,2013. Internal structure of basalt flows: insights from magnetic and crystallographicfabrics of the La Palisse volcanics, FrenchMassif Central. Geophys. J. Int. 193, 585–602.

Borradaile, G.J., Henry, B., 1997. Tectonic applications of magnetic susceptibility and itsanisotropy. Earth Sci. Rev. 42, 49–93.

Borradaile, G.J., Jackson, M., 2010. Structural geology, petrofabrics and magnetic fabrics(AMS, AARM, AIRM). J. Struct. Geol. 32, 1519–1551.

Borradaile, G.J., Werner, T., 1994. Magnetic anisotropy of some phyllosilicates.Tectonophysics 235, 233–248.

Borradaile, G.J., Keeler, W., Alford, C., Sarvas, P., 1987. Anisotropy of magneticsusceptibility of some metamorphic minerals. Phys. Earth Planet. Inter. 48,161–166.

Callot, J.P., Guichet, X., 2003. Rock texture and magnetic lineation in dykes: a simple ana-lytical model. Tectonophysics 366, 207–222.

Cifelli, F., Mattei, M., Chadima, M., Lenser, S., Hirt, A.M., 2009. The magnetic fabric in “un-deformed clays”: AMS and neutron texture analyses from the Rift Chain (Morocco).Tectonophysics 466 (1–2), 79–88.

Chadima, M., Cajz, V., Týcová, P., 2009. On the interpretation of normal and inverse mag-netic fabric in dikes: examples from the Eger Graben, NW Bohemian Massif.Tectonophysics 466, 47–63. http://dx.doi.org/10.1016/j.tecto.2008.09.005.

Fanjat, G., Shcherbakov, V., Camps, P., 2012. Influence of magnetic interactions on theanisotropy of magnetic susceptibility: the case of single domain particles packed inglobule aggregates. Stud. Geophys. Geod. 56, 751–768.

Ferré, E.C., 2002. Theoretical models of intermediate and inverse AMS fabrics. Geophys.Res. Lett. 29, 31–34.

Ferré, E.C., Martin-Hernandez, F., Teyssier, C., Jackson, M., 2004. Paramagnetic and ferro-magnetic anisotropy of magnetic susceptibility in migmatites: measurements inhigh and low fields and kinematic implications. Geophys. J. Int. 157, 1119–1129.http://dx.doi.org/10.1111/j.1365-246X.2004.02294.x.


Flanders, P.J., Schuele,W.J., 1964. Anisotropy in the basal plane of hematite single crystals.Philos. Mag. 9, 485–490.

Geoffroy, L., Callot, J.P., Auborg, C., Moreira, M., 2002. Magnetic and plagioclase linear fab-ric discrepancy in dykes: a new way todefine the flow vector using magnetic folia-tion. Terra Nova 14, 183–190.

Henry, B., 1983. Interprétation quantitative de l'anisotropie de susceptibilité magnétique.Tectonophysics 91, 165–177.

Henry, B., 1997. The magnetic zone axis: a new element of magnetic fabric for the inter-pretation of the magnetic lineation. Tectonophysics 271, 325–329.

Henry, B., Daly, L., 1983. From qualitative to quantitative magnetic anisotropy analysis:the prospect of finite strain calibration. Tectonophysics 98, 327–336.

Hirt, A.M., Almqvist, B.S.G., 2012. Unraveling magnetic fabrics. Int. J. Earth Sci. 101,613–624.

Holtzman, B.K., Kohlstedt, D.L., Morgan, J.P., 2005. Viscous energy dissipation and strainpartitioning in partially molten rocks. J. Petrol. 46, 2569–2592. http://dx.doi.org/10.1093/petrology/egi065.

Hrouda, F., 1982. Magnetic anisotropy of rocks and its application in geology andgeophysics. Geophys. Surv. 5, 37–82.

Hrouda, F., Jelinek, V., 1990. Resolution of ferrimagnetic and paramagnetic anisotropies inrocks, using combined low-field and high-field measurements. Geophys. J. Int. 103,75–84.

Jelinek, V., 1981. Characterization of the magnetic fabric of rocks. Tectonophysics 79,63–67.

Jelinek, V., 1984. On a mixed quadratic invariant of the magnetic susceptibility tensor. J.Geophys. 56, 58–60.

Jelinek, V., 1996. Theory and measurement of the anisotropy of isothermal remanentmagnetization of rocks. Trav. Geophys. 37, 124–134.

Kratinová, Z., Závada, P., Hrouda, F., Schulmann, K., 2006. Nonscaled analogue modeling ofAMS development during viscous flow: a simulation on diapir‐like structures.Tectonophysics 418, 51–61.

Kratinová, Z., Jezek, J., Schulmann, K., Hrouda, F., Shail, R.K., Lexa, O., 2010. Noncoaxial K-feldspar and AMS subfabrics in the Land's End granite, Cornwall: evidence of mag-matic fabric decoupling during late deformation and matrix crystallization. J.Geophys. Res. Solid Earth 115, B09104. http://dx.doi.org/10.1029/2009JB006714.

