The magmatic evolution of the Midcontinent rift: New ...mille066/Teaching/5100_07/Articles/Vervoort et al... · and Machado, 1992; Davis and Paces, 1990; Paces and Miller, 1993; Davis

o

ocy

0

Precambrian Research 157 (2007) 235–268

The magmatic evolution of the Midcontinent rift: Newgeochronologic and geochemical evidence from felsic magmatism

Jeff D. Vervoort a,∗, Karl Wirth b, Bryan Kennedy b,1,Travis Sandland b,1, Karen S. Harpp c

a School of Earth and Environmental Sciences, Washington State University, Pullman, WA 99164, USAb
Department of Geology, Macalester College, St. Paul, MN 55000, USAc Department of Geology, Colgate University, Hamilton, NY 13346, USA
Received 18 May 2006; received in revised form 26 November 2006; accepted 5 February 2007

Abstract

The Midcontinent Rift System (MRS) is an example of a mafic-dominated continental rift where silicic magmatism is locallysignificant. The best-preserved example of this is on the NE limb of the rift along the North Shore of Lake Superior whererhyolites comprise a large percentage (up to 25%; Green, J.C., Fitz, T.J., III, 1993. Extensive felsic lavas and rheoignimbrites in theKeweenawan Midcontinent rift plateau volcanics, Minnesota: petrographic and field recognition. J. Volcanol. Geotherm. Res. 54,177–196) of lava flows and where hypabyssal granophyric intrusive complexes are common. In this paper, we report U–Pb zirconages, Nd isotopic compositions, and major and trace-element data for seven granophyric complexes of the MRS exposed in NEMinnesota.

U–Pb zircon ages for the granophyres define two different age groups: an older group with ages from 1109 to 1106 Ma; and ayounger group with ages from 1099 to 1095 Ma. These ages coincide with the “early” and “main” magmatic stages of Midcontinentrift evolution suggested by Miller and Vervoort [Miller, J.D., Jr., Vervoort, J.D., 1996. The latent magmatic stage of the Midcontinentrift: a period of magmatic underplating and melting of the lower crust. Inst. Lake Superior Geol., 42nd Ann. Mtg., Proceedings, vol.42, pp. 33–35] and Miller and Severson [Miller, J.D., Jr., Severson, M.J., 2002. Geology of the Duluth Complex. In: Miller, J.D., Jr.,Green, J.C., Severson, M.J., Chandler, V.W., Hauck, S.A., Peterson, D.M., and Wahl, T.E. (Eds.), Geology and Mineral Potentialof the Duluth Complex and Related Rocks of Northeastern Minnesota. Minnesota Geological Survey Report of Investigations, vol.58, pp. 106–143]. Although the two groups of granophyres have similar major and trace-element compositions, they have differentNd isotopic compositions. The older, or “early stage”, granophyres have more radiogenic Nd isotopic compositions (εNd(i) = −3.7 to−0.5) whereas the younger, or “main stage”, granophyres have more crustal, unradiogenic Nd isotopic compositions (εNd(i) = −7.6 to−3.1). The age correlative Nd isotopic signatures of the granophyres are broadly consistent with the ages and isotopic compositions

f the rhyolites within the North Shore Volcanic Group (NSVG) and illustrate the episodic nature of the Midcontinent rift evolution.The early stage (1109–1106 Ma) of MRS magmatism is characterized by mafic mantle-derived magmas with minor amounts

f silicic magmas. The evolved magmas were likely derived by partial melting of either earlier formed rift related rocks or olderrust with near chondritic Nd isotopic composition. This was followed by a period of relative quiescence lasting about 5 millionears from which no significant MRS magmatism has been preserved. The main stage of MRS magmatism resumes at ca. 1100 Ma

∗ Corresponding author. Tel.: +1 509 335 5597.E-mail address: [email protected] (J.D. Vervoort).

1 Present address: Science Museum of Minnesota, 120 West Kellogg Blvd., Saint Paul, MN 55102, USA.

301-9268/$ – see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.precamres.2007.02.019

mailto:[email protected]

dx.doi.org/10.1016/j.precamres.2007.02.019

236 J.D. Vervoort et al. / Precambrian Research 157 (2007) 235–268

with voluminous basaltic volcanism and mafic intrusions. The silicic magmas produced during this stage are more abundant anddistinctly more crustal in character than during the early magmatic stage magmas. We suggest that these silicic magmas have beenderived from partial melts of more evolved crustal sources, perhaps at higher levels in the crust.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Granophyre; Midcontinent rift; Keweenawan; Granite; Mesoproterozoic; Crustal evolution; Lake superior region; Precambrian; U–Pb
geochronology; Sm–Nd isotopes
1. Introduction

The magmatism associated with continental rifts andcontinental flood basalt provinces throughout Earth’shistory is dominantly basaltic. In many of these geo-logical settings, silicic magmatism is largely absent(e.g., Columbia River Basalts; Hooper, 1997) or greatlysubordinate, while in others (e.g., Snake River Plain;Bonnichsen and Kauffman, 1987) it can be volumetri-cally significant. Silicic magmatism may be manifestin different geological settings through both high-temperature extrusive (flow, pyroclastic, rheoignimbrite)and intrusive (dike, sill, pluton) activity and canoccur coincident with all stages of mafic magma-tism, from early to late (Bryan et al., 2002). Theorigin of silicic magmas in these settings is enig-matic and despite research spanning several decades,we still do not have a complete understanding ofhow these magmas form (e.g., crustal melting, assim-ilation, fractionation), where the primary sites ofmagma generation are (upper crust, middle crust, atthe base of the crust, or mantle), or how exten-sively the magmas have been modified during transport.The reason for this complexity is that many differ-ent processes are likely to be involved and theseare manifest in different ways in different geologicalsettings. Nevertheless, constraining the genetic relation-ships between the felsic and mafic magmas is crucial forunderstanding the origin and evolution of silicic mag-matism.

