U–Pb ages, Sr–Nd- isotope geochemistry, and petrogenesis of kimberlites, kamafugites and phlogopite-picrites of the Alto Paranaíba Igneous Province, Brazil.pdf

anec,d

Ceocanese Academy of Sciences, P.O. Box 9825, Beijng 100029, China

degli Stu(IGAG),

e Instituto de Geocincias, Universidade de So Paulo, Rua

a r t i c l e i n f o

Article history:Received 16 January 2012Received in revised form 5 June 2012

Chemical Geology xxx (2012) xxxxxx

CHEMGE 16577; No of Pages 18

Contents lists available at SciVerse ScienceDirect

Chemical

l seexplained with the presence of chemically and mineralogically heterogeneous mantle sources that meltedat different pressures.

2012 Elsevier B.V. All rights reserved.

1. Introduction

The Alto Paranaba Igneous Province (hereafter APIP) extendsover an area of ~20,000 km2 in southeastern Brazil (Fig. 1). This province comprises Late Cretaceous strongly alkaline (kamafugites, kimberlites, lamprophyres) and carbonatitic rocks found as lava ows,pyroclastic successions and hypabyssal intrusions (Gibson et al.,1995; Brod et al., 2000; Arajo et al., 2001; Read et al., 2004; Gomes

minerals and elements for industrial purposes, as for examplediamonds in Trs Ranchos kimberlites, or P, Nb, Ti, REE elements asresidual and supergene enrichment over the Catalo I and II rocks(Biondi, 2005), moved the interest of a large number of researchers,who proposed different models to explain the origin and thepetrological evolution of the igneous rocks of this area. The APIPmagmatism has been related to asthenospheric or lithospheric melting caused by passage of the South American platform over a hotspotand Comin Chiaramonti, 2005; Melluso et alalkaline magmatism of APIP developed alo(Bardet, 1977; Tompkins, 1991; Biondi, 20their economic potential in terms of high

Corresponding author. Fax: +39 0812538118.E-mail address: [email protected] (L. Melluso).

0009-2541/$ see front matter 2012 Elsevier B.V. Alldoi:10.1016/j.chemgeo.2012.06.016

Please cite this article as: Guarino, V., et aphlogopite picrites of the Alto Paranaba Ig(f=0.5 2%) of such a source can produce compositions resembling the APIP rocks. The geochemical andisotopic composition of the magmas, the calculated degrees of partial melting, the composition of the calculated source and the absence of a hot spot track from Goias to Alto Paranaba igneous provinces can beAccepted 24 June 2012Available online xxxx

Keywords:KimberlitesKamafugitesPhlogopite-picritesUPb perovskite geochronologySrNd perovskite isotopesBrazilCNR, c/o Dipartimento di Scienze della Terra, Universit degli Studi di Roma La Sapienza, P.le A. Moro, 5, 00185, Roma, Italydo Lago 562, Cidade Universitria, 05508 So Paulo, Brazil

a b s t r a c t

The kimberlites, kamafugites and phlogopite picrites of Alto Paranaba Igneous Province (APIP), southernBrazil, span a range between ~91 and 78 Ma with new in situ, more tightly constrained U Pb ages. Thekimberlites show porphyritic texture with olivine xeno and phenocrysts plus phlogopite, Fe Ti Cr oxidesand perovskite microcrysts set in a carbonate rich matrix. The kamafugites are feldspar free rocks represented by ugandites and mafurites. Ugandites and mafurites are porphyritic and contain olivine and clinopyroxene phenocrysts in a ne grained groundmass composed by clinopyroxene, perovskite, apatite,magnetite, phlogopite, leucite and/or analcime in ugandites, and by olivine, clinopyroxene, amphibole,phlogopite, perovskite, magnetite, kalsilite and Ba zeolites in mafurites. Phlogopite picrites have apseudo uidal porphyritic texture with olivine phenocrysts and abundant phlogopite microcrysts in agroundmass composed by olivine, spinel, apatite, perovskite, calcite and rare garnet. New in situ Sr andNd isotopic data on perovskites (87Sr/86Sri=0.70467 0.70565 and 143Nd/144Ndi=0.51222 0.51233) fallwithin the known ranges of APIP rocks (87Sr/86Sri=0.70431 0.70686; 143Nd/144Ndi=0.51205 0.51280).The APIP magmas derived from a source assemblage made up of an old metasomatized mica carbonategarnet lherzolite, that did not suffer interaction with convective mantle, nor with any hypotheticalmelts derived from a Trindade mantle plume. Geochemical modelling show that low degree meltingc Dipartimento di Scienze della Terra, Universitd Istituto di Geologia Ambientale e Geoingegneriadi di Roma La Sapienza, P.le A. Moro, 5, 00185 Roma, ItalyDipartimento di Scienze della Terra, Universit degli Studi di Napoli Federico II, via Mezzb State Key Laboratory of Lithospheric Evolution, Institute of Geology and Geophysics, ChinUPb ages, SrNd- isotope geochemistry,kamafugites and phlogopite-picrites of th

Vincenza Guarino a, Fu-Yuan Wu b, Michele LustrinoCelso de Barros Gomes e, Excelso Ruberti e, Colomboa

j ourna l homepage: www.e., 2008). The Cretaceousng a NW SE lineament05). These rocks, withpercentage of precious

rights reserved.

l., U Pb ages, Sr Nd isotopneous Province, Brazil, Chemd petrogenesis of kimberlites,Alto Paranaba Igneous Province, Brazil

, Leone Melluso a,, Pietro Brotzu a,lso Gaeta Tassinari e, Darcy Pedro Svisero e

none 8, 80134 Napoli, Italy

Geology

vie r .com/ locate /chemgeo(Trindade Plume, Gibson et al., 1995; Thompson et al., 1998; Carlsonet al., 2007; Bulanova et al., 2010). Such a mantle plume would havebeen successively channelled towards SE, in areas characterized byrelatively thinned lithosphere, to form the Serra do Mar Igneous Province and the Trindade Martin Vaz volcanic lineament (Thompson etal., 1998). This plume model is essentially based upon a supposedtrack of igneous activity, with ages decreasing from NW (APIP) tothe SE (the easternmost sector of the Serra do Mar; Thompson et

e geochemistry, and petrogenesis of kimberlites, kamafugites and. Geol. (2012), doi:10.1016/j.chemgeo.2012.06.016

2 V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxal., 1998; Bennio et al., 2002, 2003; Brotzu et al., 2005, 2007), butwithout strong geochemical and/or petrological constraints.

We present new in situ U Pb ages and Sr Nd isotopic ratios onperovskites. New bulk rock Sr Nd isotopic ratios are also reportedfor phlogopite picrite rocks of Catalo II complex. In addition, we expand the data on the mineral chemical and bulk rock major and traceelement data of APIP rocks from the districts of Limeira, Lemes,Indai, Pantano, Presidente Olegario, Faco, So Jos dos Talhados,Osmar, Serra, Serra do Bueno, Velosa and Catalo II (Fig. 1), somereported here for the rst time, thus completing and improving thedata set of Melluso et al. (2008).

2. Geological setting and previous age determinations

The APIP is a large igneous province located across SE MinasGerais and SW Gias on a late Precambrian mobile belt (the BrasiliaBelt), between the NE margin of the Paran Basin and the SW marginof the So Francisco Craton (Almeida et al., 2000; D'Agrella Filho etal., 2011; Peucat et al., 2011, and references therein). The Late Cretaceous Cenozoic igneous rocks of southeastern Brazil are alignedalong two main trends: the rst is oriented along NW SE (APIP),whereas the second trends W E (Serra do Mar Igneous Province).The mac potassic to ultrapotassic magmatism of APIP consists of different igneous forms including plugs, dykes, lava ows, pipes, pyroclastic deposits and plutonic complexes. The most continuous andextensive cover of lava ows and tuff beds of the APIP is the Matada Corda formation, extending over an area of ~8000 km2 and anaverage thickness of ~100 120 m (Leonardos et al., 1991; Sgarbi etal., 2000; Read et al., 2004).

Fig. 1. Localization of APIP complexes studied, southern Brazil (modied after Melluso et al.,we are reported our new UPb perovskite data ages (in bold) and the data ages taken by: So(IV); Gibson et al., 1995 (V).

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, ChemThese complexes have been thoroughly investigated by means ofgeochronology. Sonoki and Garda (1988) obtained K Ar ages on Catalo I (85 Ma, whole rock), Pantano (79.7 Ma, whole rock), SerraNegra (83.4 83.7 Ma, biotite separates), Salitre (82.5 86.3 Ma, biotite separates and 79.0 94.5 Ma, whole rock), Arax (77.4 97.6 Ma,biotite separates) and Tapira (71.2 87.2 Ma, biotite separates);Machado (1991) reported a Rb Sr age of 83 Ma for the Catalo II igneous complex; Gibson et al. (1994) reported a 40Ar/39Ar ages for Serrado Bueno (904 Ma on groundmass); Sgarbi et al. (2004) dened UPb perovskite ages for the Mata da Corda kamafugites (75 81 Ma)and Gibson et al. (1995) reported K Ar ages of Carmo do Paranaba(750 and 83.61.4 Ma on phlogopite separates).

3. Analytical techniques

Major and trace elements (Sc, V, Cr, Ni, Cu, Zn, Rb, Sr, Y, Zr, Nb, Ba)of APIP rocks were obtained using a Philips PW1400 X ray uorescence spectrometer at C.I.S.A.G., University of Napoli Federico II, following methods described by Melluso et al. (2005). Major elementsfor Catalo II rocks are obtained with ICP OES methods at Actlabs(Canada). Lanthanides (REE) and other trace elements from theAPIP rocks have been analysed using ICP MS at Actlabs (Canada).Mineral analyses were performed at C.I.S.A.G., University of NapoliFederico II, utilizing an Oxford Instruments Microanalysis Unit: aJEOL JSM 5310 electron microscope equipped with an INCAx actEDS detector operating at 15 kV accelerating voltage, 50 100 nAbeam current and variable spot size. Details of standards are providedin Melluso et al. (2010). Bulk rock Sr and Nd isotope analyses wereperformed at the Geochronological Research Center, University of

2008). In this gures are summarized all ages referred of the APIP complex, specicallynoki and Garda, 1988 (I); Machado, 1991 (II); Gibson et al., 1994 (III); Sgarbi et al., 2004


able 1ew major (wt.%) and some trace elements (ppm) analyses of the APIP rocks.

Rock Geographiccoordinates

Locality References Sample SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 LOI sum Mg# K2O/Na2O

C.I. Sc V Cr Ni Cu Zn Rb Sr Y Zr Nb Ba

Kimberlite 1830 S4729 W

Limeira This work LIMI-3 XRF 28.4 2.0 2.2 11.6 0.2 27.5 12.7 1.0 1.2 2.3 7.9 97.2 84 1.1 1.1 30 186 1449 1278 59 85 117 2389 37 578 186 3325


Lemes This work LEM XRF 29.2 2.5 2.0 10.7 0.2 27.6 11.0 0.1 1.0 2.7 10.8 97.8 83 11.4 1.1 18 181 1513 1288 59 89 108 2291 36 609 178 2860


Indiai This work IN I-1 XRF 26.7 2.3 2.5 11.4 0.2 23.6 14.8 0.3 1.5 3.1 10.5 96.9 82 4.3 1.1 30 193 1315 1116 61 92 57 2417 38 705 188 3420

Kimberlite Indiai This work IN I-6 XRF 25.3 2.3 2.5 11.7 0.2 23.4 15.4 0.4 0.8 3.3 10.0 95.4 82 2.0 1.1 26 157 1224 1130 62 105 53 2376 38 715 204 4092Kimberlite 1835 S

4650 WPantano This work PANT-1 XRF 32.3 4.0 3.1 13.5 0.2 20.8 12.0 1.4 2.6 1.0 6.9 97.9 78 1.9 1.4 28 115 1428 945 131 96 144 1673 21 681 240 3658


Trs Ranchos Melluso et al.(2008)

TR-14 XRF 31.9 0.9 2.0 9.7 0.2 27.0 8.3 0.9 2.2 2.7 11.4 97.1 86 2.5 1.1 20 269 1707 1498 43 58 141 2756 19 266 223 6556


Limeira Melluso et al.(2008)

LIMI-1 XRF 27.9 2.0 2.2 11.6 0.2 27.8 12.5 0.2 1.4 2.4 10.4 98.8 84 6.0 1.0 21 189 1456 1283 57 86 111 2271 35 544 173 3225