Lamali, A., Merabet, N., Henry, N., Maouche, S., Graine-Tazerout, K., Mekkaoui, A., Ayache,M., 2013. Polyphased geodynamical evolution of the Ougarta (Algeria) magmaticcomplexes evidenced by paleomagnetic and AMS studies. Tectonophysics 588, 82–89.

Leblanc, M., 1974. Le grand dyke de dolérites de l'Anti.Atlas et le magmatisme jurassiquedu Sud-Marocain. CR Acad. Sci. Paris 278 (D), 2943–2946.


http://refhub.elsevier.com/S0040-1951(14)00198-X/rf0255





























http://dx.doi.org/10.1111/j.1365-246X.2004.02294.x














http://dx.doi.org/10.1093/petrology/egi065

http://dx.doi.org/10.1093/petrology/egi065















http://dx.doi.org/10.1029/2009JB006714







Lexa, O., Štípská, P., Schulmann, K., Baratoux, L., Kröner, A., 2005. Contrasting texturalrecord of two distinct metamorphic events of similar P–T conditions and differentdurations. J. Metamorph. Geol. 23 (8), 649–666.

Lutgens, F.K., Tarbuck, E.J., Tasa, D.G., 2011. Essentials of Geology, 11th Edition. PearsonEducation Limited, United Kingdom.

Machek, M., Roxerová, Z., Závada, P., Silva, P.F., Henry, B., Dědeček, P., Petrovský, E.,Marques, F.O., 2014. Intrusion and deformation effects in rock salt hostinglamprophyre dyke, case study from Loulé diapir, Portugal (this volume).

Martin-Hernandez, F., García-Hernández, M.M., 2010. Magnetic properties and anisotropyconstant of goethite single crystals at saturating high fields. Geophys. J. Int. 181,756–776.

Martín-Hernandez, F., Hirt, A.M., 2001. Separation of ferrimagnetic and paramagnetic an-isotropies using a high-field torsion magnetometer. Tectonophysics 337, 209–222.

Martín-Hernandez, F., Hirt, A.M., 2004. A method for the separation of paramagnetic, fer-rimagnetic and haematite magnetic subfabrics using high-field torque magnetome-try. Geophys. J. Int. 157, 117–127.

Martín-Hernández, F., Bominaar-Silkens, I., Dekkers, M.J., Maan, J.K., 2006. High-field can-tilever torque magnetomery as a tool for single crystal magnetocrystalline anisotropydeterminations. Tectonophysics 418, 21–30.

McCabe, C., Jackson, M., Ellwood, B.B., 1985. Magnetic anisotropy in the Trenton lime-stones: results of a new technique, anisotropy of anhysteretic susceptibility. Geophys.Res. Lett. 12, 333–336.

Moreira, M., Geoffroy, L., Pozzi, J.P., 1999. Ecoulement magmatique dans les dykes dupoint chaud des Açores: étude préliminaire par anisotropie de susceptibilitémagnétique ASM dans l'Île de Sao Jorge. C. R. Acad. Sci., Ser. IIa Sci. Terre Planets329, 15–22. http://dx.doi.org/10.1016/S1251-8050(99)80222-5.

Park, Y., Means, W.D., 1996. Direct observation of deformation processes in crystalmushes. J. Struct. Geo. 18 (6), 847–858.

Porath, H., Raleigh, C.B., 1967. An origin of the triaxial basal-plane anisotropy in hematitecrystals. J. Appl. Phys. 38, 2401–2402.

Potter, D.K., Stephenson, A., 1988. Single‐domain particles in rocks and magnetic fabricanalysis. Geophys. Res. Lett. 15, 1097–1100.

Rochette, P., 1988. Inverse magnetic fabric carbonate bearing rocks. Earth Planet. Sci. Lett.90, 229–237.

Rochette, P., Jackson, M.J., Aubourg, C., 1992. Rockmagnetism and the interpretation ofanisotropy of magnetic susceptibility. Rev. Geophys. 30, 209–226.

Rochette, P., Aubourg, C., Perrin, M., 1999. Is this magnetic fabric normal? A review andcase studies in volcanic formations. Tectonophysics 307, 219–234.

Roperch, P., Taylor, G.K., 1986. The importance of gyromagnetic remanence in alternatingfield demagnetization. Some new data and experiments on G.R.M. and R.R.M.Geophys. J. R. Astron. Soc. 87, 949–965.

Schelley, D., 1985. Determining palaeo-flow directions from groundmass fabrics in theLyttelton radial dykes, New Zealand. J. Volcanol. Geotherm. Res. 25, 69–79.

Schmidt, V., Hirt, A.M., Rosselli, P., Martín-Hernández, F., 2007. Separation of diamagneticand paramagnetic anisotropy by high-field, low-temperature torque measurements.Geophys. J. Int. 168, 40–47. http://dx.doi.org/10.1111/j.1365-1246X.2006.03202.x,02007.