The Midcontinent Rift System (MRS) is an exam-ple of a mafic-dominated system where coeval silicicmagmatism is locally significant. Throughout much ofthe MRS, magmatism was dominantly mafic with onlyminor amounts of evolved compositions (e.g., Moreyand Mudrey, 1972; Basaltic Volcanism Study Project,1981; Green, 1982a; Brannon, 1984; Dosso, 1984; Paces,1988; Nicholson and Shirey, 1990; Lightfoot et al.,
1991; Klewin and Berg, 1991; Shirey et al., 1994; Wirthet al., 1997; Vervoort and Green, 1997). An excep-tion to this is along the northeast limb of the MRS innortheastern Minnesota where rhyolites comprise a sig-
nificant percentage (10–25%; Green and Fitz, 1993) ofthe total volume of erupted material. The Lake Superiorregion also has excellent exposures of MRS intrusiverocks. The Duluth and Beaver Bay complexes wereintruded beneath and into the extrusive rocks of therift in a series of complex subvolcanic units (Miller etal., 2002). These are dominantly gabbroic in composi-tion but range from troctolite to anorthosite to granite(granophyres). The large percentage of silicic composi-tions, in conjunction with the occurrence of intrusive andextrusive MRS rocks exposed at different crustal levels,provides an excellent opportunity for a detailed exam-ination of the relationships between felsic and maficmagmatism. The mafic and felsic intrusive and extru-sive rocks within the Midcontinent Rift System may bethe best exposed of all continental rifts. In this paperwe report major and trace-element, Nd isotopic, andU–Pb zircon data from a series of granophyre intrusionsand related rocks from the Midcontinent Rift System innortheast Minnesota. Specifically, we analyzed 81 sam-ples from seven granophyre intrusions as well as ninesamples from closely related volcanic and hypabyssalplutonic rocks. The first order goal of this investiga-tion is to constrain the genesis of rhyolitic magmas inmafic-dominated continental rift systems. This will alsoallow us to address other second-order questions relatedto the Midcontinent Rift System and rhyolite–granitegenesis in continental rifts: (1) what is the mechanismby which felsic magmas are produced in the MRS?(2) What is the age relationship between mafic andfelsic magmatism in the MRS? And finally (3) whatdo these temporal and compositional relationships tellus about the overall magmatic and chemical evolutionof the MRS? This study follows a similar Nd isotopeand trace-element study of the rhyolites of the NorthShore Volcanic Group (NSVG) exposed along the northshore of Lake Superior in Minnesota (Vervoort andGreen, 1997). We will show that the granophyre Nd
data are consistent with results from the rhyolite studyand together demonstrate an increasing role of evolvedcrustal sources in the generation of silicic magmas withtime.

rian R

2

2

oApK(cluGe1flfi1eetit1aM1

rN2sra1PVgpsMiSt(

2R

M

J.D. Vervoort et al. / Precamb

. Geological background

.1. Geological setting of the MCR

The Midcontinent Rift System is the most conspicu-us feature on gravity and aeromagnetic maps of Northmerica where it appears as prominent patterns of geo-hysical anomalies extending northward from centralansas to Lake Superior, and eastward into Michigan

Fig. 1). The origin of these midcontinent geophysi-al anomalies (e.g., King and Zietz, 1971) has beenargely attributed to ‘incipient rifting’ in response to anpwelling mantle plume (e.g., Burke and Dewey, 1973;reen, 1983; Van Schmus and Hinze, 1985; Hutchinson

t al., 1990; Nicholson and Shirey, 1990; Shirey et al.,994; Nicholson et al., 1997). At least 15 km of volcanicows are interpreted from geophysical studies to havelled the rift basin (Behrendt et al., 1988; Cannon et al.,989; Hinze et al., 1990; Hutchinson et al., 1990; Allent al., 1997). The total volume of MRS volcanic rocks isstimated to be ∼2 × 106 km3 (Cannon, 1992), althoughhe rift is now buried by Paleozoic rocks along much ofts extent. The overwhelming majority of MRS magma-ism, based on U–Pb geochronology, occurred between109 and 1086 Ma (Davis and Sutcliffe, 1985; Heamannd Machado, 1992; Davis and Paces, 1990; Paces andiller, 1993; Davis and Green, 1997; Zartman et al.,

997).In the Lake Superior region, the Midcontinent

ift formed in Archean and Proterozoic rocks of theorth American craton. These consisted of northern.8–2.65 Ga granite-greenstone terranes (Card, 1990),outhern 3.5–2.55 Ga high-grade granite-gneiss ter-anes (Bickford et al., 2006; Schmitz et al., 2006),nd early Proterozic sediments (e.g., Hemming et al.,995) related to the Penokean Orogen. During theenokean Orogeny (ca. 1.85 Ga; Van Schmus, 1980;an Wyck and Johnson, 1997) the southern Archeanneiss terranes were variably deformed and metamor-hosed, and thick sequences of synorogenic foredeepediments were deposited in northern Wisconsin and

innesota. In northeastern Minnesota, basal volcanic,ntrusive, and sedimentary rocks of the Midcontinent Riftystem directly overlie Neoarchean granite-greenstone

erranes and Paleoproterozoic metasedimentary rocksFig. 1).