Indai Melluso et al.(2008)

IN I-9 XRF 25.3 2.3 2.4 11.7 0.2 23.4 15.4 0.5 0.8 3.2 9.9 95.3 82 1.6 1.1 25 201 1248 1111 63 96 42 2251 38 657 206 4502


Pantano Melluso et al.(2008)

PANT-2 XRF 32.7 3.9 2.9 13.4 0.2 21.9 11.7 1.1 3.0 0.9 6.8 98.5 79 2.8 1.3 35 120 1383 929 125 94 155 1726 28 745 244 3568

Kamafugite 1943 S4603 W

Serra do Bueno This work SDB XRF 40.7 4.0 6.8 12.2 0.2 13.0 10.6 2.5 2.6 1.1 3.8 97.5 68 1.0 18 271 494 399 83 118 95 1815 34 516 119 2151


Serra This work SER XRF 40.1 4.9 6.9 11.2 0.2 10.7 12.6 1.3 1.9 0.7 5.2 95.7 68 1.5 22 306 258 139 95 84 151 1517 22 531 171 33427


So Jos dosTalhados

This work SJdT XRF 41.1 3.7 6.4 12.6 0.2 15.1 10.6 2.9 1.2 1.2 4.2 99.1 73 0.4 21 275 1042 610 78 109 125 1638 37 370 140 3598


Malaquias Melluso et al.(2008)

MAL XRF 38.1 5.1 6.6 11.5 0.2 13.5 12.9 1.2 2.9 0.9 5.1 97.8 73 2.5 26 357 548 325 130 91 115 1677 23 464 188 12382

Kamafugite 1841 S476W

Santa Rosa Melluso et al.(2008)

S.R. XRF 40.5 3.7 6.4 12.8 0.2 14.7 10.3 2.7 1.3 1.3 5.0 98.9 72 0.5 21 279 952 588 83 106 163 1591 26 434 137 3849


Veridiana Melluso et al.(2008)

VER XRF 42.2 3.6 6.8 13.7 0.2 14.6 10.6 2.8 1.2 1.3 3.6 100.6 71 0.4 22 256 820 519 76 115 161 1703 39 382 141 3153


Canas Melluso et al.(2008)

CAN XRF 42.6 3.7 7.1 13.6 0.2 14.0 10.1 2.2 1.4 1.2 5.5 101.6 70 0.6 21 177 785 520 81 116 208 1556 35 393 135 3790


Osmar This work OS XRF 33.7 6.4 5.1 15.9 0.2 15.1 10.1 0.4 3.0 0.9 6.1 97.0 71 7.3 29 196 1018 619 150 107 324 1792 26 594 222 12059


Faco This work FAC XRF 38.4 3.7 4.9 12.3 0.2 20.4 9.7 0.4 1.2 0.5 5.4 97.0 76 2.7 16 241 1178 817 93 100 120 1264 23 373 121 17804


Velosa This work V-1 XRF 35.9 5.1 6.5 11.4 0.3 10.4 11.0 1.4 2.7 1.0 6.0 91.6 67 2.0 23 309 205 146 118 113 171 2050 32 513 172 34409


PresidenteOlegario

This work PO2 XRF 38.8 3.6 4.8 12.5 0.2 20.5 10.3 1.0 1.4 0.4 5.1 98.7 79 1.3 24 278 1302 891 107 90 127 1250 19 427 133 10624

Kamafugite PresidenteOlegario

Melluso et al.(2008)

PO3 XRF 38.9 3.6 5.0 12.6 0.2 21.4 10.3 1.5 1.5 0.5 6.0 101.5 79 1.0 22 282 1320 871 100 87 148 1301 22 482 138 6856

Phl-Picrite 1802 S4752 W

Catalo II This work C2A8 ICP-OES 30.7 6.2 3.9 15.6 0.2 16.6 10.2 0.3 3.3 2.8 10.3 100.0 70 12.6

Phl-Picrite Catalo II This work C2A10 ICP-OES 29.3 6.4 4.7 15.7 0.2 14.4 11.2 0.3 4.6 3.4 9.2 99.1 67 16.3Phl-Picrite Catalo II This work C2A18 ICP-OES 30.2 4.6 4.5 12.4 0.2 18.7 11.2 0.1 4.5 1.7 10.9 99.0 77 32.0Phl-Picrite Catalo II This work C2B23 ICP-OES 31.7 6.5 4.1 15.4 0.2 15.1 8.0 0.4 4.5 1.7 10.1 97.6 69 12.9

g#=molar Mg*100/(Mg+Fe2+); Phl-Picrite=Phlogopite-picrite; C.I. =Contamination Index=(SiO2+Al2O3+Na2O)/(MgO+2*K2O), Clement, 1982.

3V.G

uarinoet

al./Chem

icalGeology

xxx(2012)

xxxxxx

Pleasecite

thisarticle

as:Guarino,

V.,et

al.,UPb

ages,SrN

d-isotope

geochemistry,

andpetrogenesis

ofkim

berlites,kam

afugitesand

phlogopite-picritesof

theAlto

ParanabaIgneous

Province,Brazil,Chem.G

eol.(2012),doi:10.1016/j.chemgeo.2012.06.016TN

M

Table 2ICP-MS trace element contents (ppm) of APIP rocks.

Rock Kimberlite Kimberlite Kimberlite Kimberlite Kimberlite Kamafugite Kamafugite Kamafugite Kamafugite Kamafugite Kamafugite Mafurite Phl-Picrite Phl-Picrite Phl-Picrite Phl-Picrite

Site Lemes TrsRanchos

Limeira Indai Pantano Serra doBueno

Malaquias Santa Rosa Canas Osmar Faco PresidenteOlegario

Catalo II Catalo II Catalo II Catalo II

References This work Melluso etal. (2008)

Melluso etal. (2008)



This work Melluso etal. (2008)



This work This work Melluso etal. (2008)

This work This work This work This work

Sample LEM TR-14 LIMI-1 IN I-9 PANT-2 SDB MAL S.R. CAN OS FAC PO3 C2A8 C2A10 C2A18 C2B23

V 158 227 149 158 69 237 241 197 252 148 226 203 216 178 172 179Cr 1710 1700 1440 1150 1260 470 459 810 706 1070 1260 1140 830 850 1040 270Co 81 75 80 77 91 55 60 65 58 77 78 76 60 51 56 51Ni 1070 1140 989 859 774 330 290 471 423 540 780 682 310 160 360 190Cu 40 34 53 61 116 60 126 95 78 140 90 90 120 130 70 130Zn 150 60 97 92 112 90 134 122 114 220 130 104 160 140 180 120Ga 9.0 6.7 10.5 11.8 13.6 15.0 14.8 15.4 15.5 16.0 11.0 11.6 16.0 19.0 15.0 15.0Ge 1.0 1.3 1.6 1.8 1.7 1.0 2.4 1.7 1.7 2.0 1.0 1.7 2.0 2.0 1.0 2.0As 13.0 27.7 5.6 10.5 5.0 21.1 16.5 5.0 55.0 5.0 5.0 5.0 5.0Rb 108 145 115 44 150 86 121 162 209 339 108 146 218 259 212 240Sr 2376 3170 2640 2540 2130 1897 2020 1780 1770 2061 1362 1410 2333 2184 4453 4077Y 34 22.3 39.5 43.1 29.1 30 32.1 32.3 37.9 28 19 22.2 40 44 28 31Zr 550 403 758 863 934 221 628 512 481 564 203 483 792 1032 556 842Nb 189 255 208 212 288 78 229 127 172 240 109 180 317 251 277 278Sn 3.0 3.1 3.3 3.0 1.0 2.8 2.3 2.7 3.0 2.0 1.9 9.0 4.0 3.0 4.0Sb 0.9 1.3 0.5 0.5 0.7 1.1 3.9 4.5 1.3 4.1Cs 2.0 2.6 1.5 1.0 1.4 1.2 5.6 33.6 42.1 5.5 1.8 3.8 2.7 2.8 2.7 2.8Ba 2639 7090 2680 3650 3100 1989 9310 2920 3350 10,120 15,880 5620 3874 4257 3440 6234La 303 420 345 396 313 145 303 222 220 266 143 168 323 307 308 272Ce 623 722 739 798 626 298 646 463 459 513 284 332 674 584 611 518Pr 73.2 54.3 63.7 68.3 51.8 36.2 54.4 39.8 38.9 57.5 31.6 28.7 78.3 69.1 71.4 57.9Nd 206 152 209 227 159 111 178 136 125 156 90 99 230 246 202 200Sm 32.5 22.8 37.2 41.1 27.3 19.1 31.7 22.8 22.5 23.3 14.3 16.9 37.6 34.2 32.7 27.1Eu 8.5 5.6 9.5 10.4 6.9 5.1 8.2 6.0 6.1 6.0 3.9 4.6 9.9 9.3 8.6 7.2Gd 19.3 12.1 22.0 23.7 16.6 12.3 18.1 14.0 14.4 13.3 8.9 11.2 25.5 23.5 22.0 18.6Tb 2.2 1.3 2.4 2.7 1.8 1.5 2.1 1.7 1.8 1.5 1.1 1.2 2.8 2.6 2.0 1.9Dy 9.1 5.2 10.0 10.7 7.0 6.7 8.3 7.4 8.2 6.5 4.7 5.1 10.6 10.8 8.2 8.0Ho 1.4 0.8 1.4 1.5 1.0 1.1 1.1 1.2 1.3 1.0 0.8 0.8 1.5 1.7 1.1 1.2Er 3.2 2.0 3.2 3.2 2.4 2.8 2.7 2.8 3.3 2.5 1.9 2.0 3.4 3.9 2.5 3.0Tm 0.4 0.2 0.4 0.4 0.3 0.3 0.3 0.3 0.4 0.3 0.2 0.2 0.4 0.5 0.3 0.4Yb 1.8 1.3 2.0 1.9 1.6 1.9 1.8 1.8 2.5 1.6 1.2 1.5 1.9 2.3 1.4 1.8Lu 0.2 0.1 0.2 0.2 0.1 0.2 0.2 0.2 0.3 0.2 0.2 0.2 0.2 0.3 0.2 0.2Hf 13.7 9.0 15.0 17.3 20.2 4.9 14.2 13.3 12.9 14.8 6.7 10.7 20.8 23.3 13.6 20.6Ta 13.0 9.0 12.1 13.3 17.0 8.3 16.5 8.6 8.2 17.4 9.9 9.2 16.9 14.4 18.4 17.9W 2.0 1.7 1.2 1.0 2.5 1.3 1.0 1.0 1.0 1.0 1.0 1.0Tl 0.3 0.2 0.2 0.2 0.1 0.2 0.3 0.2 0.3 0.1 0.2 0.3 0.4 0.2 0.2Pb 9.0 15.2 5.2 8.9 6.4 5.0 9.2 14.0 11.6 9.0 5.0 6.3 24.0 151.0 16.0 97.0Th 24.6 33.2 24.7 26.9 26.1 9.9 28.1 18.5 20.9 26.8 11.8 13.1 35.4 30.0 35.0 26.8U 5.5 18.2 5.1 5.7 5.7 3.5 5.1 3.4 4.1 5.9 3.6 3.0 7.0 7.6 6.0 6.1(La/Yb)N 113 225 116 140 133 51 113 84 59 112 80 77 115 90 148 102

4V.G

uarinoet

al./Chem

icalGeology

xxx(2012)

xxxxxx

Pleasecite

thisarticle

as:Guarino,

V.,et

al.,UPb

ages,SrN

d-isotope

geochemistry,

andpetrogenesis

ofkim

berlites,kam

afugitesand

phlogopite-picritesof

theAlto

ParanabaIgneous

Province,Brazil,Chem.G

eol.(2012),doi:10.1016/j.chemgeo.2012.06.016

mes

5V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxFig. 2. Photomicrographs of the samples studied. Kimberlites: Indai (a); Limeira (b); LeSo Paulo, on a VG354 Micromass thermal ionization mass spectrometer (details in Guarino et al., 2012). Perovskite major elementcompositions were obtained using a JEOL JAX8100 microprobe with15 kV accelerating potential and 12 nA beam current. Perovskitetrace element compositions (including REE) were obtained usinglaser ablation inductively coupled plasma mass spectrometry(LA ICP MS). The U Pb ages are obtained through two different techniques: SIMS analysis using a CAMECA IMS 1280 (for Trs Ranchos,Pantano, Lemes, Malaquias, Presidente Olegario and Catalo II) andlaser ablation with an Agilent 7500a ICP MS (for Limeira and Indai).The analytical details for age determination can be found in Wu et al.(2010), which demonstrated that the ages obtained through twotechniques are identical within the analytical errors for the individualsamples. In this paper, we adopt the 207Pb corrected 206Pb/238Uweighted rather than the lower intercept ages because the formerhas better precision. Perovskite Sr Nd isotopic analyses were undertaken using a Neptune MC ICP MS. Perovskite analyses (elementalcomposition, U Pb age and Sr Nd isotopic ratio) were undertakenat the Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing (China). Detailed analytical procedures (accuracy, precision and standard utilized) and the choice of adopting the 206Pb/238U weighted ages are widely described and discussed in Yang etal. (2009), Wu et al. (2010) and Li et al. (2010). X ray diffractometry(XRD) was utilized to identify kalsilite in the powders of somesamples, because this phase was not found during the microprobeanalytical work.

