Oxidation State of the Lithospheric Mantle below the

JOURNAL OF PETROLOGY

VOLUME 55

NUMBER 12

PAGES 2457^2480

2014

doi:10.1093/petrology/egu063

Oxidation State of the Lithospheric Mantle below the Massif Central, France LAURA UENVER-THIELE1*, ALAN B. WOODLAND1, HILARY DOWNES2 AND RAINER ALTHERR3 1

INSTITUT FU«R GEOWISSENSCHAFTEN, JWG UNIVERSITY, FRANKFURT AM MAIN, GERMANY

2

DEPARTMENT OF EARTH AND PLANETARY SCIENCES, BIRKBECK, UNIVERSITY OF LONDON, LONDON WC1E

7HX, UK 3

RECEIVED DECEMBER 19, 2013; ACCEPTED OCTOBER 27, 2014

Oxidation state is a sensitive indicator of geochemical processes within the upper mantle. Here we report results of a regional study of the oxidation state of spinel peridotite xenoliths from 45 volcanic centers distributed over 20 000 km2 in the Massif Central, France. The log fO2 values relative to the fayalite^magnetite^ quartz oxygen buffer (FMQ) were determined from the equilibrium between the Fe-bearing components in olivine, orthopyroxene and spinel, with the Fe3þ content of spinel measured either by Mo« ssbauer spectroscopy or by electron microprobe using secondary spinel standards. For the entire suite of samples, log fO2 values range between FMQ ^ 047 and FMQ þ166. Our data confirm the presence of two distinct lithospheric mantle domains, previously reported in the literature, lying north and south of 45830’N, respectively. The northern domain, with its refractory bulk composition, tends to record more oxidized conditions, having log fO2 values mostly at or above FMQ þ1. The log fO2 in the southern domain is more variable, including values below FMQ. Assuming that increasing equilibration temperatures among xenoliths reflect increasing depths of origin, samples from the northern domain suggest that the shallower part of the subcontinental lithospheric mantle (SCLM) is somewhat more oxidized than at deeper levels. On the other hand, such a general observation cannot be made for the southern domain. The high log fO2 values of harzburgites suggest that they are more sensitive to resetting of their oxidation state by metasomatism than lherzolites. In terms of modally metasomatized xenoliths, the ‘melt’ leading to the addition of clinopyroxene apparently had a higher oxidation state (log fO24FMQ þ1) than the agent responsible for crystallization of amphibole (log fO2 FMQ þ 06). Furthermore, amphibolebearing and amphibole-free peridotites exhibit the same range in

*Corresponding author. E-mail: [email protected]

fO2. Cryptic metasomatism can also reset oxidation state, sometimes very effectively. Metasomatic processes are probably the reason why the xenolith suite from the Massif Central records relatively high log fO2 values compared with ‘normal’ non-cratonic SCLM. This study demonstrates the utility of using oxidation state to help characterize and delineate domains in the lithospheric mantle.

KEY WORD: oxidation state, mantle peridotite, metasomatism, lithospheric mantle, Massif Central

I N T RO D U C T I O N The oxidation state of the Earth’s upper mantle has been intensively investigated and discussed over the last 25 years (e.g. Frost & McCammon, 2008). The importance of this parameter is partly due to the fact that the speciation of volatiles (i.e. CÔ^H fluids) is a direct function of the prevailing oxygen fugacity (fO2), thus influencing many geochemical processes such as partial melting, degassing of the Earth, diamond formation and metasomatism (e.g. Wood et al., 1990; Ballhaus & Frost, 1994). Metasomatic interactions in the subcontinental lithospheric mantle (SCLM) are frequently observed to produce an increase in fO2 (e.g. Mattioli et al., 1989; Woodland et al., 1992, 1996, 2006; Ballhaus, 1993), although this is not always the case (e.g. Chen et al., 1991). It is apparent that, although the fO2 of a rock can preserve a signature reflecting its

ß The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@ oup.com

Downloaded from http://petrology.oxfordjournals.org/ by guest on December 20, 2015

INSTITUT FU«R GEOWISSENSCHAFTEN, UNIVERSITA«T HEIDELBERG, 69120 HEIDELBERG, GERMANY


VOLUME 55

DECEMBER 2014

THE MASSIF CENTRAL AND THE SCLM The Massif Central constitutes one of the largest exposures of crystalline basement rocks in Western Europe that owes its origin to the Variscan orogeny. These rocks were subsequently affected by the development of the European Cenozoic rift system, which produced the Limagne Graben among other structures (Fig. 1; Merle & Michon, 2001). The Massif Central is marked by the presence of numerous Tertiary^Quaternary volcanic fields and isolated volcanoes. This volcanic activity commenced in the Late Eocene and continued through to the Holocene (Maury & Varet, 1980) and involved the eruption of predominantly alkali basalts and basanites (Wilson & Downes, 1991). The different fields were active over different discrete periods of time (Fig. 2; Maury & Varet, 1980; Briot & Harmon, 1989). Inspection of Figs 1 and 2 reveals no correlation between eruption age and geographical location (for example, the youngest activity is found not only in the north, near Clermont-Ferrand, but also in the Vivarais field, 150 km to the SE). During some periods, more than one field was active (see Fig. 2). Whether or not the tectonic activity and volcanism of the Massif Central is related to a rising mantle plume is still open to debate. Granet et al. (1995) and Sobolev et al. (1997) presented some compelling geophysical evidence for uplift of the lithosphereâsthenosphere boundary, which supports the existence of a plume. In contrast, Merle & Michon (2001) suggested that the major stage of volcanism that began 14 Myr ago could have been related to thermal erosion of the lithosphere. In any case, there is ample evidence for magmas having traversed the uppermost mantle and crust over a time span of more than 25 Myr. This provides numerous opportunities for melt^rock interaction to have taken place, potentially modifying the compositional character of the lithospheric mantle. Such effects are over and above those geochemical processes that may have occurred since the Variscan-age amalgamation of the SCLM beneath the Massif Central (Lenoir et al., 2000b; Wittig et al., 2006).

SAMPLI NG As this is a regional study, we have endeavored to obtain xenolith samples over the widest possible geographical area. Samples were investigated from 45 volcanic centers spanning an area of 20 000 km2 (Fig. 1). The localities are distributed over a total of 12 volcanic fields that were active during periods from 20 Ma to recent times (Fig. 2). Although no systematic correlation between eruption age and geographical location is apparent, the xenolith-bearing localities in the northern domain tend to be older than 4 Ma, except for the Puy de Beaunit locality, which is significantly younger (54 7 ka; Boivin et al., 2009).

2458


past history, more recent ‘event(s)’ can reset this parameter (e.g. Woodland et al., 2006). Thus, regional variations in oxidation state can be used to help assess the extent of metasomatic interactions and delineate regions of the mantle that have experienced different (or similar) geochemical histories. Tertiary and Quaternary volcanism in the French Massif Central has sampled the underlying SCLM in the form of xenoliths over a wide geographical area (20 000 km2). Such an extensive distribution of peridotite xenoliths provides a unique opportunity to investigate regional variations in mantle structure and composition, including oxidation state. In general, the bulk composition of the SCLM ranges from lherzolite to harzburgite, implying that partial melting and melt extraction processes are important. The occurrence of additional phases such as amphibole or phlogopite, or different bulk compositions such as wehrlite and pyroxenite, are indicators of further processes that have chemically modified the SCLM after its stabilization. In addition, other peridotite samples have experienced ‘cryptic’ metasomatism, as revealed by enrichments in trace elements [e.g. light rare earth elements (LREE), Sr] and/or modification of isotopic signatures (e.g. Zangana et al., 1997; Lenoir et al., 2000b; Downes et al., 2003). Textural and geochemical studies of mantle-derived xenoliths suggest the existence of two distinct domains in the SCLM beneath the Massif Central that were amalgamated during the Variscan orogeny (Lenoir et al., 2000b). The boundary between the two mantle domains appears to run approximately east^west at 45830’N (Lenoir et al., 2000b; Downes et al., 2003). The domain to the north is generally more refractory, but has undergone a pervasive re-enrichment of LREE. In contrast, the southern domain has a more juvenile (less depleted) composition but is less enriched or is even depleted in LREE. For example, these two domains exhibit differing signatures in terms of their Ce/Sm, Eu/Yb, Nb/Ta and Zr/Hf ratios (Lenoir et al., 2000b). In addition, different types and degrees of metasomatism are apparent across the Massif Central. In this context, the different histories of these two juxtaposed SCLM domains may also be reflected in their oxidation state, with local variations also to be expected owing to metasomatic interactions. Therefore, we have undertaken a study to investigate the oxidation state of the SCLM beneath the Massif Central over the widest geographical area possible and to determine if these two juxtaposed mantle domains record different oxidation states. In addition, we sought to assess how oxidation state might have been influenced by metasomatic processes and how this parameter can be used to place further constraints on lithospheric evolution in general and more specifically on that of the European SCLM.

NUMBER 12

UENVER-THIELE et al.

SCLM OXIDATION STATE


Fig. 1. Geological sketch map of the Massif Central illustrating the major volcanic fields. Also shown are the locations of the studied samples along with the 45830’N boundary between northern and southern mantle domains as defined by Lenoir et al. (2000b). Abbreviations for volcanic centres: Ac, Alleyrac; As, Alleyras; AV, Aurelle^Verlac; B, Burzet; CB, Cre“te de Blandine; CH, Les Champs; CN, Cha“teau Neuf; Con, Conac; Fr, Fraisse; JC, St. Jean le Centenier; L, Longueroque; LA, Les Angles; LB, Le Buisson; LCx, Le Cheix; LCh, Les Chanats; LCr, Le Chier; LEs, L’Estrade; LF, Le Fau; LJ, Labastide de Juvinas; LP, La Prade; Mal, Marais de Limagne; Mb, Montboissier; MBAR, Montbar; MBo, Marais de Bore¤e; MC, Mont Coupet; MGR, Mont Gros; MP, Mont Peylenc; MPS, Malpas; MPx, Marais de Praclaux; MtB, Mont Briancon; NC, Nappe de Cussac; PB, Puy de Beaunit; PBe, Puys de Bessoles; PH, Puy de la Halle; PL, Pont Labeaume; PR, Puy du Roi; PV, Puy de Vergnes; R, Rochelambert; Ri, Ringues; RdL, Rocher du Lion; RP, Ray Pic; SS, Sommet de la Sape'de; St, Sauterre; StMa, Saint Maurice; VP, La Vestide du Pal.