.2. Character of magmatism of the Midcontinent
ift System
Volcanic, sedimentary, and plutonic rocks of theidcontinent Rift System are well exposed in the

esearch 157 (2007) 235–268 237

Lake Superior region (Fig. 1) and are divided intothe Keweenawan Supergroup and Midcontinent riftIntrusive Supersuite, respectively (Miller et al., 2002).Rocks of the Keweenawan Supergroup have largelybeen divided into lower and upper units based onpaleomagnetic polarity, which changed from reversedto normal magnetic polarity at about 1105–1102 Ma(Green, 1982b; Van Schmus et al., 1982; Halls andPesonen, 1982; Miller et al., 1995; Davis and Green,1997). Volcanism throughout the Midcontinent riftwas bimodal and tholeiitic in composition, although itis predominantly mafic (e.g., Green, 1982a; BasalticVolcanism Study Project, 1981; Green, 1983). Iso-topic evidence has been interpreted to indicate varyingcontributions from asthenosphere, lithospheric man-tle, and crustal sources (Brannon, 1984; Dosso, 1984;Sutcliffe, 1987; Paces and Bell, 1989; Nicholson andShirey, 1990; Klewin and Berg, 1991; Lightfoot etal., 1991; Shirey et al., 1994; Nicholson et al., 1997;Wirth et al., 1997; Vervoort and Green, 1997). Silicicrocks occur as rhyolitic flows and granitic intrusionsand typically comprise <10% of the volcanic sectionin central graben of the MRS Lake Superior region(Nicholson, 1992). In the Portage Lake Volcanics ofthe Keweenawan Peninsula region (MI), rhyolites aredivided into two groups based on mineralogy andgeochemical composition (Nicholson, 1992), and areinterpreted to have originated by contamination of tholei-itic melts with Archean crust and partial melting ofArchean crust (Nicholson and Shirey, 1990). Rhyoliticrocks of the North Shore Volcanic Group in northeast-ern Minnesota are notable because of their thickness(up to 250 m), lateral extent (up to 40 km), and rela-tively high volumetric proportions (∼25%) comparedwith other parts of the MRS (Fitz, 1988; Green andFitz, 1993). In addition, nearly all NSVG rhyolites con-tain magmatic groundmass trydymite, and some containtextural evidence of rheomorphic flow after eruption asash-flow tuffs. The low viscosities required for floware attributed to high temperature, high fluorine con-tents, and low oxidation state at the time of eruption(Green and Fitz, 1993). Silicic magmatism is also man-ifest in granitic intrusions along the north shore ofLake Superior in northeastern Minnesota. Originallyknown as the “red rock” series of the Duluth Com-plex (Winchell, 1899; Grout et al., 1959), these rockswere emplaced along the basal and mid-level portionsof the North Shore Volcanic Group (Miller et al., 2002).
These granitic intrusions have similar dimensions to therhyolitic units (up to 1 km thick and 150 km2 in arealextent) and are characterized by having granophyric tex-tures.


Fig. 1. Geologic map of the Midcontinent Rift System in northeastern Minnesota showing the locations of granophyre complexes. Early Stageucturaligher st

granophyre complexes comprise the earliest phase and uppermost strassociated with younger hypabyssal intrusive complexes and occur at hMiller et al. (2001).

2.3. Nature of the granophyres

The granophyre bodies of northeastern Minnesotahave been mapped as part of various studies, but theyhave been the subject of relatively few modern petro-logic studies. There has been uncertainty of the geneticrelationship of the various granophyre units to the intru-sive units of the Duluth and Beaver Bay Complexes.Recent field, geochronologic, and aeromagnetic stud-ies (reviewed by Miller et al., 2002), however, haveconstrained the extent of the Duluth Complex and the
relationships between its constituent parts, including thegranophyric complexes. The intrusions that comprise theDuluth Complex have been classified into four seriesbased on age, magnetic polarity, lithology, and inter-
level of the Duluth Complex. Main stage granophyre complexes areructural levels within the overlying volcanic sequence. Modified from

nal structure (Miller and Severson, 2002). Intrusionsemplaced during the early stage of MRS magmatism,based on their reversed magnetic polarity, include theearly gabbro series and the felsic series (early gra-nophyres of this study). These early intrusions and therelated Logan Sills are concentrated in the eastern partsof the Duluth Complex, although many felsic seriesintrusions also occur along the roof zone of the cen-tral parts of the Complex (Fig. 1). Intrusions emplacedduring the main stage of MRS magmatism, based ontheir normal magnetic polarity, belong to the anorthositic
series and layered series (Fig. 1). Although field rela-tions consistently show that the leucogabbroic rocksof the anorthositic series were emplaced earlier thanthe discrete, differentiated intrusions of the layered

rian R

sea

watsmotpC(o1Bpsysabf

opb1L1W1gtgakrftfia

thheED(n


eries, U–Pb ages of the two series indicate both weremplaced contemporaneously at around 1099 Ma (Pacesnd Miller, 1993).

The intrusions that comprise the Duluth Complexere largely emplaced into the reversely polarized lavas

t the base of the NSVG. In contrast, the intrusions ofhe Beaver Bay Complex and related hypabyssal intru-ions were emplaced into normal polarity lavas in theedial levels of the volcanic pile (Fig. 1). The intrusions

f the Beaver Bay complex have a similar range of mafico felsic compositions as is found in the Duluth Com-lex. U–Pb ages of mafic intrusions of the Beaver Bayomplex indicate an emplacement age around 1096 Ma

Paces and Miller, 1993), which is similar to the agesf the volcanic rocks they intrude (Davis and Green,997). Existing geochronologic data indicate that theeaver Bay Complex is younger than the Duluth Com-lex. Based on field relationships, however, there areome portions of the Duluth Complex that appear to beounger than the Beaver Bay Complex. These relation-hips have not yet been confirmed with geochronologynd so it is clear that the details of the relationshipsetween these two intrusive complexes have not beenully constrained.

Granophyre bodies that roughly define the roof zonef the Duluth Complex have reverse paleomagneticolarity and include, from east to west: (1) the Cucum-er Lake granophyre (Babcock, 1959; Grout et al.,959); (2) the Misquah Hills granophyre; (3) the Wineake monzodiorite–Beth Lake granophyre (Davidson,977a,b,c,d,e; Davidson and Burnell, 1977); (4) thehitefish Lake granophyre (Boerboom and Miller,

994); (5) the Isabella granophyre; (6) the Mt. Weberranophyre (Bonnichsen, 1971; Venzke, 1994); and (7)he Fairbanks/Brimson granophyres. Most of these earlyranophyre bodies are well exposed, except the Isabelland Fairbanks/Brimson granophyre bodies, which arenown only from drill core and aeromagnetic data,espectively. Where exposed, contacts between the earlyelsic series and the early gabbroic series are grada-ional suggesting that the felsic rocks were emplacedrst, and then intruded by the gabbroic rocks (Millernd Severson, 2002).