4. Classication and petrography

The new major and trace element data of APIP rocks are reportedin Tables 1 and 2. All the APIP rocks have high LOI (3.6 11.4 wt.%) and

Osmar (h); Presidente Olegario (i); Faco (l); Velosa (m). Phlogopite-picrites: Catalo II (n)Pntano and Trs Ranchos are kimberlites; Canas, Malaquias, Santa Rosa, Presidente Olegar

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chem(c); Pntano (d). Kamafugites: Serra do Bueno (e); Serra (f); So Jos dos Talhados (g);low to very low SiO2 content (25.3 42.6 wt.%). Taking into accounttheir mineral assemblage and the bulk rock chemical composition,the ultrabasic rocks with olivine macrocrysts and carbonated matrixare classied as kimberlites (Woolley et al., 1996; Le Maitre, 2002;Tappe et al., 2005). Using the geochemical classication of Foley etal. (1987), the bulk of the APIP rocks can be classied as kamafugites.

The APIP ugandites (Serra do Bueno, Malaquias, Serra, Santa Rosa,So Jos dos Talhados, Veridiana, Canas) are ultrabasic (i.e., SiO243 wt.%) volcanic rocks with olivine and clinopyroxene phenocrystsin a groundmass made up of clinopyroxene, perovskite, apatite,opaque, phlogopite, leucite and/or analcime. The APIP mafurites(Osmar, Presidente Olegario, Faco, Velosa) show the same phenocryst assemblage set in a groundmass of olivine, clinopyroxene, amphibole, phlogopite, perovskite, magnetite, secondary Ba zeolites(harmotome) and kalsilite (detected by XRD). Leucite grains areoften altered into analcime aggregates and kalsilite grains may havebeen replaced by Ba zeolites (harmotome), as suggested by Sgarbi etal. (2000). Hereafter, we refer ugandites and mafurites as kamafugitesfor simplicity.

The Catalo II dykes are characterized by the phlogopite rich,SiO2 poor (SiO2=29.3 31.7 wt.%), MgO rich (MgO=14.4 18.7 wt.%),TiO2 rich (4.6 6.5 wt.%), Al2O3 poor (3.9 4.7 wt.%), ultrapotassic (K2O/Na2O=12.6 32) rocks. Their mineralogical, bulk rock composition andtheir association with carbonatites, allowed us to classify them as ultramac lamprophyres (IUGS classication; Tappe et al., 2005). For clarity,following Gibson et al. (1995) and Brod et al. (2000), we still continueto call them as phlogopite picrites. A detailed petrographic descriptionof the lithotypes is reported below (see also Fig. 2).

The APIP kimberlites, kamafugites and ultramac lamprophyrescrop out in the same areas (Fig. 1), as also observed elsewhere (e.g.,North Atlantic Craton, Tappe et al., 2008, 2011; Tanzanian Craton,

. Others photomicrographs are in Fig. 2 of Melluso et al. (2008) where Indai, Limeira,io and Veridiana are kamafugites.


6 V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxFoley et al., 2012; Kaapvaal Craton, Donnelly et al., 2011; SuperiorCraton, Birkett et al., 2004; and Wyoming Craton, O'Brien et al.,1995). The association of ultramac lamprophyres with carbonatitesis also noted (cf. Tappe et al., 2005, 2006; Woolley and Kjarsgaard,2008; Tappe et al., 2009.

4.1. Kimberlites

Kimberlites from the Trs Ranchos plug have a seriate porphyritictexture with olivine xenocrysts plus a phenocryst assemblage of olivine and rare Cr spinel set in a microcrystalline groundmass mainlyformed by olivine, serpentine, phlogopite, magnetite, perovskite andpotassic richterite microcrysts and a calcitic matrix. The Indai plugs(Fig. 2a) have a seriate porphyritic texture with olivine xenocrystsplus a phenocryst assemblage made up of olivine, phlogopite, ilmenite and chromite. Phlogopite occurs as large (up to 3 mm in size) anddiscrete laths or forming ne grained aggregates with olivine. The

Fig. 3. a) APIP olivine diagram; b) IVSi-IVAl-IVFe diagram for tetrahedral cations in APIP phlogAPIP spinels (Kapsiotis et al., 2009); e) APIP ilmenite diagram (Wyatt et al., 2004); f) REE/c

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chemne grained groundmass is formed by olivine, monticellite, apatite,perovskite, baddeleyite and carbonates. The Limeira (Fig. 2b) andLemes (Fig. 2c) plugs showa seriate porphyritic texture,with xenocrystsof olivine and rare phlogopite, often partially altered to amixture of clayminerals and carbonate. The phenocryst and microphenocryst assemblage is made up of partially iddingsitized olivine, phlogopite, spineland euhedral calcite crystals up to few mm in size. Spinel is often mantled by perovskite. Themicrocrystalline groundmass is made up of olivine, monticellite, magnetite, ilmenite, perovskite, apatite, serpentineand carbonates. The samples of the Pntano plug (Fig. 2d) have porphyritic texture; themain phenocryst phases are olivine, rarely foundalso inxenocrysts, perovskite and phlogopite. Olivine grains and perovskitemicrocrysts are often rounded in shape. The groundmass is composedof perovskite, magnetite, chromite, olivine, apatite, phlogopite andcarbonates. No pyroxene has been found.

After Woolley et al. (1996), the APIP kimberlites are classied asgroup I kimberlites on the basis of: 1) volatile rich composition, as

opites; c) APIP clinopyroxene diagram; d) Cr/(Cr+Al) vs. Mg/(Mg+Fe2+) diagram forhondrite of APIP perovskites (Boynton, 1984).


7V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxrecorded by the high LOI content (6.9 11.4 wt.%); 2) variable butgenerally high CO2 and H2O, given by the abundant presence ofcarbonates, calcite veins and silicate pseudomorphs (as serpentine);3) ultrabasic composition (SiO2=25.3 32.7 wt.%); 4) slightly to stronglypotassic composition (K2O=0.8 3 wt.%; K2O/Na2O=1.1 11.4),5) inequigranular texture with pheno xenocrysts of olivine, phlogopite,Mg ilmenite, chromite, perovskite, set in a ne grained matrix.

4.2. Kamafugites

4.2.1. UganditesThe Serra do Bueno diatreme rocks (Fig. 2e) have a porphyritic

texture with olivine phenocrysts set in a ne grained groundmasscomposed of phlogopite, clinopyroxene, perovskite, magnetite andanalcime. Phlogopite phenocrysts may be coated by ne grained aggregates of magnetite. The Malaquias and Serra (Fig. 2f) plugs havea porphyritic texture, with rare olivine and clinopyroxene phenocrysts plus microcrysts of clinopyroxene, perovskite, opaque, phlogopite and leucite (often analcimized). Clinopyroxene shows acicular orprismatic shape, and is often zoned. The perovskite grains have agranular shape. Fine grained aggregates of magnetite and phlogopitelocally occur. The very ne grained groundmass consists of the samephases observed as in the microcrysts. The Santa Rosa, So Jos dosTalhados (Fig. 2g), Veridiana and Canas plugs have a porphyritic texture, with often altered and corroded olivine and clinopyroxene phenocrysts set in a groundmass of clinopyroxene, phlogopite, apatite,perovskite, analcime and magnetite.

4.2.2. MafuritesThe Osmar dyke (Fig. 2h) is porphyritic with olivine, clinopyroxene,

perovskite and magnetite phenocrysts. Some large olivine grains arerounded and more or less altered to iddingsite. Perovskite and magnetite grains are very small, reaching 1 mm in size. Clinopyroxene isoften anhedral. The groundmass is formed by the same phases observedas phenocrysts, with olivine, phlogopite, magnetite and secondaryBa zeolite (harmotome). Composite cumulitic xenoliths are formed byaggregates of phlogopite and magnetite, mantled by olivine and phlogopitemicrocrysts and then rimmed by perovskite grains. Globular to irregular ocelli containing analcime and Ba zeolites occur. The PresidenteOlegario lavas (Fig. 2i) have a pseudo uidal porphyritic texture, withphenocrysts of olivine, spinel and phlogopite. Olivine is variable in sizeand partially or totally altered. Magnetite and phlogopite locally formne grained aggregates. The ne grained groundmass is composed byacicular microcrysts of clinopyroxene, amphibole, phlogopite, perovskite,magnetite and chromite. Ba zeolites (harmotome) are secondary phases.The Faco plug (Fig. 2l) has a porphyritic texture, with phenocrystsand microphenocrysts of olivine. The groundmass is composed by clinopyroxene, magnetite, phlogopite, perovskite microcrysts. Ba zeolites(harmotome) are secondary.

The ne grained mafurites of the Velosa plug (Fig. 2m) show porphyritic texture with microcrysts of altered olivine, clinopyroxeneand phlogopite. The groundmass is composed by clinopyroxene,amphibole, phlogopite, magnetite, perovskite microcrysts, withBa zeolites (harmotome) as secondary minerals. Clinopyroxene isacicular and is often coated by aggregates of perovskite and magnetite. Poikilitic phlogopite includes clinopyroxene, perovskite andmagnetite grains.

4.3. Phlogopite picrites

The coarse to very ne grained Catalo II dykes (Fig. 2n) show apseudo uidal porphyritic texture and are dominated by elongatemicrocrysts of phlogopite. Olivine joins phlogopite in the phenocrystassemblage. Microcrystalline groundmass is composed by olivine,

spinel, apatite, perovskite, calcite and rare garnet.

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chem5. Mineral chemistry

5.1. Olivine group minerals

Olivine (Supplementary Table 1, Fig. 3a) of kimberlites is found asphenocrysts [Fo8189, where Fo is forsterite content=100*Mg/(Mg+Fe) in atoms] and xenocrysts (Fo9092). In terms of forsteritecontent, olivine in the APIP kimberlites falls within the range of olivine in kimberlites worldwide (Fo8495; Mitchell, 1986). Kamafugiteshave olivine phenocrysts and microphenocrysts, and rarelyxenocrysts. The forsterite content ranges from Fo76.2 to Fo89.7 forpheno and micro phenocrysts, while it clusters around Fo9091 forxenocrysts in Osmar mafurite. Olivine pheno and microcrysts of thephlogopite picrites cluster around Fo8889, with variable CaO contents(0.1 to 1 wt.%). Monticellite (Supplementary Table 1) has been foundin the groundmass of kimberlites. At Indai, monticellite has Mg#=91.7 92.2, whereas at Limeira and Lemes, the Mg# varies from 88to 90 and from 74 to 84, respectively, thus showing a signicant compositional range.

5.2. Mica

Phlogopite in kimberlites has a wide range in Al (5.5 13.3 wt.%Al2O3), Ti (0.2 5.3 wt.% TiO2), Fe (2.8 10.2 wt.% FeOtot), F (0.11.2 wt.%) and Mg# [Mg#=Mg/(Mg+Fe2+)=0.91 1]. Phlogopitesin kamafugites exhibit an even larger range in Al (2.76 12.52 wt.%Al2O3), Ti (0.14 4.92 wt.% TiO2), Fe (6.3 21.9 wt.% FeOtot) and F(0.42 1.58 wt.%) concentrations, coupled with lower Mg# (0.63 1).In phlogopite picrites, phlogopite has variable range in Al (1.2913.12 wt.% Al2O3), Ti (0.44 4.04 wt.% TiO2), Fe (7.42 33.92 wt.% FeOtot) and F (0.1 1.97 wt.%) and a much wider Mg# (0.44 0.96; Supplementary Table 2).