2459


VOLUME 55

NUMBER 12

DECEMBER 2014

A total of nine localities lie within the northern domain. Depending on the type of eruptive center, the xenoliths were collected from lava flows, cinder cones or maar deposits. Sample selection was predicated on macroscopic observation of textures and xenolith size (45 cm). Even in the field, different textural types could be identified, from coarse-grained isotropic textures to fine-grained textures with variable degrees of foliation development. A total of 120 xenoliths were investigated for their oxidation state. A subset of the samples was previously characterized petrologically and geochemically in the studies of Werling (1994), Zangana (1995), Werling & Altherr (1997), Zangana et al. (1997) and Downes et al. (2003). This provided us with a limited amount of geochemical data from which we could make a first assessment of the relationship between metasomatism and oxidation state. This included the Ray Pic locality (RP in Fig. 1), from which 17 samples with different textures and trace element signatures were previously described by Zangana et al. (1997).

T H E P E R I D OT I T E S Except for one wehrlite sample (RP87-7), all the investigated xenoliths are spinel-bearing lherzolites and harzburgites. Because clinopyroxene contents are generally low and variably distributed, we did not consider that point counting of thin sections would accurately reflect the modal mineralogy for distinguishing between harzburgites and lherzolites. Instead, we applied the following criteria: harzburgites generally contain olivine with forsterite contents Xfo4091 and coexisting spinel with Cr2O3 22 wt %. Using the nomenclature of Mercier & Nicolas (1975), the xenoliths were further classified into three main textural groups: protogranular, porphyroclastic and equigranular. A secondary recrystallized texture

(complex protogranular or complex porphyroclastic) was also observed in a few samples from three localities [Puy de Beaunit (PB), Sauterre (ST) and Le Cheix (LCx); see Fig. 1]. These samples are all from the northern domain and are harzburgites. Mercier & Nicolas (1975) considered these textural types to be linked, so that one texture can develop from another through deformation. The degree of deformation increases from protogranular to equigranular textures and is attended by a significant reduction in grain size. A crude foliation defined by linear concentrations of spinel is observed in most samples with equigranular textures. The samples investigated in this study most commonly exhibit protogranular or porphyroclastic textures or are transitional between these two textures (protogranular 46%, protogranular^porphyroclastic 25%, porphyroclastic 17%, porphyroclasticêquigranular 7%, equigranular 4%). Thus, the majority of our samples are only weakly deformed, if at all. Some protogranular samples exhibit pyroxene^spinel clusters that are considered to be a characteristic of this textural type (Lenoir et al., 2000b). Such clusters occur particularly in samples from northern domain localities such as Puy de Beaunit (PB), Montboissier (MB) and Les Champs (CH). In the southern domain, only one protogranular sample with pyroxene^spinel clusters was identified from Ringue (Ri-D). It is considered that these clusters result from the decompression reaction of garnet þ olivine during mantle upwelling to produce the clinopyroxene þ orthopyroxene þ spinel assemblage (Nicolas et al., 1987). Spinel always occurs as a subordinate phase in the peridotites and can either have a similar grain size to the coexisting silicates or be finer grained and occur interstitially. Spinel usually has an irregular grain shape with either vermicular (Fig. 3a and b) or ‘holly leaf ’ form (Fig. 3c

2460


Fig. 2. Summary of the eruption ages of the volcanic centers in the 12 volcanic fields of the Massif Central [data after Maury & Varet (1980) and Nehlig et al. (2003)].



and d). In other cases, it occurs as very small (02 mm) subhedral to anhedral grains and can define a foliation in equigranular samples. Although grain shape varies with texture, as observed by Mercier & Nicolas (1975), more than one spinel type can occur in a single sample, especially for those with transitional textures. Minor amounts (2%) of brown amphibole occur in a few lherzolite xenoliths, particularly from the Deve's, Velay and Vivarais volcanic fields of the southern domain. Although amphibole and/or phlogopite has been reported in xenoliths from northern domain localities (Downes & Dupuy, 1987; Wilson & Downes, 1991; Yoshikawa et al., 2010), only one sample in our suite from Puy de Beaunit (PB) contains minor amphibole (sample 53A-LU6). Amphibole occurs predominantly in peridotites with protogranular or protogranular^porphyroclastic textures. In samples from L’Estrade (LEs in Fig. 1), amphibole is associated with spinel and has the same grain size as the coexisting anhydrous mineral phases. Otherwise, it occurs as small, dispersed interstitial grains not necessarily associated with spinel.

A N A LY T I C A L T E C H N I Q U E S Major elements Major element compositions of each of the minerals (spinel, olivine, orthopyroxene, clinopyroxene and amphibole) were determined with a JEOL JXA-8900 superprobe

at the Universita«t Frankfurt am Main. Operating conditions were 15 kV accelerating voltage, beam current of 20 nA and a 3 mm spot diameter. Data were collected in wavelength-dispersive mode with normal counting times of 20^40 s on the peak and 15^40 s on background (depending on element) and using a mix of natural and synthetic standards. Measurements were made on thin sections or polished grain mounts. The data are reported in Supplementary Data Electronic Appendix 1 (supplementary data are available for downloading at http://www.petrology.oxfordjournals.org).

Determination of Fe3þ/Fe in spinel For the most part, the Fe3þ content of spinel was determined by 57Fe Mo«ssbauer spectroscopy at the Universita«t Frankfurt am Main. Handpicked mineral separates (2^ 6 mg) were first washed in HF (24 h) and dilute HCl (5 h) to remove any silicate material or secondary magnetite from the grain boundaries. Each clean separate was ground under acetone, mixed with a small amount of sugar and packed into a pre-drilled hole in a 1mm thick Pb disc that was subsequently closed on both sides with tape. The sample diameter was calculated so that absorber thicknesses remained 5 mg Fe cm2 to avoid potential saturation effects (Woodland et al., 2006). Mo«ssbauer spectra were obtained at room temperature operating in transmission and constant acceleration mode with a nominal 50 mCi 57Co source in a Rh matrix. Calibration of the velocity scale was carried out relative to 25 mm thick a-Fe.

2461


Fig. 3. Backscattered electron images illustrating different types of spinel. (a) Vermicular spinel in sample ST-A from Sauterre; (b) vermicular spinel surrounded by clinopyroxene forming a spinel^pyroxene cluster (sample 53A-LU6, Puy de Beaunit); (c) holly leaf spinel partially surrounded by clinopyroxene (sample 65/2, St. Jean le Centenier); (d) holly leaf spinel (sample CON-C, Conac).


VOLUME 55

NUMBER 12

DECEMBER 2014

M AJ O R E L E M E N T C H E M I S T RY Olivine Forsterite contents range from Xfo ¼ 0887 to 0919 in our suite (except for sample 73-LU1 with Xfo ¼ 0877; Table 2). Olivine is compositionally homogeneous in a given sample with variations in forsterite content only very rarely exceeding 00015. Ni concentrations range from 0005 to 0009 cations per formula unit (c.p.f.u.), regardless of whether the xenolith is a harzburgite or lherzolite. These concentrations are typical for mantle-derived peridotites (e.g. McDonough & Rudnick, 1998). Fig. 4. Mo«ssbauerPspectra of spinel illustrating extremes in (a) high and (b) low Fe3þ/ Fe ratios.

Mirror-image spectra were collected over 512 channels using a velocity range of 5 mm s1. Spinel spectra are composed of two broad absorption peaks (Fig. 4a and b) and were fitted following the spectral model described by Woodland et al. (2006), using two doublets with pseudoVoigt peak form for Fe2þ and a single Lorentzian doublet for Fe3þ. The relative peak widths of the Fe2þ doublets were constrained to be equal. Spectral fitting was performed using the NORMOS software package (Wissel Elektronik GmbH, Germany). Hyperfine parameters determined for each sample are presented in Supplementary Data Electronic P Appendix 2. The uncertainty in the computed Fe3þ/ Fe is estimated P to be 001 absolute. Combining the measured Fe3þ/ Fe with the total Fe content from microprobe analyses allowed 3þ the Fe and Fe2þ contents of spinel to be determined. 3þ P Fe / Fe and Fe3þ contents are reported in Table 1. In some cases, measurement of Fe3þ in spinel by Mo«ssbauer spectroscopy was precluded either by the presence of alteration rims or by Pthe lack of sufficient spinel. For these samples, the Fe3þ/ Fe of spinel was determined

Orthopyroxene Orthopyroxenes are all Al- and Fe-rich enstatite with XMg ranging between 088 and 092. The values are similar to those for coexisting olivine, and suggest equilibrium partitioning of Fe and Mg between these two phases (e.g. von Seckendorf & O’Neill, 1993). Al contents span a range of 004^022 c.p.f.u., with the highest Al values found in the lherzolites (see Electronic Appendix 1). No differences in Al content or XMg are apparent between samples from the northern and southern domains. However, orthopyroxene in amphibole-bearing samples tends to have elevated Al contents, exceeding 013 c.p.f.u. (Fig. 5a). Ca contents range from 001 to 006 c.p.f.u. The highest Ca contents in orthopyroxene occur in three northern domain xenoliths, two lherzolites from Montboissier (Mb1, Mb4) and one harzburgite from Le Cheix (11-1). Major and minor element compositions of orthopyroxene are provided in Electronic Appendix 1.

Clinopyroxene Clinopyroxenes are Al-rich chromian diopsides with Mg/ (Mg þ Fe), XMg ¼ 088^094. Cr contents range from 002 to 005 c.p.f.u. Clinopyroxene in samples from the southern domain often have Al contents 402 c.p.f.u., whereas the Al content of northern domain clinopyroxenes is mostly

2462


by microprobe using a set of secondary spinel standards following the procedure outlined by Wood & Virgo (1989) and is indicated in Table 1. Corrections to the stoichiometP rically calculated value of Fe3þ/ Fe ranged from þ002 (underestimated) to 007 (overestimated), the differences being apparent between microprobe sessions. P Thus, although stoichiometrically calculated Fe3þ/ Fe for spinel has the potential to yield reasonable results (see also Canil & O’Neill, 1996), small day-to-day differences in the microprobe calibration can lead to systematic errors that will arbitrarily bias samples measured during a given session. These systematic errors would go unnoticed if no secondary standards were employed as a check. This approach does have somewhat larger errors in Fe3þ/Fe compared with measurements made by Mo«ssbauer spectroscopy (0025; Woodland et al., 1992, 1996).