Younger granophyre bodies that are associated withhe Beaver Bay Complex and other miscellaneousypabyssal intrusive units (Grout and Schwartz, 1933)ave normal paleomagnetic polarity and include, fromast to west: (1) Pine Mountain granophyre; (2)
agle Mountain granophyre (Davidson, 1977a,b,c,d,e;avidson and Burnell, 1977); (3) Finland granophyre
Miller et al., 1993); (4) Silver Creek diabase and gra-ophyre; and (5) Sawmill Lake gabbro and granophyre.

esearch 157 (2007) 235–268 239

These granophyre bodies were emplaced at higher struc-tural levels within the NSVG extrusive rocks, and abovethe roof zone of the Duluth Complex (Fig. 1).

Classification of the MRS granophyric rocks usingmodal proportions is complicated by the abundant (<20modal%) micrographic intergrowths of alkali feldsparand quartz that are commonly present. We have adoptedthe approach recommended by Miller et al. (2002) thatclassification of the granophyric rocks be based on CIPWnormative mineral compositions and the quartz-alkalifeldspar-plagioclase-feldspathoid (QAPF) rock classi-fication scheme of Streckeisen (1976) and LeMaitre(1989).

The granophyre bodies, although dominantly graniticin composition, are quite heterogeneous throughoutindividual intrusions. Typically the basal zones of thegranophyres are more mafic (monzodioritic) and gradeupward to granitic compositions in their middle andupper portions. The roof zones of the granophyres arealso typically more mafic in composition. It is not knownif these compositional differences might result fromsegregation during transport and emplacement, partialmelting and mixing along margins of the complexes andwallrocks, or in situ fractional crystallization and differ-entiation. Most of the MRS granophyres have not beenstudied other than as part of regional studies. Four excep-tions to this are the Eagle Mountain (Nelson, 1991),Mt. Weber (Venzke, 1994), Finland (Miller and Ripley,1996; Miller and Chandler, 1997), and Cucumber Lakegranophyres (Jerde, 2003a,b).

Nelson (1991) conducted detailed mapping, petrog-raphy, and geochemical studies to examine the originof the Eagle Mountain granophyre. Field relations indi-cate that rhyolite flows are the oldest units in theregion and that younger gabbro and granophyre wereemplaced beneath it. The lower contact of the gra-nophyre is gradational with ferrodiorite over a distanceof approximately 150 m where the two phases appearintermixed. Nelson (1991) described the Eagle Moun-tain granophyre as fine-grained and variably porphyritic,consisting of plagioclase phenocrysts and minor maficminerals (ca. 10%) in a groundmass of granophyricintergrowths surrounding euhedral albite or orthoclase.Based on geochemical arguments and the similari-ties in mineral and whole-rock compositions, Nelson(1991) suggested that the granophyre and ferrodioriteare genetically related by low-pressure (<2 kb) fractionalcrystallization. Granophyre compositions suggest that
crystallization occurred at 990 ◦C near the 1 kb graniticsystem minima (Nelson, 1991).
Venzke (1994) examined the relationships betweenseveral mafic intrusions (Greenwood Lake, Bald Eagle,

brian R
240 J.D. Vervoort et al. / Precam
and Cloquet Lake) and intervening granophyric gran-ite in the Greenwood Lake region. He concluded thatthe various units in the Greenwood Lake intrusion arerelated by fractional crystallization, but he considered itunlikely that the granophyric granites were generated byfractional crystallization of the same parental melts butinstead may have resulted from liquid immiscibility.

Miller and Ripley (1996) and Miller and Chandler(1997) studied layered intrusions of the Duluth andBeaver Bay Complexes using field, drill core, geophys-ical, and geochemical data. Although several features(e.g., transitional rock types, conformable compositionallayering, and mineral and whole-rock compositions)suggest that the layered mafic Sonju Lake Intrusionand Finland Granophyre might be related by fractiona-tion, the relatively large volume of granophyre comparedto mafic rocks argues against a petrogenetic relation-ship. Rather, these authors concluded that the FinlandGranophyre was an earlier intrusion, which served asa density barrier to the emplacement of the youngerSonju Lake mafic magma. The Sonju Lake magma con-sequently underplated, partially melted and assimilatedthe lower portions of the granite, giving rise to the tran-sitional intermediate zone that now separates the twointrusions.

Jerde (2003a,b) examined the Cucumber Lake gra-nophyre and associated Crocodile Lake gabbro to studythe relationships between these commonly associatedrocks in northeastern Minnesota. His results indicate thatthe densities of rocks in the underlying gabbro intrusionsincrease with depth below the granophyre body, consis-tent with processes of density-controlled emplacementrather than in situ fractionation processes. By this inter-pretation, the mafic and felsic rocks are unrelated exceptalong their intermediate contact zones formed by partialmelting.

3. Methods

Representative samples were collected from sevengranophyre complexes in northeastern Minnesota: fourgranophyre bodies with reversed paleomagnetic polar-ity (Cucumber Lake, Misquah Hills, Wine Lake, andMt. Weber), and three with normal paleomagnetic polar-ity (Pine Mtn., Eagle Mtn., and Finland). In two ofthe better-exposed granophyres (Cucumber Lake andMisquah Hills) we conducted detailed sampling of thegranophyres, their gradational border zones, and the
surrounding country rocks, to examine geochemicalvariations perpendicular to contacts. We also collectedsamples along-strike within several granophyre bodiesto check for lateral geochemical and isotopic variations
esearch 157 (2007) 235–268

at the same structural level within individual granophyreintrusions. Ninety samples were selected for whole-rockmajor and trace-element analysis by X-ray Fluorescence(XRF), including seventeen samples from a 125 m longdrill core into the base of the Finland Granite. Subsetsof these samples were selected for trace-element analy-ses by ICP-MS (n = 47), Nd isotope studies (n = 59) andU–Pb zircon geochronology (n = 7).