The APIP micas show the typical substitution IVAl3+ IVFe3+, thatdenes the phlogopite/tetra ferriphlogopite variation, indicating deciency in the sum of the common tetrahedral cations (i.e., Si+ IVAl3+

b8; Brod et al., 2001; Supplementary Table 2). This IVAl deciency isdue to the very low Al contents of the host magma (Al2O3=2.06.9 wt.%; average=4.4 wt.%; Table 1; Mitchell, 1995). The variation intetrahedral cations, and the different enrichment in Fe3+, is shown inFig. 3b. Phlogopites of kimberlites, kamafugites and phlogopite picritesdisplay a decrease in the Si content with increasing IVFe3+; on the otherhand, phlogopites of phlogopite picrites, kimberlites and kamafugitesshow an opposite trend, with increasing of Si and IVFe3+ cations(Fig. 3b), the highest contents in Fe3+ is found in the rims of kimberlitephlogopites and in the rims and in the microcryst present in thephlogopite picrites and kamafugites. This different Fe3+ content indicatesthat: a) kimberlites do not show signicant variations of oxygen fugacity,so the presence of Fe3+ into tetrahedral site is a primary characteristicof magmas and b) in kamafugites and phlogopite picrites Fe3+ ofphlogopite shows a variation indicating more oxidizing conditions(e.g., Wones and Eugster, 1965).

5.3. Clinopyroxene

Clinopyroxene (Supplementary Table 3) is present in kamafugitesas pheno and microcrysts, in some cases corroded and variably altered. It is mostly diopside with rarer augite (Wo5139En4327Fs346;Fig. 3c). As already pointed out by Sgarbi et al. (2000) and Mellusoet al. (2008), the low Al contents (0.10 4.18 wt.% Al2O3) are insufcient to ll the tetrahedral site in most clinopyroxenes, hence Fe3+

is sometimes required (up to 0.11 atoms Fe3+ in the tetrahedralsite for four cations together with all Si and Al). This feature isnoted also in other localities where clinopyroxene crystallizes fromkamafugitic magma (e.g., Italy: Cellai et al., 1994; Tappe et al., 2003;Rwanda and Uganda: Sahama, 1974; Aoki et al., 1985; Dawson et

al., 1985; De Mulder, 1985; Tappe et al., 2003; Western Qinling,


Gansu Province, China: Yu et al., 2001; Serbian ultrapotassic province,Prelevic et al., 2005).

5.4. Oxide minerals

5.4.1. SpinelXenocryst spinel of kimberlites (Fig. 3d) is Al rich (34.9 48.4 wt.%

Al2O3; Mg#=0.62 0.78), whereas spinel crystallized from magmaranges from magnesio chromite to chromite [Cr#=Cr/(Cr+Al)=0.76 0.87; Mg#=0.40 0.53], Cr Ti rich magnetite (5.0 30.2 wt.%Cr2O3; 3.2 14.5 wt.% TiO2), magnetite and Ti magnetite (7 27 mol%ulvspinel; 3.55 15.04 wt.% TiO2). Magnetite and Ti magnetite(2.27 19.75 wt.% TiO2) are the dominant spinels in uganditesand mafurites (4 38 mol% ulvspinel). Chromite (Cr#=0.97 0.98;Mg#=0.27 0.29; Fig. 3c) and Cr Ti rich magnetite (5.0412.12 wt.% Cr2O3; 13.40 16.81 wt.% TiO2) are rarely present in thekamafugites. Ti magnetite (8 30 mol% ulvspinel, 4.1 17.4 wt.%TiO2), magnesio chromite (Cr#=0.86 0.87; Mg#=0.52 0.58) andchromite (Cr#=0.85; Mg#=0.45) have been analysed in phlogopitepicrites (Fig. 3d; Supplementary Table 4).

5.4.2. IlmeniteIlmenite is found in APIP kimberlites (ilmenite=80 93 mol%;

Fe2+TiO3=49 68 mol%), and has high geikielite content (MgTiO3=28 46 mol%) and variable eskolaite (0 8 wt.% Cr2O3). The ilmenitein the APIP kimberlites falls in the eld of kimberlitic ilmenite(Fig. 3e; Wyatt et al., 2004), being characterized by relatively highMgO (6.3 12.1 wt.%) at a given TiO2 (43.1 53.5 wt.%) (Supplementary Table 4).

5.4.3. PerovskitePerovskites of kimberlites, kamafugites and phlogopite picrites

show a similar composition, as they have high perovskite component(CaTiO3, 75.5 98.1 mol%) coupled with variable loparite (Ce0.5Na0.5-TiO3, 0.3 22.2 mol%) and tausonite (SrTiO3, 0.2 3.9 mol%), low thorutite(Th0.5TiO3, 0 0.8 mol%), latrappite (CaFe0.5Nb0.5O3, 0 2.9 mol%), lueshite(NaNbO3, 0 1.1 mol%) and Ce orthoferrite (CeFeO3, 0 2.5 mol%;Supplementary Table 5). The new LA ICP MS trace element concentrations on perovskite of kimberlites, kamafugites andphlogopite picrites are reported in Supplementary Table 6. Perovskitesare strongly enriched in light REE (e.g., La ~5700 25,900 ppm in

8 V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxFig. 4. UPb ages of perovskites obtained by SIMS and laser ablation method

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chems for APIP kimberlites (a), kamafugites (b) and phlogopite-picrites (c).


cont

9V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxFig. 4b(kimberlites, ~2700 4400 ppm in kamafugites and ~6900 7800 ppm inphlogopite picrites) and have strongly fractionated REE patterns inchondrite normalized diagrams, with La up to ~80,000 times chondriteand high LREE/HREE ratios (LaN/YbN ~700 2700 for kimberlites, ~260700 for kamafugites and ~770 870 for phlogopite picrites), matchingthe values reported by Melluso et al. (2008) in kimberlite andkamafugite perovskites (Fig. 3f).

5.5. Minor phases

Amphibole is a very rare interstitial phase of Trs Ranchos kimberlite, Presidente Olegario and Velosa kamafugites. The amphiboles arehigh Mg# (0.92 0.94 in kimberlites and 0.84 0.88 in kamafugites)titanian potassian richterite, following the classication of Leake etal. (1997; Supplementary Table 7). The presence of this amphiboleis known to occur also in West Qinling kamafugites in China (Yu etal., 2001) and in Torngat ultramac lamprophyres in Canada (Tappeet al., 2004, 2008). Ca rich garnet is a late crystallized phase of theCatalo phlogopite picrites. The compositions are rich in TiO2 (8.2115.12 wt.%), and are mostly made up of andradite (Ca3Fe2Si3O12;8 59 mol%) and morimotoite (Ca3TiFeSi3O12; 8 70 mol%), withlower morimotoite Mg (Ca3TiMgSi3O12; 4 21 mol%) and schorlomite(Ca3Ti2SiFe2O12; 11 15 mol%) end members calculated followingLocock, 2008 (Supplementary Table 8). Similar garnet present inTorngat ultramac dykes (in northern Labrador and northeasternQuebec; Tappe et al., 2004, 2008). Apatite is F (2.6 2.8 wt.%) andSr rich (3.1 3.3 wt.% SrO). Leucite occurs in Malaquias ugandites asrounded groundmass microcrysts, and is typically transformed intoanalcime. Kalsilite was identied through XRD analyses in theOsmar, Presidente Olegario, Faco and Velosa samples. Harmotomeand analcime are the most common secondary phase in kamafugiticrocks (Fig. 4).

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Cheminued).6. New UPb perovskite age determinations

New in situ U Pb ages on perovskite have been obtained from selected localities. The main results are shown in Fig. 1 and listed inTable 3 and in Supplementary Table 9. Perovskite (CaTiO3) is a primary magmatic mineral, and crystallizes at an early stage, so it recordsthe primary geochemical and isotopic features with respect to theemplacement and origin of the kimberlitic magma. Its resistance toalteration is useful to preserve the pristine magmatic features of thecrystallizing melt (e.g., Paton et al., 2007; Wu et al., 2010; Tappeand Simonetti, 2012).

The Pntano, Lemes, Indai and Trs Ranchos kimberlites and twosamples of the Catalo phlogopite picrites have U Pb ages clusteringbetween 87 and 80 Ma, while Malaquias and Presidente Olegariokamafugites cluster between 81 and 78 Ma. Only two samples showslightly high values in the U Pb ages: Limeira kimberlite (916 Ma)and one of the Catalo phlogopite picrites (904 Ma) (Figs. 4a,b,c).

In Fig. 1, it is possible to see that our new age data are higher forkimberlites (~91 80 Ma) compared with previous APIP kimberliteU/Pb zircon ages (~87 80 Ma; Davis, 1977). Our new age data forkamafugites (~81 78 Ma) fall within the U/Pb perovskite ages(~81 75 Ma) of the Mata da Corda kamafugites (Sgarbi et al., 2004).As concerns Catalo II phlogopite picrites, our new age data (823and 834 Ma) are close to the age (83 Ma) estimated by Machado(1991). Only one Catalo II phlogopite picrite shows a slightly olderage (904 Ma).

7. Geochemistry

7.1. Major elements

The APIP kimberlites, kamafugites and phlogopite picrites arecharacterized by variable and quite exotic melt compositions


10 V. Guarino et al. / Chemical Geology xxx (2012) xxxxxx(Figs. 5 8). These rocks are all ultrabasic (SiO2=25.3 42.6 wt.%),MgO rich (MgO=10.4 27.8 wt.%), high Mg# (Mg# 67 86), CaO rich(8.0 15.4 wt.%), Al2O3 poor (2.0 7.1 wt.%), Na2O poor (0.1 2.9 wt.%)and potassic to ultrapotassic in character (K2O=0.8 4.6 wt.%;K2O/Na2O ranging from 0.4 to 32.0 in all but one sample). Thetotal alkali content (Na2O+K2O) remains relatively low, rangingfrom 1.1 to 5.2 wt%. The relatively low K2O is a typical characteristic of uncontaminated kimberlites worldwide (Kjarsgaard et al.,2009; Kamenetsky et al., in press); while the relatively low K2Oof kamafugites is probably not a pristine value, likely being relatedto the effects of post magmatic alteration (zeolitization; Melluso et al.,2008). High and variable LOI (3.6 11.4 wt.%) are largely due to theabundant presence of volatile bearing phases such as carbonates, serpentine, phlogopite, zeolites and amphibole. The investigated rocksfall within the eld of the literature APIP rocks (Gibson et al., 1994;Bizzi et al., 1995; Gibson et al., 1995; Carlson et al., 1996; Gibson et al.,1997; Arajo et al., 2001; Gomes and Comin Chiaramonti, 2005;Melluso et al., 2008; Figs. 5 and 6). The kimberlites typically show thehighest MgO and the lowest SiO2 contents, whereas the kamafugiteshave the lowest MgO and the highest SiO2 (Fig. 5). With decreasingMgO, only Al2O3 shows good negative correlation (Fig. 6). TiO2 and

Fig. 4c (continued).

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, ChemSiO2 also are negatively correlated with MgO, but with major scatter(Fig. 6). The highest TiO2, Fe2O3(tot) and K2O, and the lowest Na2O contents are recorded in the phlogopite picrites.

7.2. Trace element geochemistry

Trace element concentrations are variable, also as a consequenceof the highly porphyritic nature of the samples. The high concentrations of Ni (~140 1500 ppm) and Cr (~260 1700 ppm) arewell correlated with MgO whole rock concentration (Fig. 7a,b),with kimberlites being the most enriched in these elements. Elementssuch as Sc, V and Co do not show any clear correlation with MgO, witha large overlap among the different rock types. The Large Ion Lithophile Element (LILE) content is high, especially for Sr (~12504450 ppm) and Ba (~2150 34,400 ppm; Fig. 7c,d). Also the concentration of high eld strength elements (HFSE) is generally high,with no correlation with MgO and no substantial differences amongthe various lithotypes (Fig. 8). The APIP rocks show scattered variation of Light and Middle REE with MgO, but a clear negative correlation between heavy REE and MgO (Fig. 8a,b,c,d). Ytterbium and Lushow an asymptotic increase with decreasing MgO, with a sharp increase in concentration at MgO b10 wt.% (Fig. 8e,f). Other trace elements do not show any particular feature compared with MgO asconsequence of the porphyritic nature of the rocks and theirpost magmatic mobilization of the most mobile elements. The APIProcks show highly fractionated REE patterns [(La/Yb)N=51 225],with La abundances 500 1000 times chondrite and relatively lowHREE abundance (Lu=~5 10 times chondrite; Fig. 9a,b), near to observed values in Neoproterozoic Brauna kimberlites emplaced in thenortheastern part of the So Francisco Craton, Brazil (Donatti Filhoet al., this issue). Kimberlites and phlogopite picrites show slightlymore REE fractionated patterns than kamafugites.