Table 1: Representative spinel compositions and Fe3þ contents determined by Mo« ssbauer spectroscopy or with secondary spinel standards P Fe3þ/ Fe

9939

0057(5)

0229

10006

0064(5)

0233

9997

0130(9)

0289

1767

9935

0125(8)

0340

038

1964

9939

0065(6)

0230

013

035

2075

10051

0087(6)

0311

b.d.

029

2002

9792

0100(6)

0341

008

040

2003

9815

0065(5)

0258

1552

007

018

1601

10001

0082(8)

0220

923

1120

010

038

2112

10016

0058(5)

0242

1159

1299

010

040

2008

9964

0085(6)

0295

1257

1355

012

036

1972

10074

0072(6)

0244

3205

3359

1750

014

023

1655

10012

0121(9)

0281

015

5535

1105

1147

003

042

2088

9968

0070(5)

0281

n.d.

010

5637

938

1099

011

032

2058

9814

0064(10)

0266

007

013

4904

1887

1169

012

034

2030

10063

0083(5)

0306

28-5

007

010

5595

1031

1204

012

038

2074

9975

0082(6)

0293

32-4

003

006

5993

724

1098

012

039

2100

9978

0038(5)

0160

34-1y

n.d.

006

5614

1071

1206

003

042

2030

10000

0074(5)

0283

39-1y

n.d.

007

5301

1348

1295

010

042

1931

9963

0083(6)

0290

40A-2

002

004

5903

843

1117

012

044

2101

10027

0041(5)

0172

40A-LU3

004

007

5771

982

1193

b.d.

038

2090

10089

0065(5)

0254

41B-1y

003

006

5945

818

1190

012

037

2056

10069

0037(3)

0147

41B-3

003

008

5511

1206

1135

010

033

2021

9928

0056(5)

0222

41B-5

004

006

5729

1078

1098

012

035

2060

10024

0049(9)

0206

43-2

006

014

5444

1274

1202

011

036

2073

10063

0076(5)

0290

43-LU3

005

009

5465

1308

1169

b.d.

037

2083

10075

0067(5)

0263

44-LU3

004

014

4939

1605

1439

011

036

1890

9940

0087(7)

0265

45-1*

006

009

5455

1110

1258

011

041

2015

9907

0076(6)

0275

45-4y

n.d.

021

4745

1796

1315

008

029

1930

9887

0110(6)

0368

45-6y

n.d.

007

5609

1019

1223

007

050

2013

9958

0076(5)

0285

52-1y

n.d.

019

5104

1537

1246

008

040

1924

9905

0066(6)

0235

52-8

005

009

5780

879

1124

008

040

2127

9974

0062(5)

0255

53A-1y

n.d.

040

2570

4164

1463

000

024

1678

9984

0119(7)

0328

53A-LU1 (1.sp)*

008

007

3543

2960

1643

018

025

1747

9954

0142(8)

0363

53A-LU1 (2.sp)*

005

006

3291

3206

1824

024

020

1602

9980

0144(9)

0327

53A-LU6*

007

014

4002

2561

1439

015

033

1915

9995

0109(7)

0327

53A-LU8*

005

016

1832

4777

1934

023

016

1431

10036

0155(10)

0311

54A-1 (1.sp)*

007

006

4888

1628

1382

016

036

2027

9995

0166(6)

0445

54A-1 (2.sp)*

005

006

4823

1669

1475

013

034

2023

10051

0101(7)

0328

54A-1 (3.sp)*

007

009

4980

1554

1375

012

034

1942

9917

0072(6)

0234

55A-2y

n.d.

016

5369

1210

1224

003

039

2061

9963

0082(5)

0305

60-1

006

014

5701

925

1094

012

041

2180

9975

0063(5)

0265

60-2

009

009

5714

913

1159

011

042

2032

9889

0070(5)

0274

SiO2

TiO2

Al2O3

Cr2O3

FeO

MnO

NiO

MgO

1-4

001

006

5708

986

1145

014

044

2035

1-LU3*

005

002

5718

924

1272

013

040

2031

6-2y

n.d.

011

2966

3628

1818

010

012

1501

11-1y

n.d.

053

3849

2627

1554

003

031

11-LU2*

005

005

5394

1236

1282

013

14A-5*

006

003

5114

1523

1279

21-1y

n.d.

015

4809

1602

1283

21-2y

n.d.

012

5707

883

1141

22-1y

n.d.

016

3663

3118

22-5*

006

003

5804

22-7

003

003

5432

25-1*

003

006

5432

26B-3

003

b.d.

27-1y

n.d.

27-2y 28-4

Total

(continued)

2463


Fe3þ

Sample


VOLUME 55

NUMBER 12

DECEMBER 2014

Table 1: Continued SiO2

TiO2

Al2O3

Cr2O3

FeO

MnO

NiO

MgO

Total

Fe3þ

P Fe3þ/ Fe

61-2

007

011

5808

903

1134

010

039

2099

10013

0059(5)

0242

62-LU3

005

010

5970

728

1102

011

044

2120

9992

0053(5)

0223

63-1

006

008

5871

772

1101

012

038

2012

9820

0045(5)

0187

63-LU2

007

012

5864

842

1087

009

044

2148

10016

0050(9)

0230

64-1

007

014

5677

959

1047

008

034

2114

9862

0051(5)

0222

64-2

011

016

5068

1651

1143

012

034

1979

9917

0063(5)

0245

65-2

004

006

4789

2072

1223

016

030

1925

10064

0046(6)

0167

65-LU3

007

013

5757

868

1119

010

040

2069

9886

0070(5)

0285

65-LU4

006

003

5266

1492

1156

010

037

2029

9999

0055(5)

0215

70-LU2

003

010

5093

1725

1135

012

035

2035

10048

0060(5)

0240

71-2y

n.d.

007

5185

1620

1164

008

039

1981

10027

0057(5)

0223

73-LU1*

003

007

5791

815

1344

012

037

1932

9944

0054(6)

0184

78-LU3b*

003

002

4443

2225

1483

015

027

1781

9980

0081(7)

0238

80-LU3 (1.sp)*

005

003

5310

1175

1419

011

034

1953

9912

0096(6)

0306

80-LU3 (2.sp)*

004

006

5036

1414

1476

014

030

1884

9864

0099(7)

0298

CH11

002

006

5245

1475

1047

010

034

1999

9819

0079(6)

0274

CON-A*

005

004

3651

2926

1769

022

023

1666

10065

0123(8)

0294

CON-B

007

017

5229

1504

1179

011

035

2090

10076

0080(5)

0307

CON-C

005

008

5607

1029

1145

012

039

2069

9915

0071(5)

0282

CON-D

008

009

5664

1058

1116

011

043

2091

10000

0058(5)

0238

FR10 (1.Sp)

004

056

4169

2557

1274

015

027

1821

9926

0068(6)

0230

FR10 (2.Sp)

007

035

4434

2284

1213

013

028

1832

9848

0065(6)

0230

FR11

004

060

3936

2439

1767

020

032

1629

9888

0090(9)

0211

LA-A

005

005

5081

1603

1288

011

038

2030

10062

0105(5)

0366

LA-F (1.Sp)

007

005

5034

1575

1230

012

034

2053

9952

0095(6)

0345

LA-F (2.Sp)

008

011

4921

1661

1266

014

037

2044

9964

0098(6)

0345

LEs1

009

028

4785

1633

1504

016

033

1929

9937

0118(7)

0345

LEs2

007

011

5811

815

1161

011

039

2122

9979

0075(5)

0299

LEs3

006

011

5725

888

1103

014

041

2092

9882

0059(5)

0244

LEs4

008

021

5615

959

1224

013

042

2091

9973

0095(6)

0331

LEs5

004

009

5695

857

1165

011

043

2071

9858

0070(5)

0275

LEs6

007

008

5443

1208

1121

013

038

2095

9934

0077(5)

0312

MAL30

008

020

5574

1128

1197

015

037

2056

10037

0084(5)

0323

MB1

005

003

4600

1873

1492

017

035

1881

9907

0119(7)

0346

MB4

003

005

5027

1525

1323

013

038

1955

9891

0096(6)

0319

MB50

003

004

4929

1696

1307

014

033

1931

9919

0085(10)

0285

MB57 (1.Sp)

004

004

3570

3047

1539

018

021

1763

9969

0130(8)

0350

MB57 (2.Sp)

003

003

3872

2746

1466

016

024

1799

9931

0122(7)

0350

MB-B

005

002

4803

1844

1300

012

033

1930

9931

0077(6)

0261

MBAR-A

005

011

5852

832

1114

013

041

2133

10004

0071(5)

0297

MBAR-B

004

007

5813

993

965

012

040

2132

9970

0095(5)

0459

MBAR-D

003

007

5403

1304

1218

012

037

2044

10031

0069(5)

0257

MBAR-E

003

009

5978

817

1066

010

043

2144

10071

0038(5)

0167

MBo-B

003

010

5780

908

1223

015

042

2107

10090

0070(6)

0265

MBo-C

005

012

5465

1128

1205

011

041

1976

9843

0078(6)

0292

MBo-D

004

012

5427

1275

1237

010

035

2084

10087

0080(6)

0295

(continued)

2464


Sample



Table 1: Continued SiO2

TiO2

Al2O3

Cr2O3

FeO

MnO

NiO

MgO

Total

Fe3þ

P Fe3þ/ Fe

MBo-E

005

012

5608

1039

1319

013

042

2081

10121

0103(6)

0361

MC-B

003

007

3153

3396

1859

022

017

1635

10092

0165(9)

0362

MC-C

006

005

5423

1229

1244

015

038

1939

9901

0061(6)

0222

MC-D*

004

005

5711

903

1233

011

037

2018

9924

0062(5)

0231

MC-E

004

005

4887

1878

1244

013

029

1915

9978

0061(6)

0230

MC-F

005

004

5615

1096

1128

011

039

2119

10020

0074(5)

0301

MC-G*

003

004

2835

3847

1745

023

017

1511

9985

0111(9)

0257

MGR1

006

011

5688

1050

1129

014

041

2058

10000

0086(5)

0351

MP-B

005

012

6129

585

1132

012

047

2159

10084

0057(5)

0238

MP-C

004

010

5817

767

1276

012

046

2109

10043

0072(6)

0260

MP-D

004

012

4958

1774

1316

014

033

1961

10073

0086(6)

0293

MPS-B

007

009

5793

924

1081

013

039

2053

9921

0054(5)

0231

PH1

003

002

4481

2049

1487

015

029

1808

9875

0086(7)

0248

RdL-A

004

003

5785

892

1102

008

039

2073

9905

0039(5)

0164

RdL-B

006

016

5424

1170

1318

014

042

1999

9991

0095(6)

0327

RdL-C

005

007

5878

816

1098

011

038

2104

9959

0056(5)

0236

Ri-A

005

010

5567

1179

1179

014

037

2081

10072

0062(5)

0241

Ri-B

006

013

5786

837

1168

011

039

2115

9976

0060(5)

0239

Ri-D

008

014

4169

2566

1330

013

026

1897

10023

0092(6)

0299

RP83-70z

n.d.

n.d.