Samples were crushed in a Bico-Braun jaw crusher,separated into two aliquots using a sample splitter, andground to fine powder (<75 �m) using a shatterbox withsteel and tungsten carbide (WC) bowls. The aliquotground in WC was used for preparation of samples formajor element analyses, and the powders ground in thesteel mill were used for trace element and isotope anal-yses.

The concentrations of major elements (SiO2, TiO2,Al2O3, Fe2O3, MnO, MgO, CaO, Na2O, K2O, and P2O5)were determined by XRF from fused lithium borate glassbeads. Loss on Ignition (LOI) values were calculatedfrom the percent weight loss after ignition (1000 ◦C for1 h) relative to the dried (105 ◦C for 2 h) sample powders.One gram of ignited rock powder was mixed with fivegrams of dried lithium tetraborate/metaborate (12:22)flux, and 0.01 g ammonium nitrate (NH3NO4). The sam-ple mixture was then fused in platinum alloy (95%Pt–5% Au) crucibles with two drops 50% hydrobromicacid solution (HBr) using a Claisse® Fluxy. Fusedsamples were cast in 32 mm diameter platinum alloymolds which produced glass discs approximately 5 mmthick.

Trace-element concentrations (Co, Ni, Zn, Ga, Ba,Rb, Sr, Y, Zr, Nb, Ce, Pb, Th, and U) were determinedfrom pressed powder pellets. Ten grams of rock pow-der were combined with 15–20 drops of 2% polyvinylalcohol (PVA). The mixture was then placed in a 40 mmdiameter stainless steel die and formed into a 5 mm thickpellet by applying six tons pressure for 60 s in a manualpress.

Major and trace-element concentrations were deter-mined using a Philips PW-2400 X-ray FluorescenceSpectrometer with a Rh-anode, end window X-raytube, and Philips Super-Q analytical software atMacalester College. Elemental concentrations weredetermined by comparing X-ray intensities for eachelement in a sample unknown with those from >40reference samples from the US Geological Survey(USGS), Canadian Certified Reference materials Project
(CCRMP), Geological Survey of Japan (GSJ), and Cen-tre de Recherches Petrographiques et Geochmiques(CRPG). Trace-element concentrations were correctedfor matrix effects using the Rh tube Compton K�

rian R

s(aT

mgcgrrmspwassccGafr(

n5fritscdturaisdthNrreapav


catter peak ratio method. The analytical precision±1 sigma), calculated from analyses of 6 glass discsnd pellets of international standards, are reported inables 2 and 3.

Rare earth elements (REE) and selected trace ele-ents (Hf, Pb, Th, U) were determined for representative

ranophyre and country rock samples by inductivelyoupled plasma mass spectrometry (ICP-MS) at Col-ate University. Rocks were melted by mixing 0.4 g ofock powder (<75 �m) with 1.6 g of lithium metabo-ate flux, 0.01 g LiBr, and 0.01 g NH3NO4. The meltedixtures were poured into 100 mL of nitric acid and

tirred until dissolved. An aliquot of the dissolved sam-le (2.5 mL) was then diluted to 50 mL with deionizedater. Trace-element concentrations were determined onHewlett Packard HP4500 ICP-MS at Colgate Univer-

ity following the method of Harpp et al. (2003). Internaltandards (133Cs, 115In, 182W) were used to monitor andorrect for mass-dependent instrument drift. Elementaloncentrations were calculated by comparison with USeological Survey (DTS-1, BIR-1, W-2, G-2, STM-1)

nd CRPG (AC-E) reference materials. The precisionor most elements is estimated to be <3% based oneplicate analyses of W-2 and AC-E run as unknownsTable 3).

We determined U–Pb zircon ages for seven gra-ophyres and determined Nd isotopic compositions of1 samples from these granophyres and eight samplesrom volcanic and hypabyssal plutonic rocks spatiallyelated to the granophyres. All U–Pb and Sm–Ndsotopic measurements were performed on a VG-354hermal ionization mass spectrometer at the Univer-ity of Arizona. Procedures for zircon separation, ionhromatography, U–Pb analysis, and data reduction areescribed in Gehrels and Boghossian (2000). Conven-ional concordia plots and age calculations were donesing Isoplot 3 (Ludwig, 2003). Procedures for wholeock dissolution and spiking, elemental separations, andnalysis of the Sm and Nd isotopes work are describedn Patchett and Ruiz (1987). Over the course of thistudy, the 143Nd/144Nd value of the LaJolla Nd stan-ard was 0.511869, similar to the long-term average forhe Arizona lab. All 143Nd/144Nd values reported hereave been corrected relative to 0.511860 for the LaJollad standard. The external precision of the 143Nd/144Nd

atio, based on the reproducibility of standards andeplicate analysis, is approximately 0.000018 or 0.35psilon Nd units. Initial ε values for the samples
Ndre calculated using the ages in Table 1 and assumingresent-day CHUR values of 143Nd/144Nd = 0.512638nd 147Sm/144Nd = 0.1966 and a 147Sm decay constantalue of 6.54 × 10−12.
esearch 157 (2007) 235–268 241

4. Results

The results of preliminary petrographic, major andtrace element, and isotope studies related to this study aredescribed in a series of undergraduate theses (Phillips,2000; Kennedy, 2000; Sandland, 2001) and conferenceproceedings (Harris and Wirth, 2000; Kennedy et al.,2000; Phillips and Wirth, 2000, 2001; Vervoort et al.,2000, 2001; Sandland et al., 2001).