The primitive mantle normalized incompatible element patternsof the APIP rocks show a general bell shaped pattern peaking at Baand with troughs at K (Fig. 9c,d). Compared with neighbouring REE(Nd and Sm), the APIP rocks show also slight troughs at Hf andZr. Nb and Ta show relatively high concentrations (~100 500 timesprimitive mantle; Fig. 7c,d). Other common features among thedifferent rocks are the relatively low abundance of Cs and Rb, andsmall troughs at Pb. Peaks at Pb in the patterns of phlogopite picrites(Fig. 9c) can be considered as an primary feature, in a way similar tothe Group II kimberlites (Becker and le Roex, 2006). The Late Proterozoic Brauna kimberlite eld (Donatti Filho et al., this issue) shows asimilar, less enriched pattern with respect to APIP kimberlites (Fig. 9c).

7.3. Sr and Nd isotope compositions

Sr and Nd isotopic analyses have been obtained with in situmass spectrometry on thirteen perovskite separates, six from kimberlites, four from kamafugites and three from phlogopite picrites(Table 3). There is complete overlap of values among the differentrock types, all showing initial Sr>0 and 87Sr/86Sri ranging from0.70431 to 0.70686. Twelve perovskite samples cluster in a narrowerSr isotopic range (87Sr/86Sri=0.70506 0.70565); and initial Ndare all negative, clustering around 5.1 (total range from 4.14 to6.10). Only the Trs Ranchos perovskite shows a slightly differentSr isotope (87Sr/86Sri=0.70467) and initial Nd (3.85). The isotopicdata of the APIP perovskites analysed here fall in the same areadepicted by bulk rock values of APIP kimberlites and kamafugitesand, remarkably, of the new bulk rock analyses of Catalo phlogopitepicrites reported in Table 3 (cf. Fig. 9). The APIP isotopic data (Fig. 10)have a similar composition to the Brauna kimberlites (Donatti Filho etal., this issue), and show an intermediate composition between Group Iand Group II kimberlites. The Sr isotopic ratios of the APIP rocks havepoor negative correlation with bulk rock MgO, whereas Nd isotopes

show poor correlation with MgO (Fig. 11).


Table 3New in-situ UPb age and Sr-Nd analyses on perovskite for some APIP rocks and new SrNd data for phlogopite-picrites of Catalo II district.

Rock Locality References Sample Analyseson:

Type ofanalyses:

206Pb/238Uweightedages (Ma)

87Rb/86Sr

87Sr/86Sr

2 s 147Sm/144Nd

143Nd/144Nd

2 s 87Sr/86Sr i

143Nd/144Nd i

Nd(i)

Kimberlite Limeira This work LIMI-1 Perovskite Laser 916 0.00043 0.70523 0.00002 0.0720 0.51229 0.00001 0.70523 0.51225 5.4Kimberlite Tres Ranchos This work TR-14 Perovskite SIMS 873 0.01700 0.70467 0.00006 0.0531 0.51236 0.000013 0.70467 0.51233 3.9Kimberlite Pantano This work PANT-1 Perovskite SIMS 841 0.00022 0.70541 0.00002 0.0717 0.51235 0.00006 0.70540 0.51232 4.2Kimberlite Pantano This work PANT-2 Perovskite SIMS 821 0.00029 0.70541 0.00003 0.0682 0.51234 0.000024 0.70541 0.51230 4.5Kimberlite Lemes This work LEM Perovskite SIMS 842Kimberlite Indaia This work IN I-1 Perovskite Laser 805 0.00021 0.70523 0.00002 0.0711 0.51230 0.000012 0.70523 0.51227 5.3Kamafugite Malaquias This work MAL Perovskite SIMS 792 0.00028 0.70565 0.00003 0.0746 0.51231 0.00009 0.70565 0.51227 5.2Kamafugite Presidente

OlegarioThis work PO2 Perovskite SIMS 812 0.00014 0.70508 0.00003 0.1038 0.51234 0.00008 0.70508 0.51230 4.6


This work PO2 Perovskite SIMS 791 0.00018 0.70507 0.00003 0.1005 0.51233 0.00008 0.70506 0.51229 4.9


This work PO3 Perovskite SIMS 784 0.00008 0.70535 0.00001 0.0790 0.51228 0.00007 0.70535 0.51225 5.7

Phl-Picrite Catalao II This work C2A8 Perovskite SIMS 834 0.00153 0.70545 0.00006 0.0694 0.51226 0.000013 0.70544 0.51222 6.1Phl-Picrite Catalao II This work C2A18 Perovskite SIMS 904Phl-Picrite Catalao II This work C2B23 Perovskite SIMS 823 0.00600 0.70554 0.00006 0.0677 0.51229 0.00008 0.70551 0.51225 5.4Phl-Picrite Catalao II This work C2A8 Whole-rock 0.49434 0.70571 0.00005 0.0984 0.51227 0.000011 0.70553 0.51221 6.2Phl-Picrite Catalao II This work C2A10 Whole-rock 0.91174 0.70619 0.00010 0.0837 0.51223 0.000013 0.70531 0.51223 5.9Phl-Picrite Catalao II This work C2A18 Whole-rock 0.48017 0.70573 0.00003 0.0974 0.51227 0.000013 0.70557 0.51223 5.8Phl-Picrite Catalao II This work C2B23 Whole-rock 0.73486 0.70580 0.00006 0.0816 0.51228 0.000010 0.70599 0.51219 6.7

Phl-Picrite=Phlogopite-picrite.

Fig. 5.Major elements (wt.%) vs. SiO2 (wt.%) variation diagrams for APIP rocks analysed here. APIP literature data are taken by Gibson et al. (1994, 1995, 1997); Bizzi et al. (1995);Carlson et al. (1996); Arajo et al. (2001).

11V. Guarino et al. / Chemical Geology xxx (2012) xxxxxx

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotope geochemistry, and petrogenesis of kimberlites, kamafugites andphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chem. Geol. (2012), doi:10.1016/j.chemgeo.2012.06.016

12 V. Guarino et al. / Chemical Geology xxx (2012) xxxxxx8. Discussion

The chemical andmineralogical composition of the APIPmagmas canbe considered quite exotic compared with the much more widespreadbasaltic magmas emplaced in neighbouring districts (e.g., tholeiitic andalkaline rocks of the Paran Basin). The kimberlite whole rock compositions can be considered only a proxy of the original melt composition,being these rocks xenocryst bearing (e.g., forsterite and diopside mantlephases) and heavily weathered (presence of silicate pseudomorphs plussecondary calcite veins), although they are uncontaminated [Contamination Index=C.I. =b1.5; C.I. Clement, 1982; 1.5 value is of Nowicki et al.,2008; Table 1]. These characteristics make difcult to identify magmaticcompositions of the original kimberlitic magma (Becker and le Roex,2006; Kamenetsky et al., 2009; Kjarsgaard et al., 2009; Arndt et al.,2010; Mitchell and Tappe, 2010).

The APIP kimberlites, kamafugites and phlogopite picrites arestrongly silica undersaturated magmas, enriched in CaO and volatiles(CO2 and H2O), but not strongly so in alkalies. These compositionscannot be generated by melting four phase lherzolitic mantle, butare more likely related to a volatile rich peridotitic source assemblagemetasomatized by phlogopite and carbonate rich veins (Rosenthalet al., 2009; Melluso et al., 2011). Partial melting of carbonated peridotite can produce low SiO2 melts (Hirose, 1997; Gudnnsson and

Fig. 6.Major elements (wt.%) vs. MgO (wt.%) variation diagrams for APIP rocks. APIP literatual., 1996; Gibson et al., 1997; Arajo et al., 2001.

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, ChemPresnall, 2005; Dasgupta et al., 2007). The CaO enrichment may bederived mainly from a high carbonate content.

Melting of APIP mantle source is considered as having taken placewithin the phlogopite rich veins, similarly to what proposed byGibson et al. (2006) for the alkaline rocks from the western marginof the Paran Etendeka igneous province. The incompatible elementsof APIP rocks mostly show strong enrichment (Fig. 9c,d), but troughsat K, Rb and Cs may indicate the presence of residual phlogopite during partial melting.

Experimental studies point out that low degree melts of a carbonated lherzolite, in the presence of H2O, are carbonatitic (Brey et al.,2008; Litasov et al., 2008; Dasgupta et al., 2009; Foley et al., 2009;Litasov, 2011). With increasing degree of partial melting, the silicate component in the melt increases, producing melts mixedcarbonatitic silicatic melts resembling kimberlites, kamafugitesand other similar low SiO2 rocks, resembling phlogopite picritecompositions. The MgO/CaO vs. SiO2/Al2O3 diagram (Gudnnssonand Presnall, 2005; Fig. 12) was used to determine the possiblepressure and depth of formation for APIP rocks. In order to obtaina reliable whole rock composition representative of a liquid, olivinexenocrysts were removed from the analysed kimberlites (Supplementary Table 10), following the methods of Nielsen and Sand(2008) and Nielsen et al. (2009). Pressure data (Fig. 12) are mostly

re data are taken by Gibson et al., 1994; Bizzi et al., 1995; Gibson et al., 1995; Carlson et


Fig. 7. Ni, Cr, Sr, Ba (ppm) vs. MgO (wt.%) variation diagrams for APIP rocks. APIP literature data are taken by Gibson et al. (1994, 1995, 1997); Bizzi et al. (1995); Carlson et al.(1996); Arajo et al., 2001.

Fig. 8. LREE (La and Ce; 8a,b) MREE (Eu and Gd; 8c,d) HREE (Yb and Lu; 8e,f) (ppm) vs. MgO (wt.%) variation diagrams for APIP rocks analysed here. APIP literature data aretaken by Gibson et al., 1994, 1995, 1997; Bizzi et al. (1995); Carlson et al. (1996); Arajo et al. (2001).

13V. Guarino et al. / Chemical Geology xxx (2012) xxxxxx

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotope geochemistry, and petrogenesis of kimberlites, kamafugites andphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chem. Geol. (2012), doi:10.1016/j.chemgeo.2012.06.016

ugits, ka

14 V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxaround 5 GPa for kimberlites, about 4 5 GPa for phlogopite picritesand about 3 5 GPa for kamafugites.

Fig. 9. a,b) Chondrite-normalized rare earth element pattern for APIP kimberlites, kamafization values of Lyubetskaya and Korenaga, 2007) element pattern for APIP kimberlitekimberlites is reported for comparison (Donatti-Filho et al., this issue).8.1. The role of partial melting in the APIP rocks

The origin of mac ultramac, alkaline and carbonatitic rocks incontinental settings is usually explained by low to moderate degrees

Fig. 10. 143Nd/144Ndi vs. 87Sr/86Sri diagram for the new perovskite data. APIP kimberliteand kamafugite elds are taken by Gibson et al. (1995); Carlson et al. (1996); Thompsonet al. (1998); Arajo et al. (2001); Gomes and Comin-Chiaramonti (2005). Catalophlogopite-picrite eld is taken by: this paper (open rhombs; Table 3); Gibson et al.(1995); Gomes and Comin-Chiaramonti (2005). Isotopic eld of Brauna kimberlites istaken by Donatti-Filho et al. (this issue). Group I and Group II southern Africa kimberliteelds are of: Smith, 1983; Fraser and Hawkesworth (1992); Tainton (1992); Clark(1994); Nowell et al. (2004); Coe (2004); Becker and le Roex (2006). Trindade eld istaken by Halliday et al. (1992). Uganda kamafugite eld is taken by Rosenthal et al.(2009). Serra do Mar eld is taken by Thompson et al. (1998); Bennio et al. (2002);Lustrino et al. (2003); Brotzu et al. (2005, 2007).

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chemof partial melting of a carbonated mantle source (e.g., Wyllie, 1980;Eggler, 1989; Tainton and McKenzie, 1994; le Roex et al., 2003;

es and phlogopite-picrites (Boynton, 1984). c,d) Primitive mantle-normalized (normal-mafugites and phlogopite-picrites. In panels a and c the compositional eld of BraunaGudnnsson and Presnall, 2005; Melluso et al., 2011). The experimental study of Dalton and Presnall (1998a,b) pointed out thatkimberlitic magmas can be produced through ~1% partial meltingof carbonate bearing garnet lherzolite at 5 6 GPa, where lowerdegrees of melting (~0.3%) can produce carbonatitic melts. Theserelatively high pressures of formation and low degrees of partialmelting are compatible with the strong enrichment in incompatibleelements of kimberlites and the strongly depleted HREE typical ofmetasomatically enriched sources (e.g., Wass and Rogers, 1980).