5450

1243

1161

n.d.

n.d.

2074

9913

0079(5)

0308

RP83-71z

n.d.

n.d.

5307

1353

1247

n.d.

n.d.

2005

9962

0086(6)

0309

RP87-2z

n.d.

n.d.

5924

894

1049

n.d.

n.d.

2046

9948

0050(5)

0209

RP87-4z

n.d.

027

4169

2498

1464

n.d.

n.d.

1804

9901

0113(7)

0329

RP87-6z

n.d.

n.d.

5581

1109

1142

n.d.

040

2076

9926

0073(5)

0292

RP87-7z

n.d.

n.d.

3185

3433

1653

n.d.

n.d.

1630

9901

0151(8)

0369

RP91-15z

n.d.

n.d.

5823

835

1242

n.d.

n.d.

2026

9895

0071(5)

0262

RP91-16z

n.d.

n.d.

5498

1188

1212

n.d.

n.d.

2026

9962

0066(5)

0247

RP91-17z

n.d.

n.d.

5046

1639

1255

n.d.

n.d.

1955

10140

0084(6)

0295

RP91-18z

n.d.

n.d.

4018

2575

1556

n.d.

n.d.

1753

9924

0132(8)

0356

RP91-2z

n.d.

023

3589

2966

1762

n.d.

n.d.

1622

9965

0140(9)

0327

RP91-21z

n.d.

n.d.

4599

2101

1303

n.d.

035

2102

9925

0086(6)

0292

RP91-22z

n.d.

n.d.

5305

1459

1142

n.d.

032

2027

9884

0073(5)

0290

RP91-3z

n.d.

018

4274

2288

1427

n.d.

n.d.

1918

9895

0122(6)

0368

RP91-4z

n.d.

n.d.

5551

1099

1132

n.d.

n.d.

2102

9939

0069(5)

0277

RP91-5z

n.d.

n.d.

5800

861

1274

n.d.

n.d.

1960

10019

0069(6)

0245

RP91-8z

n.d.

n.d.

5698

1082

1107

n.d.

n.d.

2052

9899

0044(5)

0182

ST2 (1.Sp)

003

004

5246

1237

1416

014

035

1930

9889

0111(7)

0347

ST2 (2.Sp)

003

003

5122

1301

1465

011

036

1910

9855

0116(6)

0347

ST-A (1.sp)*

003

017

3709

2842

1708

021

023

1787

10110

0133(8)

0334

ST-A (2.sp)*

003

011

3855

2747

1718

018

030

1797

10181

0128(8)

0323

ST-A (3.sp)*

003

014

3873

2729

1676

018

028

1805

10148

0122(8)

0315

ST-B*

006

006

5268

1366

1261

010

036

2063

10019

0071(6)

0257

STMA-B

004

007

3219

3572

1588

020

019

1742

10174

0118(8)

0310

*Fe3þ measured with secondary standards. yMicroprobe data from Werling (1994). zMicroprobe data from Zangana (1995). SpinelPcompositions are in wt % and Fe3þ contents are in c.p.f.u. (with error given in parentheses). The uncertainty in Fe3þ/ Fe is 001. n.d., not determined; b.d., below detection.

2465


Sample


VOLUME 55

NUMBER 12

DECEMBER 2014

Table 2: Summary of results indicating sample rock type and texture, along with calculated temperature, log fO2 values and forsterite contents and ferrosilite activities in coexisting olivine and orthopyroxene, respectively Sample

Locality*

Type of deposit

Rock type

Texture

T (8C)

Xfo

aFeSiO3

log fO2 (FMQ)

Petite Chaıˆne de la Sioule PH

B

H

pr

938

0907

000812

091

54A-1 (1.sp)

ST

B

L

p

1089

0898

000741

061

54A-1 (2.sp)

ST

B

L

p

1089

0898

000741

043

54A-1 (3.sp)

ST

B

L

p

1089

0905

000741

035

ST2 (1.Sp)

ST

B

L

pr

812

0902

000893

161

ST2 (2.Sp)

ST

B

L

pr

812

0902

000893

166

ST-A (1.sp)

ST

B

H

cpr

841

0910

000687

122

ST-A (2.sp)

ST

B

H

cpr

841

0910

000687

121

ST-A (3.sp)

ST

B

H

cpr

841

0910

000687

114

ST-B

ST

B

L

p

1023

0906

000762

048

53A-1

PB

P

H

cpr

1116

0915

000674

087

53A-LU1(1.sp)

PB

P

H

cpr

861

0905

000775

113

53A-LU1(2.sp)

PB

P

H

cpr

861

0905

000775

115

53A-LU6

PB

P

H, A

cp

1068

0913

000658

097

53A-LU8

PB

P

H

cpr

861

0906

000770

104

6-2

Mb

B, P

H

pr

750

0910

000677

107

MB1

Mb

B, P

L

pr

858

0904

000782

132

MB4

Mb

B, P

L

pr–p

1027

0911

000859

141

MB50

Mb

B, P

H

pr

770

0908

000778

116

MB57 (1.Sp)

Mb

B, P

H

pr

895

0917

000637

143

MB57 (2.Sp)

Mb

B, P

H

pr

895

0917

000637

142

MB-B

Mb

B, P

H

cp

807

0909

000718

088

STMA-B

StMa

P

H

pr

926

0917

000605

114

CH11

CH

B

L

pr

783

0910

000779

125

FR10 (1.Sp)

Fr

B

H

p

919

0905

000742

019

FR10 (2.Sp)

Fr

B

H

p

919

0905

000742

019

FR11

Fr

B

H

p

931

0902

000760

068

Chaıˆne des Puys

Livradois

Forez

Mont Dore 11-1

LCx

B

H

cpr

1198

0898

000901

070

11-LU2

LCx

B

L

p

869

0900

000855

050

78-LU3b

PBe

B

H

pr–p

816

0912

000723

094

14A-5

LA

B, P

H

pr–p

970

0911

000682

089

LA-A

LA

B, P

H

p

980

0910

000687

116

LA-F (1.Sp)

LA

B, P

L

p–e

957

0907

000760

101

LA-F (2.Sp)

LA

B, P

L

p–e

957

0907

000760

103

21-1

NC

B

L

pr–p

1037

0906

000755

086

21-2

NC

B

L

e

852

0896

000888

050

80-LU3 (1.sp)

LCh

P

L

p–e

879

0901

000847

112

Cezallier

Cantal

(continued)

2466


PH1



Table 2: Continued Sample

Locality*

Type of deposit

Rock type

Texture

80-LU3 (2.sp)

LCh

P

L

p–e

1-4

PR

P

L

p

1-LU3

PR

P

L, A

pr–p

25-1

LCr

P

L

p–e

26B-3

As

B

H

27-1

MPx

P

27-2

MPx

28-4

T (8C)

log fO2 (FMQ)

Xfo

aFeSiO3

0901

000847

108

805

0906

000791

073

884

0901

000898

069

806

0899

000925

081

p

779

0914

000622

122

L

pr

1001

0905

000790

055

P

L

pr

1048

0897

000835

010

R

P

H

pr–p

1041

0911

000751

054

28-5

R

P

L, A

p

1088

0897

000882

045

52-1

LB

B

L

pr

857

0898

000866

032

52-8

LB

B

L

pr

984

0904

000798

050

55A-2

MtB

P

L

pr

997

0900

000822

057

60-1

CN

B

L

pr–p

1078

0904

000757

028

60-2

CN

B

L, A

pr

963

0898

000843

072

61-2

Ac

P

L, A

pr

951

0895

000902

018

CON-A

Con

P

H

pr

958

0899

000886

084

CON-B

Con

P

L

pr

1046

0909

000699

066

CON-C

Con

P

L

pr–p

1012

0903

000775

052

CON-D

Con

P

L

pr–p

983

0901

000789

016

LEs1

LEs

P

L, A

e

1064

0890

001077

093

LEs2

LEs

P

L, A

pr

1035

0894

000878

034

LEs3

LEs

P

L, A

pr

906

0896

000889

028

LEs4

LEs

P

L, A

pr

1089

0887

001025

055

LEs5

LEs

P

L, A

pr

998

0895

000889

040

LEs6

LEs

P

L, A

pr

1016

0906

000738

059

34-1

Mal

P

L, A

p–e

940

0894

000976

052

MAL30

Mal

P

L

pr

1055

0900

000843

064

MBAR-A

MBAR

P

L, A

pr

910

0899

000838

063

MBAR-B

MBAR

P

L

pr

889

0906

000800

075

MBAR-D

MBAR

P

L, A

pr–p

840

0905

000800

080

MBAR-E

MBAR

P

L, A

pr

920

0895

000869

047

22-1

MC

P

H

e

721

0908

000840

075

22-5

MC

P

L

p

947

0900

000822

022

22-7

MC

P

L

pr–p

889

0905

000823

115

MC-B

MC

P

H

pr–p

834

0912

000674

161

MC-C

MC

P

L

p

841

0897

000923

046

MC-D

MC

P

L

p

866

0895

000956

047

MC-E

MC

P

L

p

921

0912

000687

056

MC-F

MC

P

L

p

911

0903

000816

075

MC-G

MC

P

H

pr–p

871

0917

000614

100

MGR1

MGR

P

L, A

pr–p

988

0902

000897

092

MPS-B

MPS

B

L

pr–p

1000

0897

000851

021

879

Deve`s

RdL

B

L, A

p

830

0899

000896

005

RdL

B

L

pr–p

996

0895

000875

075

RdL-C

RdL

B

L

pr–p

990

0898

000849

015

(continued)