4.1. U–Pb zircon data

U–Pb data for 59 zircon analyses from seven gra-nophyre samples are presented in Table 1 and shown inconventional Wetherill-type U–Pb concordia diagramsin Fig. 2(a–g). Most of these analyses (39) were per-formed on single zircon grains, but due to the small sizeof zircons from some samples, a few of the analysesrequired multigrain fractions (see Table 1). All grainswere abraded in a Krogh-style air abrasion device toimprove concordancy. Analyses were conducted usingconventional isotope dilution and thermal ionizationmass spectrometry (Gehrels and Boghossian, 2000).

Analyses from all samples range from concordant tostrongly discordant and plot along well-defined lineararrays (MSWD values from 0.2 to 1.3). Because most ofthe grains are discordant, we use the upper intercept ofunforced regressions to determine crystallization ages ofthe granophyres. These ages, and their uncertainties, aresummarized in Fig. 3 and shown in their spatial contextwithin the rift in Fig. 2(h). Lower intercepts are near zerofor all samples with some small variation. Nevertheless,the regression lines fit to the data were not forced throughthe origin of concordia. The lower intercept ages for eachof the samples are shown with their upper intercept agesin Fig. 2(a–g).

In abstract, the crystallization ages for the gra-nophyres fall into two groups: an older group withages from 1109 to 1106 Ma; and a younger group withages from 1099 to 1095 Ma. These ages are consistentwith the U–Pb zircon ages published for Midcontinentrift rocks in the Lake Superior region (e.g., Davis andSutcliffe, 1985; Heaman and Machado, 1992; Davis andPaces, 1990; Paces and Miller, 1993; Davis and Green,1997; Zartman et al., 1997; Schmitz et al., 2003) andcoincide with the “early” and “main” magmatic stages(1109–1106 Ma and 1100–1094, respectively) of Mid-continent rift evolution suggested by Miller and Vervoort
(1996) and Miller and Severson (2002). This generalchronology has also been recognized by other workers(e.g., Davis and Green, 1997; Nicholson et al., 1997). Wewill follow this general two-stage chronologic division in

242J.D

.Vervoortetal./Precam

brianR

esearch157

(2007)235–268

Table 1U–Pb isotopic data and ages

Grain type Grain weight (�g) Pbc (pg) U (ppm) Isotope ratios rho Ages

206Pbm/204Pb 206Pbc/208Pb 206Pb*/238U 207Pb*/235U 207Pb*/206Pb* 206Pb*/238U 207Pb*/235U 207Pb*/206Pb*

Early Stage granophyre complexes (reversed paleomagnetic polarity)GP99-29 (Cucumber Lake granophyre); 1106.8 ± 2.8 Ma

1C 22 14 190 3063 9.9 0.165422 ± 0.54 1.73882 ± 0.62 0.07623 ± 0.30 0.875 987 1023 1101 ± 61C 21 34 232 1578 8.7 0.180579 ± 0.52 1.89922 ± 0.63 0.07626 ± 0.35 0.831 1070 1081 1102 ± 71C 14 7 64 1435 6.5 0.175690 ± 1.14 1.84681 ± 1.21 0.07623 ± 0.40 0.943 1043 1062 1101 ± 81C 15 22 231 2020 5.6 0.061969 ± 0.84 0.59097 ± 0.95 0.06891 ± 0.44 0.888 389 471 896 ± 91C 12 9 98 1450 6.4 0.180249 ± 0.94 1.89252 ± 1.05 0.07615 ± 0.45 0.903 1069 1079 1099 ± 91C 14 12 123 1750 5.3 0.178979 ± 0.69 1.88149 ± 0.85 0.07623 ± 0.50 0.807 1061 1075 1101 ± 101C 21 9 89 2380 6.6 0.186124 ± 0.61 1.96122 ± 0.75 0.07642 ± 0.40 0.845 1100 1102 1106 ± 85D 50 11 67 3535 6.4 0.185851 ± 0.50 1.95826 ± 0.58 0.07642 ± 0.30 0.855 1099 1101 1106 ± 65D 41 9 46 2557 5.8 0.187352 ± 0.64 1.97063 ± 0.71 0.07630 ± 0.30 0.906 1107 1106 1103 ± 6

GP99-57 (Misquah Hills granophyre); 1106.0 ± 4.8 Ma2F 10 23 393 1867 2.8 0.181935 ± 0.49 1.91483 ± 0.58 0.07634 ± 0.30 0.855 1078 1086 1104 ± 62F 8 5 48 762 8.4 0.172194 ± 2.26 1.81527 ± 2.32 0.07645 ± 0.60 0.966 1024 1051 1107 ± 122F 7 4 72 1473 6.8 0.176011 ± 0.96 1.85234 ± 1.07 0.07634 ± 0.45 0.907 1045 1064 1104 ± 92F 10 8 158 2037 7.2 0.162042 ± 0.72 1.70723 ± 0.77 0.07642 ± 0.25 0.946 968 1011 1106 ± 52F 12 7 148 2287 6.0 0.140344 ± 0.78 1.47520 ± 0.84 0.07623 ± 0.30 0.934 847 920 1101 ± 62F 9 8 152 1845 7.9 0.163735 ± 0.92 1.72642 ± 1.01 0.07649 ± 0.40 0.918 978 1018 1108 ± 8