A non modal partial melting modelling has been used to constrainthe sources of the APIP rocks. The best t between the source ofGroup I kimberlites calculated by le Roex et al. (2003) and the calculated source of the APIP rocks are obtained with low degreemelting (f=0.5 2%; Fig. 13). A hypothetical kimberlitic composition has been estimated by removing the composition of olivinexenocrysts (Supplementary Table 10). In Fig. 13a, the calculatedsource for kimberlites (new data in Supplementary Table 10) andphlogopite picrites shows a similar pattern, very similar to thesource estimated for Group I kimberlites by le Roex et al. (2003).The calculated source for APIP kamafugites (Fig. 13b) is also similarto the APIP kamafugite source (f=1%) suggested by Gomes andComin Chiaramonti (2005).

Many studies point out the absence of orthopyroxene in thesource of highly alkaline, ultramac magmas similar to the APIProcks (e.g., Tappe et al., 2007 and references therein). This is reasonable hypothesis taking into account the extreme silica undersaturation of kimberlitic and kamafugitic melts, and suggesting thatthe mantle source was metasomatized by carbonate rich melts ofuids metasomatized the mantle during their ascent, as a consequence of metasomatic reactions which may have taken place atdepths 2.5 3 GPa (Tappe et al., 2007): CaMg(CO3)2 (dolomiticmelt)+4MgSiO3 (orthopyroxene)=CaMgSi2O6 (clinopyroxene)+


2Mg2SiO4 (olivine)+2CO2. In addition, the absence of phenocrystclinopyroxene in the kimberlites is probably related to a low pressurereaction: CaMgSi2O6 (clinopyroxene)=CaMgSiO4 (monticellite)+SiO2

Fig. 11. 87Sr/86Sri and 143Nd/144Ndi vs. MgO (wt.%) diagrams for APIP rocks analysed here. APCarlson et al., 1996; Gibson et al., 1997; Arajo et al., 2001.

15V. Guarino et al. / Chemical Geology xxx (2012) xxxxxx(liquid), constraining silica activity at very low values (e.g., Barker,2001). The presence of monticellite in the APIP kimberlites supports thisview.

8.2. Involvement of mantle plume

Three main petrogenetic models have been hypothesized for theorigin of the APIP magmas. Two of these invoke the presence of mantle plumes and the third explains their peculiar compositions withpartial melting of metasomatized lithospheric mantle sources.

(1) On the basis of the common Dupal Sr Nd Pb isotopic anomalyof the Brazilian alkaline rocks, Bizzi et al. (1995) evidenced agenetic link with some South Atlantic OIB and seamountswith Dupal geochemical signatures (e.g., Walvis Ridge andRio Grande Rise), proposing that the source of the APIP rockswas modied by the Tristan da Cunha hot spot at ~130 Ma.

(2) Using similar geochemical arguments, and on the basis of aSE directed decrease of the igneous activity (from ~90 Ma inthe NW most Ipor Province towards ~80 Ma in the APIP andnally ~60 Ma in the SE most sectors of the Serra do Mar Igneous Province), several authors proposed a connection with adifferent mantle plume/hot spot, whose present day manifestation has been identied in the Trindade Martin Vaz archipelago (Gibson et al., 1995; Thompson et al., 1998). In particular,Fig. 12. MgO/CaO vs. SiO2/Al2O3 diagram (Gudnnsson and Presnall, 2005) for APIProcks. The APIP kimberlites plotted are taken by in Supplementary Table 9.

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chemthe decrease in U Pb perovskite ages from Santo Antonio daBarra (Gois; ~89 Ma) to Mata da Corda kamafugites (AltoParanaba; ~80 Ma) has been interpreted by Sgarbi et al.(2004) as a hot spot track related to the NW directed motionof the south American platform.

(3) On the basis of a detailed Sr Nd Pb Os isotopic study, Carlsonet al. (1996, 2007) pointed out different lithospheric mantlesources in the genesis of the Goias and Alto Paranaba alkaline rocks. These sources are considered to have stronglyinuenced, or entirely determined, the chemical compositionof the investigated rocks. The 187Os/188Os of the kamafugiticrocks higher than that observed for peridotite suggests amixtureof lithospheric peridotite of varying age, veined and/or interlayered with various olivine poor components (Carlson etal., 2007). The isotopic similarity between APIP rocks andBrauna kimberlites shown in Fig. 9 conrm not signicant isotopic changes in the mantle during a large span of the time forthe magmatism in southern Brazil, between the 642 MaBrauna kimberlites (Donatti Filho et al., this issue) and the91 78 Ma APIP rocks. Also our new U Pb perovskite age determinations disagree with a hot spot track model. Indeed,we have measured ages as old as those analysed in the GoiasProvince (~91 90 Ma), testifying an almost coeval magmaproduction in the Goias and Alto Paranaiba provinces thatare more than 400 km apart. The volcanic activity lasted inthe APIP probably up to Late Cretaceous (our younger U Pbperovskite ages=78 Ma) or even Paleocene (~61 Ma; Read etal., 2004, and references therein). We believe that these age constraints, associated with the exotic chemical composition of themagma, requiring low to very low degrees of partial melting,are best interpreted as the effect of chemically and mineralogically heterogeneous mantle sources, whose key features were

IP literature data are taken by Gibson et al., 1994; Bizzi et al., 1995; Gibson et al., 1995;acquiredwhen theAmazonian and So Francisco Craton collided,during the Brasiliano orogenetic cycle. As a natural consequence,the involvement of a thermal perturbation caused by the presence of a hot spot or a mantle plume is not required, being therelatively high amount ofmagmaproduced as consequence of elevated homologous temperature rather than elevated absolutetemperature (e.g., Anderson, 2011).

The estimated depth of formation of the investigated APIP rocks(Fig. 12) are mostly clustered around 3 and 5 GPa. This depth rangecorresponds to the Low Velocity Zone (LVZ), a well known geophysical feature commonly located worldwide in the depth interval of~60 250 km (~2 8 GPa). At these depths both vp and vs show astrong decrease of wave speed, a characteristics interpreted as thepresence of few percent (1 3%) of melts (e.g., Thybo, 2006). Thestrong anisotropy recorded in the LVZ is compatible with the presence of melts arranged in lamellae in a network of deformed solid


16 V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxperidotitic matrix (e.g., Conrad et al., 2010; Anderson, 2011). Themelt should be consequence of the presence of low temperaturemelting minerals like dolomite (at ~2 4 GPa) and magnesite (at P>4 GPa) or phlogopite, while the lamellae shape could be related toshearing of the lithospheric plates over the more plastic deepermantle.

9. Conclusions

The new U Pb perovskite ages of kimberlites (~91 80 Ma),kamafugites (~81 78 Ma) and phlogopite picrites (~90 82 Ma)more tightly constrain the age of APIP magmatism. The new Sr Ndisotopic ratios measured on perovskites indicate slightly radiogeniccompositions, similar to those reported for other APIP kimberliticand kamafugitic rocks and for Brauna kimberlites, not compatiblewith classical mantle plume models based on isotopic grounds.

These rocks are ultrabasic, mac ultramac, strongly silicaundersaturated potassic to ultrapotassic, Ca Ti rich and Al poormeltswith high REE and other incompatible trace elements. These features suggest an origin in the Low Velocity Zone of the upper mantle,with variable interaction with lithospheric mantle, without anykind of geochemical or isotopic relationship with the allegedTrindade Martin Vaz or Tristan da Cunha (supposed) mantle

Fig. 13. Primitive mantle-normalized trace element pattern for APIP rocks, but for kimberlcomposition eld (f=0.52%) has been calculated for kimberlites and phlogopite picrphlogopite-picrite. Group I kimberlite source (f=0.41.5%) is calculated after le Roex et awith a non-modal melting mode starting from Osmar and Presidente Olegario samples. Thfrom parental magmas calculated by Gomes and Comin-Chiaramonti (2005). Normalizinnon-modal melting mode, the source mineralogy and their modal and melting proportions,used in the melting modelling is a garnet lherzolite (modal proportions: olivine 63%, ortorthopyroxene 5%, clinopyroxene 5%, garnet 4%, after le Roex et al., 2003).

Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, Chemplumes (Gibson et al., 1995; Thompson et al., 1998; Sgarbi et al.,2004; Bulanova et al., 2010). The volume of magma produced,the calculated degree of partial melting and the composition ofthe APIP magmas and the absence of a hot spot track from Goiasto Alto Paranaba igneous provinces do not require any thermallyanomalous mantle, but, rather, strongly metasomatized and oxidized C H bearing sources able to stabilize carbonates and hydrous silicate minerals. Geochemical features indicate a mantlesource consisting of a carbonated lherzolite with phlogopite richveins, enriched in incompatible elements for all APIP rocks.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.chemgeo.2012.06.016.

Acknowledgements

We dedicate this paper to the late Prof. Enzo Michele Piccirillo. Enzospent much of his scientic career in studying the Brazilian volcanicrocks and was a great guide to our work in Brazil and elsewhere. M.Lustrino and L. Melluso are particularly grateful to him for havingshared his love for research and his rigorous approach to petrogeneticmodelling. Thanks to Roberto de' Gennaro and Sergio Bravi for their assistance in the microprobe and technical work. The reviews of RichardCarlson, an anonymous reviewer, and scientic and editorial input of

ites we have used the re-calculated data present in Supplementary Table 9. a) Sourceites with a non-modal melting mode, starting from Lemes kimberlite and C2A18l., 2003. b) Source composition eld (f=0.52%) has been calculated for kamafugitese APIP kamafugite source (f=1%) is calculated with non-modal melting mode startingg values are from Lyubetskaya and Korenaga (2007). The parameters used in the

together partition coefcients, are taken by le Roex et al. (2003). The source mineralogyhopyroxene 23%, clinopyroxene 12%, garnet 2%; and melting proportions: olivine 5%,


17V. Guarino et al. / Chemical Geology xxx (2012) xxxxxxSebastian Tappe were very useful for the preparation of a revisedmanuscript. The manuscript has beneted of grants from Ateneo LaSapienza to M.Lustrino, Fondi per la Ricerca Dipartimentale 2010 toL.Melluso and FAPESP grants to C.B.Gomes. M.Lustrino thanks D.L.Anderson, D.C. Presnall, G. Foulger and J.H. Natland for discussions onthe signicance of the low velocity zone in the upper mantle.

References

Almeida, F.F.M., De Brito Neves, B.B., Dal Re Carneiro, C., 2000. The origin and evolutionof the south American platform. Earth-Science Reviews 50, 77111.

Anderson, D.L., 2011. Hawaii, boundary layers and ambient mantle-geophysical con-straints. Journal of Petrology 52, 15471577.

Aoki, K., Yoshida, T., Yusa, K., Nakamura, Y., 1985. Petrology and geochemistry of theNyamuragira Volcano, Zaire. Journal of Volcanology and Geothermal Research 25, 128.

Arajo, A.L.N., Carlson, R.W., Gaspar, J.C., Bizzi, L.A., 2001. Petrology of kamafugites andkimberlites from the Alto Paranaba Alkaline Province, Minas Gerais, Brazil. Contri-butions to Mineralogy and Petrology 142, 163177.

Arndt, N.T., Guitreau, M., Boullier, A.M., le Roex, A.P., Tommasi, A., Cordier, P., Sobolev,A.V., 2010. Olivine, and the origin of kimberlite. Journal of Petrology 51, 573602.

Bardet, M.G., 1977. Geologie du diamant. Troisieme partie, gisements de diamantsd'Asia, d'Amerique, d'Europe et d'Australasie. BRGM, Memoires 83, 169.

Barker, D.S., 2001. Calculated silica activities in carbonatitic liquids. Contributions toMineralogy and Petrology 141, 704709.

Becker, M., le Roex, A.P., 2006. Geochemistry of south African on- and off-craton, groupI and group II kimberlites: petrogenesis and source region evolution. Journal of Pe-trology 47, 673703.