2467


RdL-A RdL-B


VOLUME 55

NUMBER 12

DECEMBER 2014

Table 2: Continued Sample

Locality*

Type of deposit

Rock type

Texture

T (8C)

Xfo

aFeSiO3

log fO2 (FMQ)

Ri-A

Ri

B

L

p

956

0903

000745

031

Ri-B

Ri

B

L

pr–p

1007

0893

000911

010

Ri-D

Ri

B

H

pr

1098

0911

000645

058

MBo-B

MBo

P

L, A

pr

917

0897

000882

062

MBo-C

MBo

P

L

pr

910

0902

000807

079

MBo-D

MBo

P

L

pr

935

0904

000788

081

MBo-E

MBo

P

L

pr–p

957

0902

000856

126

32-4

MP

B

L

p

832

0887

001078

032

MP-B

MP

B

L

pr

888

0893

000916

035

MP-C

MP

B

L

pr

910

0887

001017

048

MP-D

MP

B

L

pr

892

0905

000753

081

40A-2

B

B

L

p–e

840

0898

000888

003

40A-LU3

B

B

L

pr–p

907

0901

000815

056

41B-1

VP

P

L

pr

845

0889

001073

004

41B-3

VP

P

L

p

885

0906

000743

038

41B-5

VP

P

L

p

868

0901

000790

008

43-2

LF

B

L, A

pr–p

944

0900

000780

050

43-LU3

LF

B

L, A

pr

973

0906

000728

047

44-LU3

LJ

B

L, A

pr

892

0890

001002

060

45-1

SS

P

L, A

pr–p

887

0900

000877

075

45-4

SS

P

L

pr

1079

0903

000833

089

45-6

SS

P

L, A

pr–p

938

0900

000819

063

62-LU3

PL

B

L

pr

883

0898

000831

024

39-1

RP

B

L

pr–p

900

0899

000796

063

RP83-70

RP

B

L

pr–p

1067

0902

000786

054

RP83-71

RP

B

L

p

826

0907

000752

117

RP87-2

RP

B

L

p

853

0901

000868

039

RP87-4

RP

B

H

p–e

854

0906

000741

106

RP87-6

RP

B

L

pr–p

984

0900

000833

058

RP87-7

RP

B

W

pr

1021

0909

000674

119

RP91-15

RP

B

L

e

856

0893

000995

078

RP91-16

RP

B

L

p–e

850

0902

000814

068

RP91-17

RP

B

L, A

pr

1020

0908

000660

066

RP91-18

RP

B

C

p–e

893

0907

000748

140

RP91-2

RP

B

H

e

863

0901

000816

116

RP91-21

RP

B

H

pr

992

0903

000761

050

RP91-22

RP

B

L

pr

1045

0907

000706

047

RP91-3

RP

B

H

pr–p

953

0910

000690

124

RP91-4

RP

B

L, A

pr

918

0040

000785

068

RP91-5

RP

B

L

p–e

932

0893

000980

060

RP91-8

RP

B

L, A

p–e

887

0901

000817

004

Velay

Coiron 63-1

CB

B

L

pr

850

0889

000977

015

63-LU2

CB

B

L, A

pr–p

878

0896

000891

009

(continued)

2468


Vivarais



Table 2: Continued Xfo

aFeSiO3

log fO2 (FMQ)

933

0899

000798

003

985

0901

000775

004

935

0909

000732

014

1044

0902

000816

049

844

0908

000760

044

pr–p

835

0908

000685

033

pr

743

0907

000743

047

pr

822

0877

001326

011

Sample

Locality*

Type of deposit

Rock type

Texture

64-1

LP

P

L

pr

64-2

LP

P

L

pr

65-2

JC

B

L

pr–p

65-LU3

JC

B

L

pr

65-LU4

JC

B

L

pr–p

70-LU2

AV

B

L

71-2

PV

B

L

73-LU1

L

B

L

T (8C)

Aubrac

502 c.p.f.u. (Fig. 5b). This may be due to the fact that the northern domain suite includes a higher proportion of harzburgites compared with the southern domain suite. Clinopyroxene from the northern domain is characterized by a high Mg content and low Na and Ti concentrations (Fig. 5b and c; see also Downes et al., 2003). Major and minor element compositions of clinopyroxene are provided in Supplementary Data Electronic Appendix 1.

Amphibole Minor amphibole occurs in a limited number of samples (Table 2). It is homogeneous within a given sample and is Ca- and Mg-rich, and can be classified as pargasite or magnesiohastingsite following International Mineralogical Association nomenclature (Leake et al., 1997). The greatest compositional variation between samples is in their alkali contents: Na2O contents range from 265 to 423 wt % and K2O from 002 to 134 wt %. The highest K2O content was found in amphibole in sample 53A-LU6, a harzburgite from Puy de Beaunit in the northern domain. TiO2 contents also vary significantly from 129 to 307 wt % and correlate with TiO2 contents in coexisting clinopyroxene. Major element compositions of amphibole are reported in Supplementary Data Electronic Appendix 1.

Spinel Spinel composition is variable, with Cr/(Cr þAl þ Fe3þ) and Mg/(Mg þ Fe2þ) ranging from 006 to 059 and from 066 to 088, respectively (Fig. 6; Table 1). These two parameters are negatively correlated with each other, as is generally the case for suites of mantle peridotites (e.g. Fabrie's, 1979; Woodland et al., 1992, 2006). Northern domain spinels tend to have higher Cr/(Cr þAl þ Fe3þ) and slightly lower Mg/(Mg þ Fe2þ) ratios compared with spinels from the southern domain. In addition, there is a positive

relationship between Cr/(Cr þAl þ Fe3þ) in spinel and Xfo of coexisting olivine (Fig. 7), reflecting different degrees of depletion owing P to partial melting. Measured Fe3þ/ Fe varies from 016 to 046, which translates into Fe3þ contents between 004 and 019 c.p.f.u. (Table 1). Most northern domain spinels have Fe3þ contents 01 c.p.f.u. Harzburgitic samples are usually characterized by lower Fe3þ contents in spinel compared with lherzolites, owing to their higher degree of partial melting (e.g. Bryndzia & Wood, 1990). However, the harzburgites from the northern domain exhibit some of the highest Fe3þ contents (Fig. 8), indicating that they have experienced some geochemical process subsequent to the last melting event.

T H E R M O M E T RY The two-pyroxene thermometer (TBKN) of Brey & Ko«hler (1990) was used to determine equilibration temperatures, at an assumed pressure of 15 GPa. This thermometer was chosen because it is insensitive to the presence of Fe3þ in clinopyroxene and orthopyroxene (e.g. Matjuschkin et al., 2014). Therefore, we have used measured FeOtot to calculate temperatures. Our temperatures are based upon compositions obtained from the interiors of grains, which were homogeneous within a given grain. Differences in temperature calculated from core and rim compositions are generally minor and such differences have little to no impact on the results of our oxygen barometric computations (see below). Calculated temperatures vary between 700 and 12008C, with most samples lying between 850 and 9508C (Table 2). This range is similar to that reported for a suite of xenoliths from the Massif Central by Werling (1994) and Werling & Altherr (1997).

2469


*Locality abbreviations are as in Fig. 1. P, pyroclastic; B, basalt; H, harzburgitic sp-peridotite; L, lherzolitic sp-peridotite; A, amphibole-bearing peridotite; W, wehrlitic sp-peridotite; C, composite; pr, protogranular; p, porphyroclastic; cpr, complex protogranular; cp, complex porphyroclastic; e, equigranular.


VOLUME 55

NUMBER 12

DECEMBER 2014

Fig. 7. Variation in Cr# of spinel vs Xfo of coexisting olivine for samples from the northern and southern mantle domains.

Fig. 5. (a) Variations in orthopyroxene composition in terms of Al (c.p.f.u.) vs Mg# for amphibole-free and amphibole-bearing spperidotites. (b) Variation in clinopyroxene in terms of Al (c.p.f.u.) vs Mg# for samples from the northern and southern domains. (c) Na vs Ti contents (c.p.f.u.) in clinopyroxene from the northern and southern mantle domains.

OX YG E N B A RO M E T RY For spinel peridotites, fO2 can be estimated by considering the equilibrium 6Fe2 SiO4 þ O2 ¼ 3Fe2 Si2 O6 þ 2Fe3 O4

ð1Þ

Fig. 8. Fe3þ content (c.p.f.u.) in spinel vs Xfo olivine for samples from the northern and southern mantle domains. The relatively high Fe3þ content of spinels from the northern domain should be noted, although most of these samples are harzburgites.

2470


Fig. 6. Variation in spinel composition in terms of Cr# vs Mg# for samples from the northern and southern mantle domains.



DISCUSSION Variation in fO2 Variations in fO2 are often apparent between samples from different localities as well as from a single locality (Fig. 9a and b; Table 2). This observation implies that the style of volcanism (e.g. cinder cone, maar deposit or lava flow) has had no perceptible effect on the determined fO2 values of the xenoliths. In a small number of samples (eight), some compositional heterogeneity is observable between spinel grains (Table 1). However, these differences were found to have a negligible effect on the calculated log fO2, with only one sample (54A-1) yielding variations approaching the estimated uncertainties, as described above (Table 2). For the entire set of samples from the northern and southern domains, lherzolites have log fO2 values between FMQ ^ 047 and FMQ þ166 (Table 2). Harzburgites exhibit a more restricted range in log fO2, recording values4FMQ þ 05 (except sample FR10) and with many samples yielding values4 FMQ þ 09 (Table 2, Fig. 9b). Taken collectively, northern domain samples tend to record higher log fO2 values than those from the southern domain, (Fig. 9a and b).