Silva, P.F., Marques, F.O., Henry, B., Mateus, A., Lourenço, N., Miranda, J.M., 2004. Prelimi-nary results of a study of magnetic properties in the Foum‐Zguid dyke (Morocco).Phys. Chem. Earth 29, 909–920. http://dx.doi.org/10.1016/j.pce.2004.01.016.

Silva, P.F., Henry, B., Marques, F.O., Madureira, P., Miranda, J.M., 2006a. Paleomagneticstudy of the Great Foum Zguid dyke (Southern Morocco): a positive contact test re-lated to metasomatic processes. Geophys. Res. Lett. 33, L21301. http://dx.doi.org/10.1029/2006GL027498.

Silva, P.F., Henry, B., Marques, F.O., Mateus, A., Madureira, P., Lourenço, N., Miranda, J.M.,2006b. Variation of magnetic properties in sedimentary rocks hosting the FoumZguid dyke (southern Morocco): combined effects of re‐crystallization and Fe‐meta-somatism. Earth Planet. Sci. Lett. 241, 978–992.

Silva, P.F., Henry, B., Marques, F.O., Font, E., Mateus, A., Vegas, R., Miranda, J.M., Palomino, R.,Palencia‐Ortas, A., 2008. Magma flow, exsolution processes, and rockmetasomatism inthe Great Messejana–Plasencia dyke (Iberian Peninsula). Geophys. J. Int. 175, 806–824.http://dx.doi.org/10.1111/j.1365-246X.2008.03920.x.

Silva, P.F., Marques, F.O., Henry, B., Madureira, P., Hirt, A.M., Font, E., Lourenço, N., 2010.Thick dyke emplacement and internal flow: a structural and magnetic fabric studyof the deep-seated dolerite dyke of Foum Zguid (southern Morocco). J. Geophys.Res. 115, B12108. http://dx.doi.org/10.1029/2010JB007638.

Smith, J.V., 1997. Shear thickening dilatancy in crystal-rich flows. J. Volcanol. Geotherm.Res. 79 (1–2), 1–8.

Smith, J.V., 1998. Interpretation of domainal groundmass textures in basalt lavas of thesouthern Lamington Volcanics, eastern Australia. J. Geophys. Res. Solid Earth 103(B11), 27383–27391.

Smith, J.V., 2002. Structural analysis of flow-related textures in lavas. Earth-Sci. Rev. 57(3–4), 279–297.

Smith, J.V., Miyake, Y., Yamauchi, S., 1993a. Flow direction and groundmass shear zones indykes. Shimane Peninsula, Japan. Geol. Mag. 130 (1), 117–120.

Smith, J.V., Yamauchi, S., Miyake, Y., 1993b. Microshear zones in a Miocene submarinedacite dome of southwest Japan. Bull. Volcanol. 55 (6), 438–442.

Smith, J.V., Yamauchi, S., Miyake, Y., 1994. Coaxial progressive deformation textures in ex-trusive and shallow intrusive rocks, Southwest Japan. J. Struct. Geol. 16 (3), 315–322.

Stephenson, A., 1980. The measurement of the magnetic torque acting on a rotating sam-ple using an air turbine. J. Phys. E Sci. Instrum. 13, 311.

Stephenson, A., Sadikum, S., Potter, D., 1986. A theoretical and experimental comparisonof the susceptibility and remanence in rocks and minerals. Geophys. J. R. Astron. Soc.84, 185–200.

Tarling, D.H., Hrouda, F., 1993. The Magnetic Anisotropy of Rocks. Chapman and Hall,London, U.K.

Tullis, T.E., 1976. Experiments on origin of slaty cleavage and schistosity. Geol. Soc. Am.Bul. 87 (5), 745–753.

Von Dreele, R., 1997. Quantitative texture analysis by Rietveld refinement. J. Appl.Crystallogr. 30, 517–526.

Závada, P., Schulmann, K., Konopásek, J., Ulrich, J.S., Lexa, O., 2007. Extreme ductility offeldspar aggregates—melt-enhanced grain boundary sliding and creep failure: rheo-logical implications for felsic lower crust. J. Geophys. Res. Solid Earth 112. http://dx.doi.org/10.1029/2006JB004820.























http://dx.doi.org/10.1016/S1251-8050(99)80222-5


















http://dx.doi.org/10.1111/j.1365-1246X.2006.03202.x, 02007

http://dx.doi.org/10.1111/j.1365-1246X.2006.03202.x, 02007

http://dx.doi.org/10.1016/j.pce.2004.01.016

http://dx.doi.org/10.1029/2006GL027498

http://dx.doi.org/10.1029/2006GL027498




http://dx.doi.org/10.1111/j.1365-246X.2008.03920.x

http://dx.doi.org/10.1029/2010JB007638

























http://dx.doi.org/10.1029/2006JB004820


evidence for non-coaxiality of ferrimagnetic and paramagnetic fabrics, developed during magma flow...

Documents