Bennio, L., Brotzu, P., Gomes, C.B., D'Antonio, M., Lustrino, M., Melluso, L., Morbidelli, L.,Ruberti, E., 2002. Petrological, geochemical and SrNd isotopic features of alkalinerocks from the Arraial do Cabo Frio peninsula (Southeastern Brazil). Periodico diMineralogia 71, 137158.

Bennio, L., Brotzu, P., D'Antonio, M., Feraud, G., Gomes, C.B., Marzoli, A., Melluso, L.,Morbidelli, L., Morra, V., Rapaille, C., Ruberti, E., 2003. The tholeiitic dyke swarmof the Arraial do Cabo peninsula (SE Brazil): 39Ar/40Ar ages, petrogenesis and re-gional signicance. Journal of South American Earth Sciences 16, 163176.

Biondi, J.C., 2005. Brazilianmineral deposits associatedwith alkaline and alkaline-carbonatitecomplexes. In: Comin-Chiaramonti, P., Gomes, C.B. (Eds.), Mesozoic to Cenozoic AlkalineMagmatism in the Brazilian Platform. FAPESP, So Paulo, Brazil, pp. 707750.

Birkett, T.C., McCandless, T.E., Hood, C.T., 2004. Petrology of the Renard igneous bodies: hostrocks for diamond in the northern Otish Mountains region, Quebec. Lithos 76, 475490.

Bizzi, L.A., De Wit, M.J., Smith, C.B., McDonald, I., Armstrong, R.A., 1995. Heterogeneousenriched mantle materials and Dupal-type magmatism along the SWmargin of theSo Francisco Craton, Brazil. Journal of Geodynamics 20, 469491.

Boynton, W.V., 1984. Cosmochemistry of the rare earth elements: meteorite studies. In:Henderson, P. (Ed.), Rare Earth ElementGeochemistry. Elsevier, Amsterdam, pp. 63114.

Brey, G.P., Bulatov, V.K., Girnis, A.V., Lahaye, Y., 2008. Experimental melting of carbon-ated peridotite at 610 GPa. Journal of Petrology 49, 797821.

Brod, J.A., Gibson, S.A., Thompson, R.N., Junqueira-Brod, T.C., Seer, H.J., De Moraes, L.C.,Boaventura, G.R., 2000. The kamafugite-carbonatite association in the AltoParanaba Igneous Province (APIP) southeastern Brazil. Revista Brasileira deGeociencias 30, 408412.

Brod, J.A., Gaspar, J.C., de Arajo, D.P., Gibson, S.A., Thompson, R.N., Junqueira-Brod,T.C., 2001. Phlogopite and tetra-ferriphlogopite from Brazilian carbonatitecomplexes: petrogenetic constraints and implications for mineral-chemistry sys-tematics. Journal of Asian Earth Sciences 19, 265296.

Brotzu, P., Melluso, L., D'Amelio, F., Lustrino, M., 2005. Mac/ultramac potassic dykesand felsic intrusions of the Serra do Mar igneous province (SE-Brazil): a review. In:Comin-Chiaramonti, P., Gomes, C.B. (Eds.), Mesozoic to Cenozoic alkalinemagmatism in the Brazilian platform. FAPESP, So Paulo, Brazil, pp. 443472.

Brotzu, P., Melluso, L., Bennio, L., Gomes, C.B., Lustrino, M., Morbidelli, L., Morra, V.,Ruberti, E., Tassinari, C.C.G., D'Antonio, M., 2007. Petrogenesis of the Cenozoic po-tassic alkaline complex of Morro de So Joo, southeastern Brazil. Journal of SouthAmerican Earth Sciences 24, 93115.

Bulanova, G.P., Walter, M.J., Smith, C.B., Kohn, S.C., Armstrong, L.S., Blundy, J., Gobbo, L.,2010. Mineral inclusions in sublithospheric diamonds from Collier 4 kimberlitepipe, Juina, Brazil: subducted protoliths, carbonated melts and primary kimberlitemagmatism. Contributions to Mineralogy and Petrology 160, 489510.

Carlson, R.W., Esperana, S., Svisero, D.P., 1996. Chemical and Os isotopic study of Cre-taceous potassic rocks from southern Brazil. Contributions to Mineralogy and Pe-trology 125, 393405.

Carlson, R.W., Arajo, A.L.N., Junqueira-Brod, T.C., Gaspar, J.C., Brod, J.A., Petrinovic, I.A.,Holland, M.H.B.M., Pimentel, M.M., Sichel, S., 2007. Chemical and isotopic relation-ships between peridotite xenoliths and macultrapotassic rocks from SouthernBrazil. Chemical Geology 242, 418437.

Cellai, D., Conticelli, S., Menchetti, S., 1994. Crystal-chemistry of clinopyroxenes frompotassic and ultrapotassic rocks in central Italy: implications on their genesis. Con-tributions to Mineralogy and Petrology 116, 301315.

Clark, T.C., 1994. An integrated geochemical and isotopic study of the Prieska Province kim-berlites from the Republic of South Africa. M.Sc. thesis, University of theWitwatersrand.

Clement, C.R., 1982. A comparative geological study of some major kimberlite pipes innorthern Cape and Orange Free State. Ph.D. thesis, University of Cape Town.

Coe, N., 2004. Petrogenesis of the Swartruggens and Star Group II kimberlite dykeswarms, South Africa. M.Sc. thesis, University of Cape Town, 146 pp.Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, ChemConrad, C.P., Wu, B., Smith, E.I., Bianco, T.A., Tibbetts, A., 2010. Shera-driven upwellinginduced by lateral viscosity variations and asthenospheric shear: a mechanism forintraplate volcanism. Physics of the Earth and Planetary Interiors 178, 162175.

D'Agrella-Filho, M.S., Trindade, R.I.F., Tohver, E., Janikian, L., Teixeira, W., Hall, C., 2011.Paleomagnetism and 40Ar/39Ar geochronology of the high-grade metamorphicrocks of the Jequi block, So Francisco Craton: Atlantica, Ur and beyond. Precam-brian Research 185, 183201.

Dalton, J.A., Presnall, D.C., 1998a. The continuum of primary carbonatitic kimberliticmelt compositions in equilibrium with lherzolite: data from the system CaO-MgO-Al2O3-SiO2-CO2 at 6 GPa. Journal of Petrology 39, 19531964.

Dalton, J.A., Presnall, D.C., 1998b. Carbonatitic melts along the solidus of modellherzolite in the system CaO-MgO-Al2O3-SiO2-CO2 from 3 to 7 GPa. Contributionsto Mineralogy and Petrology 131, 123135.

Dasgupta, R., Hirschmann, M.M., Smith, N.D., 2007. Partial melting experiments ofperidotite+CO2 at 3 GPa and genesis of alkalic ocean island basalts. Journal of Pe-trology 48, 20932124.

Dasgupta, R., Hirschmann, M.M., McDonough, W.F., Spiegelman, M., Withers, A.C.,2009. Trace element partitioning between garnet lherzolite and carbonatite at6.6 and 8.6 GPa with applications to the geochemistry of the mantle and ofmantle-derived melts. Chemical Geology 262, 5777.

Davis, G.L., 1977. The ages and uranium contents of zircons from kimberlites and asso-ciated rocks. Carnegie Institute of Washington Yearbook 76, 631635.

Dawson, J.B., Smith, J.V., Jones, A.P., 1985. A comparative study of bulk rock andmineralchemistry of olivine melilitites and associated rocks from East and South Africa.Neues Jahrbuch fr Mineralogie, Abhandlungen 152, 143175.

De Mulder, M., 1985. The Karisimbi Volcano, (Virunga). Musee Royale de l'AfriqueCentrale, Tervuren, Belgique Annales, Serie in Octavo Sciences Geologique, 90.

Donatti-Filho, J.P., Tappe, S., Oliveira, E.P., Heaman, L., this issue. Age and origin of theNeoproterozoic Brauna kimberlites: magma generation within the metasomatizedbase of the So Francisco craton, Brazil. Chemical Geology.

Donnelly, C.L., Grifn, W.L., O'Reilly, S.Y., Pearson, N.J., Shee, S.R., 2011. The Kimberlitesand related rocks of the Kuruman Kimberlite Province, Kaapvaal Craton, South Af-rica. Contributions to Mineralogy and Petrology 161, 351371.

Eggler, D.H., 1989. Kimberlites: How do they form? in Kimberlite and Related Rocks,vol. 1. In: Ross, J., et al. (Ed.), Their Composition, Occurrence, Origin and Emplace-ment. Blackwell Sci, Malden, Mass, pp. 489504.

Foley, S.F., Venturelli, G., Green, D.H., Toscani, L., 1987. The ultrapotassic rocks: charac-teristics, classication and constraints for petrogenetic models. Earth-Science Re-views 24, 81134.

Foley, S.F., Yaxley, G.M., Rosenthal, A., Buhre, S., Kiseeva, E., Rapp, R.P., Jacob, D.E., 2009.The composition of near-solidus melts of peridotite in the presence of CO2 and H2Oat 40 and 60 kbar. Lithos 112, 274283.

Foley, S.F., Link, K., Tiberindwa, J.V., Barifaijo, E., 2012. Patterns and origin of igneous ac-tivity around the Tanzanian craton. Journal of African Earth Sciences 62, 118.

Fraser, K.J., Hawkesworth, C.J., 1992. The petrogenesis of group 2 ultrapotassic kimber-lite from Finsch mine, South Africa. Lithos 28, 327345.

Gibson, S.A., Thompson, R.N., Leonardos, O.K., Turner, S.E., Mitchell, J.G., Dickin, A.P.,1994. The Serra do Bueno potassic diatreme a possible hypabyssal equivalentof the ultramac alkaline volcanics in the late Cretaceous Alto Paranaba IgneousProvince, SE Brazil. Mineralogical Magazine 58, 357373.

Gibson, S.A., Thompson, R.N., Leonardos, O.H., Dickin, A.P., Mitchell, J.G., 1995. The LateCretaceous impact of the Trindade mantle plume: evidence from large-volume,mac, potassic magmatism in SE Brazil. Journal of Petrology 36, 189229.

Gibson, S.A., Thompson, R.N., Dickin, A.P., Mitchell, J.G., 1997. Temporal variation inmagma sources related to the impact of the Tristan mantle plume. In: Hatton, C.J.(Ed.), International Symposium on plumes, plates and mineralisation, Pretoria,Abstracts, pp. 3738.

Gibson, S.A., Thompson, R.N., Day, J.A., 2006. Timescales and mechanisms of plumelithosphere interactions: 40Ar/39Ar geochronology and geochemistry of alkalineigneous rocks from the Paran-Etendeka large igneous province. Earth and PlanetaryScience Letters 251, 117.

Gomes, C.B., Comin-Chiaramonti, P., 2005. Some notes on the Alto Paranaba igneousprovince. In: Comin-Chiaramonti, P., Gomes, C.B. (Eds.), Mesozoic to Cenozoic AlkalineMagmatism in the Brazilian Platform. FAPESP, So Paulo, Brazil, pp. 317340.

Guarino, V., Azzone, R.G., Brotzu, P., Gomes, C.B., Melluso, L., Morbidelli, L., Ruberti, E.,Tassinari, C.C.G., Brilli, M., 2012. Magmatism and fenitization in the Cretaceouspotassium-alkaline-carbonatitic complex of Ipanema So Paulo State, Brazil. Min-eralogy and Petrology 104, 4361.

Gudnnsson, G.H., Presnall, D.C., 2005. Continuous gradations among primary car-bonatitic, kimberlitic, melilititic, basaltic, picritic, and komatiitic melts in equilibri-um with garnet lherzolite at 38 GPa. Journal of Petrology 46, 16451659.

Halliday, A.N., Davies, G.R., Lee, D., Tommasini, S., Paslick, C.R., Fitton, J.G., James, D.E.,1992. Lead isotope evidence for young trace element enrichment in the oceanicupper mantle. Nature 359, 623627.

Hirose, K., 1997. Partial melt compositions of carbonated peridotite at 3 GPa and role ofCO2 in alkali-basalt magma generation. Geophysical Research Letters 24, 28372840.

Kamenetsky, V.S., Kamenetsky, M.B., Sobolev, A.V., Golovin, A.V., Sharygin, V.V.,Pokhilenko, N.P., Sobolev, N.V., 2009. Can pyroxenes be liquidus minerals in thekimberlite magma? Lithos 112, 213222.