This seems to be related to the higher proportion of harzburgites in the northern domain suite. The reason why harzburgites from both domains tend to lie at the high end of the log fO2 range must be due to metasomatic processes. Considering that harzburgites are the product of more extensive partial melting than lherzolites, it would be expected that harzburgites might yield relatively lower log fO2 values as greater amounts of Fe3þ should have been extracted (e.g. Bryndzia & Wood, 1990). However, Woodland et al. (2006) pointed out that the smaller modal amount of spinel in harzburgites means that the fO2 of such depleted compositions can be more effectively reset by metasomatic activity compared with that of lherzolites. Our new data from the Massif Central support this contention. Thus, in addition to the northern domain being characterized by more depleted bulk compositions (Lenoir et al., 2000b), it is also generally more oxidized than the southern domain, even considering the uncertainties in calculated fO2. It is noteworthy that samples from Fraisse (Fr in Fig. 9b, Table 2), which record some of the lowest log fO2 values (but are still 4FMQ) in the northern domain, are also from the oldest locality investigated in our study (20 Ma, Fig. 2; Lenoir et al., 2000a). Lherzolites of the southern domain also reveal a subtle gradient in fO2 from south to north. Samples from the southernmost volcanic fields of Aubrac and Coiron record log fO2 values mostly at the low end of the observed range, with seven out of 10 samples yielding values near FMQ (Fig. 9a, Table 2). For our limited number of samples from Cantal, which overlap in eruption age with the Coiron and Aubrac fields (Fig. 2), two of three samples record distinctly higher log fO2 values of FMQ þ1 01. Samples from the younger Deve's and Vivarais volcanic fields exhibit more variability in log fO2 values, not only between localities but also within a single locality (Fig. 9a). The texture^fO2 systematics differ between the northern and southern domains. In the north, protogranular harzburgites and lherzolites exhibit higher log fO2 values than those with porphyroclastic textures (4 FMQ þ10 compared with FMQ þ 05; Fig. 10). Thus, it appears that log fO2 values decrease with increasing degree of deformation. The small subset of samples with ‘complex’ recrystallized textures (Mercier & Nicolas, 1975) all tend to record higher log fO2 values than those with porphyroclastic textures. However, it is not clear what the timing of this recrystallization process was relative to either any metasomatic overprinting event or the resetting of oxidation state. In contrast, southern domain harzburgites and lherzolites exhibit a large variation in log fO2 from FMQ ^ 047 to FMQ þ16 with no particular relationship to textural type (Fig. 10). This is exemplified by samples from Ray Pic (Table 2). However, at Mont Coupet, where

2471


where Fe2SiO4, Fe2Si2O6 and Fe3O4 are the fayalite, ferrosilite and magnetite components in olivine, orthopyroxene and spinel, respectively. The method of calculating fO2 follows that reported by Wood et al. (1990), which uses the Nell^Wood calibration for reaction (1). Other calibrations available in the literature (e.g. O’Neill & Wall, 1987; Ballhaus et al., 1991) yield comparable results to the Nell^ Wood calibration (Wood et al., 1990; Wood,1991). To minimize the effects of uncertainties in equilibration temperature and pressure, the fO2 estimates are referenced to the fayalite^magnetite^quartz (FMQ) buffer and are reported in terms of log fO2 (see Wood, 1991). The consequence is that an uncertainty in temperature of 1008C and in pressure of 3 kbar translates to a difference in calculated fO2 of only 015 and 009 log units, respectively. Referencing fO2 values to FMQ also means that results from samples equilibrated under varying P^T conditions can be meaningfully compared. Collectively, the uncertainties from pressure, temperature and measurement of Fe3þ in spinel yield an overall uncertainty of 02^03 log units in log fO2. This is somewhat smaller than that reported in previous studies (Woodland et al., 1992, 2006) and is due to the fact that the spinels in our sample set were found to have high enough Fe3þ contents that propagating the uncertainty in the Mo«ssbauer measurements (001 in Fe3þ/ P Fe) through the calculation produces relatively smaller errors. The subset of samples for which Fe3þ in spinel was determined by microprobe using secondary standards (Table 1) has larger uncertainties that are conservatively estimated to be about 05 log units (Woodland et al., 1992).


VOLUME 55

NUMBER 12

DECEMBER 2014

we also have a reasonable number of samples, the behavior is similar to that observed in the northern domain. Thus, certain localized portions of the southern domain mantle appear to possess some relation between texture and fO2, but this appears to be the exception rather than the rule. The observed range in equilibrium temperatures obtained from our suite of xenoliths, particularly those from the same volcanic center, serves as an indicator that the rising basaltic magma has sampled the lithospheric mantle at different depths whilst en route to the surface. This is consistent with the conclusions of the extensive thermobarometric study of Werling & Altherr (1997), who suggested that the volcanism sampled the lithospheric mantle from the Moho, lying at 28^30 km (Zeyen et al.,

1997), to a depth of 70 km. Thus, our dataset reflects not only spatial variations in fO2 across the Massif Central, but also the conditions over a limited depth range, within the stability field of spinel peridotite. Unfortunately, the well-known difficulty in determining equilibration pressures for spinel peridotites compromises a detailed evaluation depth^fO2 systematics. However, we can use increasing temperature as a rough proxy for increasing depth. In the northern domain there is a tendency for samples with equilibrium temperatures 59008C to record higher log fO2 values ( FMQ þ1, Fig. 11a). This would imply somewhat more oxidizing conditions in the shallowest part of the upper mantle. In the southern domain no such correlation is apparent (Fig. 11b). Comparison

2472


Fig. 9. Variation in log fO2 for (a) lherzolites and (b) harzburgites across the Massif Central. The circles are color-coded to indicate different fO2 ranges relative to the FMQ oxygen buffer. It should be noted that one circle can represent one or more samples from a given locality.




Fig. 9. Continued.

between Fig. 11a and b provides further evidence for the northern and southern domains of the lithospheric mantle having experienced different geological histories.

Influence of metasomatism on fO2 Mantle xenoliths from the Massif Central generally exhibit evidence for having been metasomatically overprinted to various degrees, in addition to having experienced at least one partial melting event (e.g. Lenoir et al., 2000b; Downes et al., 2003; Wittig et al., 2007). Modal metasomatism, as reflected by the occurrence of disseminated secondary amphibole, is evident in some lherzolites from the southern domain volcanic fields of Deve's, Velay and Vivarais (Table 2). The log fO2 values of these samples

cluster around FMQ þ 06 03, but with some samples having lower values (Fig. 10). Although this is at the high end of the range for non-cratonic SCLM (e.g. Foley, 2011), the amphibole-bearing samples are not the most oxidized samples in our suite. In fact, at a given locality, there is essentially no difference in the range of log fO2 values between amphibole-bearing and amphibole-free samples. Ionov & Wood (1992) observed the same behavior for peridotite xenoliths from central Asia. Thus, although it is likely that the amphibole-forming process has induced changes in fO2 (e.g. Woodland et al., 1992), it appears that the presence of amphibole alone is not a direct indicator that the metasomatizing agent was strongly oxidizing. From their study of xenoliths from Marais de Limagne,

2473


VOLUME 55

NUMBER 12

DECEMBER 2014

Fig. 11. Variation of log fO2 as a function of equilibration temperature for samples from the (a) northern and (b) southern domains. The continuous and dashed lines are drawn to help visualize the distribution of the data points (see text).

2474


Fig. 10. The fO2 ranges for amphibole-free and amphibole-bearing harzburgites and lherzolites from the northern and southern domains as a function of texture (abbreviations as in Table 1).



Fig. 12. Chondrite-normalized REE patterns for clinopyroxenes from Downes et al. (2003) for several samples from Montboissier (Mb) and Fraisse (Fr) from the northern domain.

and equilibrium (1), changing the fO2 of peridotite is not a linear process, however. In their modeling of the oxidation of peridotite, Woodland et al. (2006) demonstrated that, at log fO2 values above FMQ, progressively more and more Fe3þ must be incorporated in spinel to produce even small increases in fO2. Thus, for the samples from Montboissier, a shift in log fO2 from FMQ þ10 to FMQ þ14 (i.e. samples MB4 and MB57 in Fig. 12) implies a significant addition of Fe3þ or conversion of Fe2þ to Fe3þ during metasomatic interaction, and that the metasomatizing agent must have had an oxidation state appreciably higher than that of the P ambient SCLM (this is also directly reflected in the Fe3þ/ Fe of spinel in these samples). In this context, the pre-metasomatic log fO2 of the, at the time geochemically depleted, Montboissier peridotites (Lenoir et al., 2000b), was probably much lower, potentially similar to that recorded by abyssal peridotites (FMQ ^ 10, Bryndzia & Wood, 1990). In contrast to the Montboissier xenoliths, clinopyroxene in samples FR10 and FR11 has a distinctly different REE signature (Fig. 12; Downes et al., 2003). Comparison of FR10 and FR11 indicates that stronger LREE enrichment also correlates with increasing fO2, but that the metasomatic agent was not as highly oxidative as in the MB samples. This implies that the FR samples interacted with a different type of metasomatic agent. In the southern domain, xenoliths from Ray Pic (RP) have undergone various degrees of LREE enrichment, as revealed by (Ce/Yb)N in clinopyroxene exceeding values of 05 (Zangana et al., 1997). We have determined the equilibrium fO2 for 17 of the samples of Zangana et al. and these data are plotted in Fig. 13 as a function of (Ce/Yb)N. The samples with elevated (Ce/Yb)N form two groups in terms of fO2, one with log fO2 values4FMQ þ1 and one having values around FMQ þ 05. The six samples with log fO2 values4FMQ þ1 are either harzburgites