GP99-46 (Whitefish Lake granophyre); 1109.4 ± 5.1 Ma1D 11 16 82 648 5.8 0.181678 ± 0.99 1.91525 ± 1.09 0.07645 ± 0.45 0.910 1076 1087 1107 ± 91D 14 8 47 1010 7.0 0.185619 ± 1.29 1.95833 ± 1.38 0.07653 ± 0.50 0.931 1098 1101 1109 ± 101D 18 14 117 1702 7.1 0.182457 ± 0.74 1.92120 ± 0.82 0.07638 ± 0.35 0.904 1081 1089 1105 ± 71D 15 8 35 760 6.0 0.182259 ± 1.71 1.92404 ± 1.78 0.07657 ± 0.50 0.959 1079 1090 1110 ± 101D 13 9 35 517 6.0 0.180382 ± 1.89 1.90177 ± 1.98 0.07645 ± 0.60 0.953 1069 1082 1107 ± 121D 16 26 65 453 4.9 0.182768 ± 0.91 1.92307 ± 1.07 0.07630 ± 0.55 0.857 1082 1089 1103 ± 115D 62 8 191 1675 6.1 0.182621 ± 0.77 1.93615 ± 0.83 0.07649 ± 0.30 0.932 1087 1094 1108 ± 64D 48 13 152 957 6.0 0.158952 ± 0.96 1.66249 ± 1.01 0.07585 ± 0.30 0.955 951 995 1091 ± 6

GP99-82 (Mt. Weber granophyre); 1106.2 ± 3.6 Ma1E 8 9 156 1495 7.8 0.187003 ± 0.80 1.97417 ± 0.94 0.07653 ± 0.45 0.877 1105 1107 1109 ± 91E 11 8 78 1330 7.9 0.183851 ± 1.12 2.13288 ± 1.19 0.08415 ± 0.41 0.938 1088 1159 1296 ± 81E 10 9 273 3065 5.2 0.157332 ± 0.60 1.65520 ± 0.67 0.07630 ± 0.30 0.894 942 992 1103 ± 61E 7 8 197 2040 6.7 0.183332 ± 0.73 1.93046 ± 0.80 0.07638 ± 0.30 0.927 1085 1092 1105 ± 61F 6 6 88 875 6.8 0.173334 ± 1.98 1.81038 ± 3.18 0.07577 ± 2.33 0.683 1031 1049 1089 ± 461F 8 13 233 1470 6.3 0.169775 ± 0.68 1.78486 ± 0.82 0.07626 ± 0.45 0.835 1011 1040 1102 ± 95F 21 5 45 1911 7.6 0.185018 ± 0.92 1.95332 ± 0.96 0.07657 ± 0.30 0.950 1094 1099 1110 ± 65F 24 12 72 1493 4.6 0.174320 ± 0.61 0.18312 ± 0.69 0.07619 ± 0.30 0.900 1036 1057 1100 ± 65F 18 13 188 2839 4.9 0.179472 ± 0.91 1.88754 ± 0.96 0.07626 ± 0.30 0.950 1064 1077 1102 ± 6

J.D.Vervoortetal./P

recambrian

Research

157(2007)

235–268243

Main Stage granophyre complexes (normal paleomagnetic polarity)GP99-08 (Pine Mtn. granophyre); 1095.3 ± 3.8 Ma

1C 18 9 47 763 4.2 0.128421 ± 1.50 1.35388 ± 1.65 0.07645 ± 0.70 0.904 779 869 1107 ± 141C 20 8 321 8145 5.1 0.167681 ± 0.46 1.76025 ± 0.52 0.07611 ± 0.25 0.877 999 1031 1098 ± 51C 17 43 139 553 3.3 0.161363 ± 0.74 1.69134 ± 1.14 0.07604 ± 0.80 0.712 964 1005 1096 ± 161C 20 6 40 1308 6.3 0.144741 ± 1.42 1.51860 ± 1.56 0.07611 ± 0.65 0.908 871 938 1098 ± 131C 19 10 78 1073 6.9 0.118635 ± 0.97 1.24262 ± 1.12 0.07596 ± 0.55 0.870 723 820 1094 ± 111C 16 5 114 2590 4.9 0.120928 ± 0.80 1.27047 ± 0.88 0.07619 ± 0.35 0.917 736 833 1100 ± 72B 52 28 414 8444 4.5 0.184109 ± 0.39 1.92749 ± 0.46 0.07592 ± 0.25 0.839 1089 1091 1093 ± 5

GP99-05 (Eagle Mtn. granophyre); 1098.6 ± 3.6 Ma1C 18 9 396 7105 5.5 0.148020 ± 1.11 1.54867 ± 1.16 0.07588 ± 0.35 0.953 890 950 1092 ± 71C 15 18 215 5470 4.9 0.169604 ± 0.52 1.78204 ± 0.64 0.07620 ± 0.35 0.834 1010 1039 1100 ± 71C 12 6 184 4395 5.9 0.179268 ± 0.58 1.87985 ± 0.70 0.07605 ± 0.40 0.821 1063 1074 1096 ± 81C 14 14 233 2540 4.4 0.172009 ± 0.56 1.80630 ± 0.66 0.07616 ± 0.30 0.891 1023 1048 1099 ± 61C 15 7 101 2130 5.8 0.145900 ± 0.92 1.52097 ± 1.03 0.07561 ± 0.45 0.899 878 939 1085 ± 91C 12 12 88 995 4.9 0.183839 ± 0.99 1.92646 ± 1.17 0.07600 ± 0.55 0.883 1088 1090 1095 ± 115E 26 9 85 2610 5.4 0.179548 ± 0.53 1.88215 ± 0.63 0.07603 ± 0.30 0.879 1065 1075 1096 ± 65E 35 9 75 3240 4.2 0.183769 ± 0.54 1.92701 ± 0.62 0.07605 ± 0.30 0.873 1088 1091 1096 ± 6