Kamenetsky, V.S., Kamenetsky, M.B., Golovin, A.B., Sharygin, V.V., Maas, R., in press. Ultra-fresh salty kimberlite of the Udachnaya-East pipe (Yakutia, Russia): a petrologicaloddity or fortuitous discovery? Lithos, http://dx.doi.org/10.1016/j.bbr.2011.03.031.

Kapsiotis, A., Grammatikopoulos, T.A., Tsikouras, B., Hatzipanagiotou, K., 2009.Chromian spinel composition and platinum-group element mineralogy ofchromitites from the Milia Area, Pindos Ophiolite Complex, Greece. The CanadianMineralogist 47, 10371056.e geochemistry, and petrogenesis of kimberlites, kamafugites and. Geol. (2012), doi:10.1016/j.chemgeo.2012.06.016

l Geology xxx (2012) xxxxxxKjarsgaard, B.A., Pearson, D.G., Tappe, S., Nowell, G.M., Dowall, D., 2009. Geochemistryof hypabyssal kimberlites from Lac de Gras, Canada: comparisons to a global data-base and applications to the parent magma problem. Lithos 112, 236248.

Le Maitre, R.W., 2002. A classication of igneous rocks and glossary of terms. Cam-bridge University Press, Cambridge, U.K.

le Roex, A.P., Bell, D.R., Davis, P., 2003. Petrogenesis of Group I kimberlites from Kim-berley, South Africa: evidence from bulk-rock geochemistry. Journal of Petrology44, 22612286.

Leake, B.E., Woolley, A.R., Arps, C.E.S., Gilbert, M.C., Grice, J.D., Hawthorne, F.C., Kato, A.,Kisch, A.J., Krivovichev, V.G., Linthout, K., Laird, J., Mandarino, J.A., Maresch, W.V.,Nickel, E.H., Schumacher, J.C., Smith, D.C., Stephenson, N.C.N., Whittaker, E.J.W.,Youzhi, G., 1997. Nomenclature of amphiboles: report of the subcommittee on am-phiboles of the international mineralogical association, commission on new min-erals and minerals names. The Canadian Mineralogist 35, 219246.

Leonardos, O.H., Ulbrich, M.N., Gaspar, J.C., 1991. The Mata da Corda volcanic rocks. In:Leonardos, O.H., Meyer, H.O.A., Gaspar, J.C. (Eds.), Field Guidebook. 5th Interna-tional Kimberlite Conference, Arax. : Special Publication, 3/91. CPRM, Braslia,DF, pp. 7579.

Li, Q.-L., Li, X.-H., Liu, Y., Wu, F.-Y., Yang, J.-H., Grifn, W.L., Mitchell, R.H., 2010. PreciseUPb and ThPb age determination of kimberlitic perovskites secondary ion massspectrometry. Chemical Geology 269, 396405.

Litasov, K.D., 2011. Physicochemical conditions for melting in the Earth's mantle con-taining a COH uid (from experimental data). Russian Geology and Geophysics52 (475), 492.

Litasov, K.D., Ghosh, S., Ohtani, E., 2008. Melting of carbonated peridotite and eclogiteto 32GPa and implication to origin of carbonatite- and kimberlite-like magmas inthe deep mantle. 9th International Kimberlite Conference Extended Abstract.9IKC-A-00143.

Locock, A.J., 2008. An Excel spreadsheet to recast analyses of garnet into end-membercomponents, and a synopsis of the crystal chemistry of natural silicate garnets.Computers and Geosciences 34, 17691780.

Lustrino, M., Dallai, L., Giordano, R., Gomes, C.B., Melluso, L., Morbidelli, L., Ruberti, E.,Tassinari, C.C.G., 2003. Geochemical and SrNd-O isotopic features of the Poosde Caldas alkaline massif (SP-MG, SE Brazil): relationships with the Serra do Maranalogues. Short Papers IV South American Symposium on Isotope Geology.

Lyubetskaya, T., Korenaga, J., 2007. Chemical composition of Earth's primitive mantleand its variance: 1. methods and results. Journal of Geophysical Research 112,B03211, http://dx.doi.org/10.1019/2005JB004223 (21 pp.).

Machado, D.L., 1991. Geologia e aspectos metalogenticos do complexo alcalino-carbonatitico de Catalo II (GO), thse, Universidade Estadual de Campinas (SP),Brsil, 102 pp.

Melluso, L., Morra, V., Brotzu, P., Tommasini, S., Renna, M.R., Duncan, R.A., Franciosi, L.,D'Amelio, F., 2005. Geochronology and petrogenesis of the Cretaceous AntampombatoAmbatovy complex and associated dyke swarm, Madagascar. Journal of Petrology46, 19631996.

Melluso, L., Lustrino, M., Ruberti, E., Brotzu, P., Gomes, C.B., Morbidelli, L., Morra, V.,Svisero, D.P., d'Amelio, F., 2008. Major- and trace-element composition of olivine,perovskite, clinopyroxene, Cr-Fe-Ti oxides, phlogopites and host kamafugites andkimberlites, Alto Paranab, Brazil. The Canadian Mineralogist 46, 1940.

Melluso, L., Srivastava, R.K., Guarino, V., Zanetti, A., Sinha, A.K., 2010. Mineral composi-tions and petrogenetic evolution of the ultramac-alkaline carbonatitic complexof Sung Valley, northeastern India. The Canadian Mineralogist 48, 205229.

Melluso, L., le Roex, A.P., Morra, V., 2011. Petrogenesis and Nd-Pb-Sr- isotope geochem-istry of the olivine melilitites and olivine nephelinites (ankaratrites) in Madagas-car. Lithos 127, 505521, http://dx.doi.org/10.1016/j.lithos.2011.08.003.

Mitchell, R.H., 1986. Kimberlites: mineralogy, geochemistry and petrology. Plenum,New York.

Mitchell, R.H., 1995. Kimberlites, Orangeites and Related Rocks. Plenum, New York, NY.410 pp.

Mitchell, R.H., Tappe, S., 2010. Discussion of Kimberlites and aillikites as probes of the con-tinental lithospheric mantle, by D. Francis andM. Patterson (Lithos v. 109, p. 7280).Lithos 115, 288292.

Nielsen, T.F.D., Sand, K.K., 2008. The Majuagaa kimberlite dike, Maniitsoq region, WestGreenland: constraints on anMg-rich silicocarbonatiticmelt composition fromground-mass mineralogy and bulk compositions. The Canadian Mineralogist 46, 10431061.

Nielsen, T.F.D., Jensen, S.M., Secher, K., Sand, K.K., 2009. Distribution of kimberlite andaillikite in the Diamond Province of southern West Greenland: a regional perspectivebased on groundmassmineral chemistry andbulk compositions. Lithos 1125, 358371.

Nowell, G.M., Pearson, D.G., Bell, D.R., Carlson, R.W., Smith, C.B., Kempton, P.D., Noble,S.R., 2004. Hf isotope systematics of kimberlites and their megacrysts: new con-straints on their source regions. Journal of Petrology 45, 15831612.

Nowicki, T., Porritt, L., Crawford, B., Kjarsgaard, B., 2008. Geochemical trends in kimber-lites of the Ekati property, Northwest Territories, Canada: insights on volcanic andresedimentation processes. Journal of Volcanology and Geothermal Research 174,117127.

O'Brien, H.E., Irving, A.J., McCallum, I.S., Thirlwall, M.F., 1995. Strontium, neodymium, andlead isotopic evidence for the interaction of post-subduction asthenospheric potassicmac magmas of the Highwood Mountains, Montana, USA, with ancient Wyomingcraton lithospheric mantle. Geochimica et Cosmochimica Acta 59 (21), 45394556.

Paton, C., Hergt, J.M., Phillips, D., Woodhead, J.M., Shee, S.R., 2007. New insights into thegenesis of Indian kimberlites from the Dharwar craton via in situ Sr isotope analy-sis of groundmass perovskite. Geology 35, 10111014.

Peucat, J.-J., Salomo, J., Barbosa, F., de Arajo Pinho, I.C., Paquette, J.-L., Martin, H.,Fanning, M., de Menezes Leal, A.B., Cruz, S., 2011. Geochronology of granulites fromthe south Itabuna-Salvador-Cura Block, So Francisco Craton (Brazil): Nd isotopesand UPb zircon ages. Journal of South American Earth Sciences 31, 397413.Please cite this article as: Guarino, V., et al., U Pb ages, Sr Nd isotopphlogopite picrites of the Alto Paranaba Igneous Province, Brazil, ChemPrelevic, D., Foley, S.F., Romer, R.L., Cvetkovic, V., Downes, H., 2005. Tertiary ultrapotassicvolcanism in Serbia: constraints on petrogenesis and mantle source characteristics.Journal of Petrology 46, 14431487.

Read, G., Grutter, H., Winter, S., Luckman, N., Gaunt, F., Thomsen, F., 2004. Stratigraphicrelations, kimberlite emplacement and lithospheric thermal evolution, QuiricBasin, Minas Gerais State, Brazil. Lithos 77, 803818.

Rosenthal, A., Foley, S.F., Pearson, D.G., Nowell, D.G., Tappe, S., 2009. Petrogenesis ofstrongly alkaline primitive volcanic rocks at the propagating tip of the westernbranch of the East African Rift. Earth and Planetary Science Letters 284, 236248.

Sahama, Th.G., 1974. Potassium-rich alkaline rocks. In: Srensen, H. (Ed.), The AlkalineRocks. Wiley, London, U.K, pp. 96109.

Sgarbi, P.B.A., Gaspar, J.C., Valena, J.G., 2000. Brazilian Kamafugites. Revista Brasileirade Geociencias 30, 417420.

Sgarbi, P.B.A., Heaman, L.M., Gaspar, J.C., 2004. UPb perovskite ages for Braziliankamafugitic rocks: further support for a temporal link to a mantle plume hotspottrack. Journal of South American Earth Sciences 16, 715724.

Smith, C.B., 1983. Pb, Sr and Nd isotopic evidence for sources of southern African Cre-taceous kimberlites. Nature 304, 5154.

Sonoki, I.K., Garda, G.M., 1988. Idades KAr de rochas alcalinas do Brasil Meridional eParaguai Oriental. Compilao e adaptao s novas constantes de decaimento.Boletim Instito de Geociencias Universidade So Paulo 19, 6385.

Tainton, K.M., 1992. The petrogenesis of group-2 kimberlites and lamproites from theNorthern Cape Province, South Africa. Ph.D. thesis, Cambridge.

Tainton, K.M., McKenzie, D.P., 1994. The generation of kimberlites, lamproites, andtheir source rocks. Journal of Petrology 35, 787817.

Tappe, S., Simonetti, A., 2012. Combined UPb geochronology and SrNd isotope anal-ysis of the Ice River perovskite standard, with implications for kimberlite and alka-line rock petrogenesis. Chemical Geology 304305, 1017.

Tappe, S., Foley, S.F., Pearson, D.G., 2003. The kamafugites of Uganda: a mineralogicaland geochemical comparison with their Italian and Brazilian analogues. Periodicodi Mineralogia 72, 5177.

Tappe, S., Jenner, G.A., Foley, S.F., Heaman, L.M., Besserer, D., Kjarsgaard, B.A., Ryan, B.,2004. Torngat ultramac lamprophyres and their relation to the North Atlantic Al-kaline Province. Lithos 76, 491518.

Tappe, S., Foley, S.F., Jenner, G.A., Kjarsgaard, B.A., 2005. Integrating ultramaclamprophyres into the IUGS classication of igneous rocks: rational and implica-tions. Journal of Petrology 46, 18931900.

Tappe, S., Foley, S.F., Jenner, G.A., Heaman, L.M., Kjarsgaard, B.A., Romer, R.L., Stracke, A.,Joyce, N., Hoefs, J., 2006. Genesis of ultramac lamprophyres and carbonatites atAillik Bay, Labrador: a consequence of incipient lithospheric thinning beneaththe North Atlantic Craton. Journal of Petrology 46, 12611315.

Tappe, S., Foley, S.F., Stracke, A., Romer, R.L., Kjarsgaard, B.A., Heaman, L.M., Joyce, N.,2007. Craton reactivation on the Labrador Sea margins: 40Ar/39Ar age and SrNdHfPb isotope constraints from alkaline and carbonatite intrusives. Earth andPlanetary Science Letters 256, 433454.

Tappe, S., Foley, S.F., Kjarsgaard, B

Documents

U–Pb ages, Sr–Nd- isotope geochemistry, and petrogenesis of kimberlites, kamafugites and phlogopite-picrites of the Alto Paranaíba Igneous Province, Brazil.pdf