2475


Touron et al. (2008) suggested that the modal content of amphibole increases with the degree of deformation. However, our amphibole-bearing samples do not exhibit such a relationship, as the minor amounts present (2%) occur independently of textural type, even from a single locality (Fig. 10; Table 2). The amphiboles reported by Touron et al. (2008) have rather low Ti contents (04^16 wt %), similar to or even lower than that found in our samples (see above). Thus, there is no evidence for an Fe^Titype metasomatism (Menzies & Hawkesworth, 1987) having occurred beneath the Massif Central. Another type of modal metasomatism is represented by two samples from Ray Pic, a wehrlite (RP 87-7) and a composite xenolith (RP91-18) containing clinopyroxene-rich patches (Zangana et al., 1997). Both of these samples record log fO2 values4FMQ þ1 (Table 2), indicating that this type of modal metasomatism is more oxidizing than that related to amphibole formation. Green clinopyroxene also occurs in veins with sharp contacts in samples RP 83-71 and RP 87-6 (Zangana et al., 1999). The log fO2 value of the former also lies well above FMQ þ1, but the value for RP 87-6 of FMQ þ 06 is distinctly lower (Table 2). The reason for this marked difference is not clear. Cryptic metasomatic enrichment of LREE and other highly incompatible elements has been identified in xenoliths from both domains (Zangana et al., 1997; Lenoir et al., 2000b; Downes et al., 2003). In the northern domain, trace element signatures include negative normalized anomalies in Zr, Hf, Nb and Ta, which led both Lenoir et al. (2000b) and Downes et al. (2003) to conclude that the metasomatic agent was a carbonate melt. However, in their study of peridotite xenoliths from eastern Oman, Gre¤goire et al. (2009) interpreted similar geochemical signatures to reflect metasomatism by a migrating CO2-rich alkaline mafic silicate melt. The nature of the metasomatizing agent(s) (i.e. carbonate vs silicate melt) will be discussed in a companion publication in which we present trace element compositions of coexisting clinopyroxenes from our sample suite (Uenver-Thiele et al., in preparation). Here we provide a short discussion of how cryptic metasomatism may have affected the oxidation state of the SCLM. Equilibrium fO2 values have been determined for nine samples from the study of Downes et al. (2003) and are given in Table 2. Clinopyroxene in four samples from Montboissier (MB), two lherzolites and two harzburgites, exhibits different degrees of LREE enrichment and all samples have log fO2 values FMQ þ10 (Fig. 12). In fact, samples with stronger LREE enrichment systematically exhibit higher log fO2 values, even if these values overlap somewhat when uncertainties are considered. This suggests that the metasomatic agent must have been strongly oxidizing and that even relatively small degrees of peridotite^‘melt’ interaction can reset the oxidation state. Owing to the thermodynamics of spinel


VOLUME 55

NUMBER 12

DECEMBER 2014

or those that have experienced modal addition of clinopyroxene (Fig. 13). The less oxidized group is primarily composed (but not uniquely) of lherzolites and is characterized by equilibration temperatures 49808C (Table 2), suggesting that these xenoliths may originate from a somewhat deeper level (see discussion above). The wehrlite (RP87-7) also records a high temperature, but this may be due to local heating during extensive peridotite^‘melt’ interaction. The fact that the log fO2 values of the two groups remain relatively constant with increasing (Ce/ Yb)N suggests that the oxidation state was reset by even small degrees of peridotite^‘melt’ interaction. This is consistent with our observations at Montboissier described above. Thus, metasomatic processes causing variable degrees of LREE enrichment seem to have imposed somewhat different oxidation states at different levels in the SCLM beneath Ray Pic. Whether or not the same metasomatizing agent was involved at these different depths remains an open question, which will require further geochemical data to answer.

Oxidation state of the SCLM Numerous studies have demonstrated that the Earth’s upper mantle does not have a homogeneous oxidation state and that it can vary by many orders of magnitude at various scales (e.g. Wood et al., 1990; Ballhaus & Frost, 1994; Woodland et al., 2006; Frost & McCammon, 2008; Foley, 2011). In his compilation of literature data for noncratonic SCLM, Foley (2011) gave a range of over 5 log units in fO2 with an average value of log fO2 ¼ FMQ ^ 068 (Fig. 14). This range reflects the superimposed effects of partial melting events and various types and degrees of

Fig. 14. Histograms illustrating the ranges in log fO2 for (a) a global compilation of non-cratonic lithospheric mantle (Foley, 2011) and (b) Massif Central peridotite xenoliths from this study.

metasomatic overprinting that the SCLM peridotites have experienced. Our investigation of the SCLM beneath the Massif Central confirms that oxidation state varies both spatially and with depth. However, our log fO2 values lie consistently at the high end of the global distribution, with a much narrower range (Fig. 14). Thus, although heterogeneities in oxidation state exist at different scales (Fig. 9), there appears to be a general tendency towards oxidation (log fO24FMQ þ 05) coupled with geochemical re-enrichment of the SCLM over time, as migrating melts and fluids progressively interact with the peridotites. The extent of oxidation will also be controlled by the oxygen fugacity of the invading metasomatic agent. The most affected parts of the lithosphere will be

2476


Fig. 13. Variation of (Ce/Yb)N in clinopyroxene vs equilibrium log fO2 for samples from Ray Pic. Trace element data are from Zangana et al. (1997) and have been normalized to C1 chondrite (McDonough & Sun, 1995). Equilibration temperatures are given for selected samples. A typical error bar is shown for reference. H, harzburgite; M, samples with modal addition of clinopyroxene; no label, lherzolites.



particularly susceptible to partial melting at a later point in time if heat flow increases (e.g. from interaction with a plume or a change in tectonic regime, e.g. rifting). The resulting melts will have particular geochemical signatures, reflecting this earlier history (e.g. Foley, 2008; Shaw & Woodland, 2012).

CONC LU DI NG R E M A R K S

AC K N O W L E D G E M E N T S Alain Gourgaud, Pierre Boivin, Jacques Kornprobst and Marie-Claude Kornprobst are thanked for their logistical help with fieldwork. This work benefited from discussions with Gerhard Brey, Armin Zeh and Kevin Klimm. Ilona Fin and Oliver Wienand are thanked for their help in the preparation of excellent thin sections. The paper was improved by comments from Chris Ballhaus, Michel Gre¤goire and an anonymous reviewer.

FU N DI NG This work was supported by a grant from the Deutsche Forschungsgemeinschaft to A.B.W. (Wo652/15-1).

S U P P L E M E N TA RY DATA Supplementary data for this paper are available at Journal of Petrology online.

R EF ER ENC ES Ballhaus, C. (1993). Redox states of lithospheric and asthenospheric upper mantle. Contributions to Mineralogy and Petrology 114, 331^348. Ballhaus, C. & Frost, B. R. (1994). The generation of oxidized CO2bearing basaltic melts from reduced CH4-bearing upper-mantle sources. Geochimica et Cosmochimica Acta 58, 4931^4940. Ballhaus, C., Berry, R. F. & Green, D. H. (1991). High pressure experimental calibration of the olivineôrthopyroxene^spinel oxygen geobarometer: implications for the oxidation state of the upper mantle. Contributions to Mineralogy and Petrology 107, 27^40. Boivin, P., Besson, J.-C., Briot, D., Gourgaud, A., Labazuy, P., Langlois, E., de Larouzie're, F.-D., Livet, M., Mergoil, J., Miallier, D., Morel, J.-M., Vernet, G. & Vincent, P. (2009). Volcanologie de la Cha|Œne des Puys. Edition du Parc Naturel Re¤gional des Volcans d’Auvergne. Brey, G. P. & Ko«hler, T. (1990). Geothermobarometry in four-phase lherzolites II. New thermobarometers, and practical assessment of existing thermobarometers. Journal of Petrology 31, 1353^1378. Briot, D. & Harmon, R. S. (1989). Relationships between isotopes and geological settings in volcanics from the Massif Central. Terra Abstracts 1, 343. Bryndzia, L. T. & Wood, B. J. (1990). Oxygen thermobarometry of abyssal spinel peridotites: the redox state and CÔ^H volatile composition of the Earth’s sub-oceanic upper mantle. American Journal of Science 290, 1093^1116. Canil, D. & O’Neill, H. St. C. (1996). Distribution of ferric iron in some upper-mantle assemblages. Journal of Petrology 37, 609^635. Chen, Y. D., Pearson, N. J., O’Reilly, S. Y. & Griffin, W. L. (1991). Applications of olivineôrthopyroxene^spinel oxygen geobarometers to the redox state of the upper mantle. Journal of Petrology , Special Lherzolites Issue, 291^306. Downes, H. & Dupuy, C. (1987). Textural, isotopic and REE variations in spinel peridotite xenoliths, Massif Central (France). Earth and Planetary Science Letters 82, 121^135.

2477


Our study demonstrates the utility of using oxidation state to help characterize different domains in the SCLM. Our data reveal a different character between northern and southern domains beneath the Massif Central, consistent with the geochemical differences identified by Lenoir et al. (2000b), Downes et al. (2003) and Wittig et al. (2007). Although the northern domain is generally more refractory in bulk composition than the southern domain (Lenoir et al., 2000b), it tends to record more oxidized conditions. This is in part due to the fact that such harzburgitic compositions are more sensitive to resetting of their oxidation state than lherzolites. For the most part, the lack of a clear-cut relationship between fO2 and texture is a reflection of subsequent metasomatic processes that have overprinted the SCLM beneath the Massif Central. Although rock texture should be expected to have an influence on the style and intensity of metasomatic interactions and thus fO2, any systematic relationship is most likely to be observable at a local rather than a regional scale. Considering increasing equilibration temperatures among xenoliths to reflect increasing depths of origin, samples from the northern domain suggest that the shallower part of the SCLM is somewhat more oxidized than that at deeper levels. On the other hand, such a general observation cannot be made for the southern domain, although it may hold true at a local scale. Metasomatic processes can clearly influence oxidation state, with a tendency to increase log fO2 values compared with ‘normal’ non-cratonic SCLM (Fig. 14). In terms of modally metasomatized xenoliths, the ‘melt’ leading to the addition of clinopyroxene apparently had a higher oxidation state (log fO24FMQ þ1) than the agent responsible for crystallization of amphibole (log fO2 FMQ þ 06). Cryptic metasomatism can also reset oxidation state, sometimes very effectively if harzburgitic compositions are involved. It is important to note that metasomatizing agents do not have an infinite capacity to cause oxidation and that different compositions of invading ‘melts’ are expected to possess different oxidation states. On the other hand, metasomatic processes are no doubt the reason why our xenolith suite from the Massif Central records relatively high log fO2 values compared with the Foley (2011) global compilation of oxidation state for non-cratonic SCLM. Thus, data on oxidation state can provide further constraints on geochemical processes

operating in the SCLM at various length scales from regional to local.


VOLUME 55

DECEMBER 2014

Nehlig, P., Boivin, P., de Goe«r, A., Mergoil, J., Prouteau, G., Sustrac, G. & Thie¤blemont, D. (2003). Les Volcans du Massif Central. Revue Ge¤ ologues, special issue Massif Central, 1^41. Nicolas, A., Lucazeau, F. & Bayer, R. (1987). Peridotite xenoliths in Massif Central basalts: textural and geophysical evidence for asthenospheric diapirism. In: Nixon, P. H. (ed.) Mantle Xenoliths. John Wiley, pp. 563^574. O’Neill, H. S. C. & Wall, V. J. (1987). The olivineôrthopyroxene^ spinel oxygen geobarometer, the nickel precipitation curve, and the oxygen fugacity of the Earth’s upper mantle. Journal of Petrology 28, 1169^1191. Shaw, C. S. J. & Woodland, A. B. (2012). The role of magma mixing in the petrogenesis of mafic alkaline lavas, Rockeskyllerkopf Volcanic Complex, West Eifel, Germany. Bulletin of Volcanology 74, 359^376. Sobolev, S. V., Zeyen, H., Granet, M., Stoll, G., Achauer, U., Bauer, C., Werling, F., Altherr, R. & Fuchs, K. (1997). Upper mantle temperatures and lithosphereâsthenophere system beneath the French Massif Central constrained by seismic, gravity, petrologic and thermal observations. Tectonophysics 275, 143^164. Touron, S., Renac, C., O’Reilly, S. Y., Cottin, J.-Y. & Griffin, W. L. (2008). Characterization of the metasomatic agent in mantle xenoliths from Deve's, Massif Central (France) using coupled in situ trace-element and O, Sr and Nd isotopic compositions. In: Coltorti, M. & Gre¤goire, M. (eds) Metasomatism in Oceanic and Continental Lithospheric Mantle. Geological Society, London, Special Publications 293, 177^196. von Seckendorf, V. & O’Neill, H. St. C. (1993). An experimental study of Fe^Mg partitioning between olivine and orthopyroxene at 1173, 1273 and 1473 K and 16 GPa. Contributions to Mineralogy and Petrology 113, 196^207. Werling, F. (1994). Die thermische Entwicklung des Lithospha«rischen Mantels im Bereich des Franzo«sischen Riftsystems-abgeleitet aus der Mineralchemie von Mantelxenolithen, PhD thesis, Universita«t Karlsruhe. Werling, F. & Altherr, R. (1997). Thermal evolution of the lithosphere beneath the French Massif Central as deduced from geothermobarometry on mantle xenoliths. Tectonophysics 275, 119^141. Wilson, M. & Downes, H. (1991). Tertiary^Quaternary extensionrelated alkaline magmatism in western and central Europe. Journal of Petrology 32, 811^849. Wittig, N., Baker, J. A. & Downes, H. (2006). Dating the mantle roots of young continental crust. Geology 34, 237^240. Wittig, N., Baker, J. A. & Downes, H. (2007). U^Th^Pb and Lu^Hf isotopic constraints on the evolution of sub-continental lithospheric mantle, French Massif Central. Geochimica et Cosmochimica Acta 71, 1290^1311. Wood, B. J. (1991). Oxygen barometry of spinel peridotites. In: Lindsley, D. H. (ed.) Oxide Minerals. Mineralogical Society of America, Reviews in Mineralogy 25, 417^431. Wood, B. J. & Virgo, D. (1989). Upper mantle oxidation state: ferric iron contents of lherzolite spinels by 57Fe Mo«ssbauer spectroscopy and resultant oxygen fugacities. Geochimica et Cosmochimica Acta 53, 1277^1291. Wood, B. J., Bryndzia, L. T. & Johnson, K. E. (1990). Mantle oxidation state and its relationship to tectonic environment and fluid speciation. Science 248, 337^345. Woodland, A. B., Komprobst, J. & Wood, B. J. (1992). Oxygen thermobarometry of orogenic lherzolite massifs. Journal of Petrology 33, 203^230. Woodland, A. B., Kornprobst, J., McPherson, E., Bodinier, J.-L. & Menzies, M. A. (1996). Metasomatic interactions in the lithospheric mantle: Petrologic evidence from the Lherz Massif, French Pyrenees. Chemical Geology 134, 83^112.

2478


Downes, H., Reichow, M. K., Mason, P. R., Beard, A. D. & Thirlwall, M. (2003). Mantle domains in the lithosphere beneath the French Massif Central: trace element and isotopic evidence from mantle clinopyroxenes. Chemical Geology 200, 71^87. Fabrie's, J. (1979). Spinelôlivine geothermometry in peridotites from ultramafic complexes. Contributions to Mineralogy and Petrology 69, 329^336. Foley, S. F. (2008). Rejuvenation and erosion of the cratonic lithosphere. Nature Geoscience 1, 503^510. Foley, S. F. (2011). A reappraisal of redox melting in the Earth’s mantle as a function of tectonic setting and time. Journal of Petrology 52, 1363^1391. Frost, D. J. & McCammon, C. A. (2008). Redox state of the Earth’s mantle. Annual Review of Earth and Planetary Sciences 36, 389^420. Granet, M., Wilson, M. & Achauer, U. (1995). Imaging a mantle plume beneath the French Massif Central. Earth and Planetary Science Letters 136, 281^296. Gre¤goire, M., Langlade, J. A., Delpech, G., Dantas, C. & Ceuleneer, G. (2009). Nature and evolution of the lithospheric mantle beneath the passive margin of East Oman: Evidence from mantle xenoliths sampled by Cenozoic alkaline lavas. Lithos 112, 203^216. Ionov, D. A. & Wood, B. J. (1992). The oxidation state of subcontinental mantle: oxygen thermobarometry of mantle xenoliths from central Asia. Contributions to Mineralogy and Petrology 111, 179^193. Leake, B. E., Woolley, A. R., Arps, C. E. S. et al. (1997). Nomenclature of amphiboles: report of the subcommittee on amphiboles of the International Mineralogical Association, commission on new minerals and mineral names. Canadian Mineralogist 35, 219^246. Lenoir, X., Dautria, J. M., Briqueu, L., Cantagrel, J.-M. & Michard, A. (2000a). Nouvelles donne¤es ge¤ochronoloques, ge¤ochimiques et isotopiques sur le volcanisme du Forez: relation avec l’evolution ce¤nozoı¨ que du manteaux du Massif Central. Earth and Planetary Sciences 330, 201^207. Lenoir, X., Garrido, C. J., Bodinier, J. L. & Dautria, J. M. (2000b). Contrasting lithospheric mantle domains beneath the Massif Central (France) revealed by geochemistry of peridotite xenoliths. Earth and Planetary Science Letters 181, 359^375. Matjuschkin, V., Brey, G. P., Ho«fer, H. E. & Woodland, A. B. (2014). The influence of Fe3þ on garnetôrthopyroxene and garnetôlivine geothermometers. Contributions to Mineralogy and Petrology 167, doi: 10.1007/s00410-014-0972-z. Mattioli, G. S., Baker, M. B., Rutter, M. J. & Stolper, E. M. (1989). Upper mantle oxygen fugacity and its relation to metasomatism. Journal of Geology 97, 521^536. Maury, R. C. & Varet, J. (1980). Le volcanisme tertiaire et quaternaire en France. In: Autran, A. & Dercourt, J. (eds) Evolutions ge¤ ologiques de la France. Me¤ moire du BRGM 107, 137^159. McDonough, W. F. & Rudnick, R. L. (1998). Mineralogy and composition of the upper mantle. In: Hemley, R. J. (ed.) Ultrahigh-Pressure Mineralogy. Mineralogical Society of America, Reviews in Mineralogy 37, 139^164. McDonough, W. F. & Sun, S. S. (1995). The composition of the Earth. Chemical Geology 120, 223^253. Menzies, M. A. & Hawkesworth, C. J. (1987). Mantle Metasomatism. Academic Press. Mercier, J.-C. C. & Nicolas, A. (1975). Textures and fabrics of upper mantle peridotites as illustrated by xenoliths from basalts. Journal of Petrology 16, 454^487. Merle, O. & Michon, L. (2001). The formation of the West European rift: A new model as exemplified by the Massif Central area. Bulletin de la Societe¤ Ge¤ ologique de France 172, 213^221.

NUMBER 12



Woodland, A. B., Kornprobst, J. & Tabit, A. (2006). Ferric iron in orogenic lherzolite massifs and controls of oxygen fugacity in the upper mantle. Lithos 89, 222^241. Yoshikawa, M., Kawamoto, T., Shibata, T. & Yamamoto, J. (2010). Geochemical and Sr^Nd isotopic characteristics and pressure^temperature estimates of mantle xenoliths from the French Massif Central: evidence for melting and multiple metasomatism by silicate-rich carbonatite and asthenospheric melts. In: Coltorti, M., Downes, H., Gre¤goire, M. & O’Reilly, S. Y. (eds) Petrological Evolution of the European Lithospheric Mantle. Geological Society, London, Special Publications 337, 153^175. Zangana, N. (1995). Geochemical variations in mantle xenoliths from Ray Pic, Massif Central, France, PhD thesis, University of London.

Zangana, N., Downes, H., Thirlwall, M. F. & Hegner, E. (1997). Relationship between deformation, equilibration temperatures, REE and radiogenic isotopes in mantle xenoliths (Ray Pic, Massif Central, France): an example of plume^lithosphere interaction? Contributions to Mineralogy and Petrology 127, 187^203. Zangana, N., Downes, H., Thirlwall, M. F., Marriner, G. F. & Bea, F. (1999). Geochemical variation in peridotite xenoliths and their constituent clinopyroxenes from Ray Pic (French Massif Central): implications for the composition of the shallow lithospheric mantle. Chemical Geology 153, 11^35. Zeyen, H., Novak, O., Landes, M., Prodehl, C., Driad, L. & Him, A. (1997). Refraction-seismic investigations of the Northern Massif Central (France). Tectonophysics 275, 99^117.


2479