GP99-80 (Finland granophyre); 1098.2 ± 5.5 Ma1C 16 25 167 850 4.1 0.126168 ± 0.66 1.31007 ± 0.99 0.07532 ± 0.60 0.804 766 850 1077 ± 121B 18 41 168 639 3.9 0.140066 ± 0.62 1.46153 ± 1.88 0.07569 ± 0.75 0.520 845 915 1087 ± 151B 20 205 182 164 2.3 0.139291 ± 0.52 1.44351 ± 1.67 0.07517 ± 1.51 0.452 841 907 1073 ± 301B 23 21 160 1629 4.3 0.147146 ± 0.52 1.54359 ± 0.73 0.07609 ± 0.50 0.727 885 948 1097 ± 101B 17 24 219 1506 4.4 0.177166 ± 0.49 1.85731 ± 0.68 0.07604 ± 0.45 0.749 1052 1066 1096 ± 91B 19 19 204 1320 4.3 0.101956 ± 0.74 1.05353 ± 0.88 0.07494 ± 0.45 0.860 626 731 1067 ± 91B 20 105 139 260 3.1 0.156679 ± 0.55 1.64300 ± 1.15 0.07609 ± 0.96 0.563 938 987 1097 ± 191C 21 26 83 608 3.9 0.145298 ± 0.96 1.51526 ± 1.22 0.07562 ± 0.70 0.819 875 936 1085 ± 142C 41 8 38 1856 4.5 0.147736 ± 0.90 1.54435 ± 1.03 0.07581 ± 0.50 0.874 888 948 1090 ± 102C 58 12 21 962 4.6 0.150367 ± 0.90 1.56895 ± 1.04 0.07567 ± 0.50 0.877 903 958 1086 ± 102C 73 98 27 212 2.8 0.165994 ± 0.61 1.74727 ± 1.29 0.07623 ± 1.06 0.584 990 1025 1101 ± 212C 61 10 34 2034 4.8 0.162209 ± 0.58 1.69429 ± 0.70 0.07575 ± 0.40 0.820 969 1006 1089 ± 8

Notes: Grain type: A = ∼250μ, B = ∼200μ, C = ∼150μ, D = ∼125μ, E = ∼100μ, F = ∼80μ, F = ∼60μ. All grains abraded in air abrasion device. Number refers to number of grains analyzed.206Pb/204Pb is measured ratio, uncorrected for blank, spike, or fractionation. 206Pb/208Pb is corrected for blank, spike, and fractionation. All uncertainties are at the 95% confidence level.Uncertainties in isotope ratios are in percent. Uncertainties in ages are in millions of years. Most concentrations have an uncertainty of 25% due to uncertainty in weight of grain. Constantsused: 238U/235U = 137.88. Decay constant for 235U = 9.8485 × 10−10. Decay constant for 238U = 1.55125 × 10−10. Isotope ratios are adjusted as follows: (1) mass-dependent corrections factors of:0.14 ± 0.06 %/amu for Pb and 0.04 ± 0.04 %/amu for UO2. (2) Pb ratios corrected for 0.005 ± 0.003 ng blank with 206Pb/204Pb = 18.6 ± 0.3, 207Pb/204Pb = 15.5 ± 0.3, and 208Pb/204Pb = 38.0 ± 0.8.(3) U has been adjusted for 0.001 ± 0.001 ng blank. (4) Initial Pb from Stacey and Kramers (1975), with uncertainties of 1.0 for 206Pb/204Pb, 0.3 for 207Pb/204Pb, and 2.0 for 208Pb/204Pb. Allanalyses conducted using conventional isotope dilution and thermal ionization mass spectrometry, as described by Gehrels and Boghossian (2000).


Fig. 2. 206Pb/238U vs. 207Pb/235U concordia diagrams of zircon analyses from early stage ((a) Cucumber Lake, (b) Misquah Hills, (c) WhitefishLake, (d) Mt. Weber), and main stage ((e) Pine Mtn., (f) Eagle Mtn., (g) Finland) granophyre complexes. (h) Map of the MRS in northeasternMinnesota showing the U–Pb zircon concordia ages of studied granophyre complexes.

J.D. Vervoort et al. / Precambrian Research 157 (2007) 235–268 245

Fig. 2. (Conti

Fig. 3. Summary plot of U–Pb zircon ages from granophyre com-plexes. The bands labeled “early” and “main” magmatic stages arebased on the weighted mean of the reported granophyre ages. Theseages are consistent with the evolutionary stages of Midcontinent RiftSystem as proposed by Miller and Vervoort (1996).

nued ).

presenting the age and isotope data for the granophyresbelow.

4.1.1. Early Stage granophyres4.1.1.1. Cucumber Lake granophyre (GP99-29).Results from seven single grain and two multigrain (fivegrain) analyses range from concordant to moderatelydiscordant (10%) with one highly discordant (57%)grain (Fig. 2a). The two multigrain fractions yieldconcordant analyses with 207Pb/206Pb ages of 1106 ± 6and 1103 ± 3 Ma. A regression line fit through all anal-yses defines an upper intercept age of 1106.8 ± 2.8 Ma(MSWD = 1.2).

4.1.1.2. Misquah Hills granophyre (GP99-57). Anal-
yses of six fractions, each consisting of two ∼80 �mgrains, range from slightly to moderately discordant(1–11%, with one at 22%; Fig. 2(b)). A regression linefit through all six points yields an upper intercept age

brian R
of 1106.0 ± 4.8 Ma (MSWD = 0.43). This is consistentwith the 207Pb/206Pb ages of the five least discordantpoints, which range from 1108 to 1104 Ma (average1105.8 Ma).

4.1.1.3. Whitefish Lake granophyre (GP99-46). Analy-ses of six single grain and two multigrain (4 and 5 grain)fractions yield data that are concordant, or nearly so, withone moderately discordant (11%) analysis (Fig. 2(c)). Aregression line fit through seven points, excluding oneslightly discordant point (a five grain fraction set off fromthe main population), yields an upper intercept age of1109.4 ± 5.1 Ma (MSWD = 0.25). Including all points inthe regression yields an age (1112.0 ± 6.9 Ma) that is sta-tistically identical to the seven point regression but withhigher error and a higher MSWD (2.0) and low proba-bility of fit (0.061). The excluded point may representan inherited older component in this fraction similar towhat is seen in one of the grains in the Mt. Weber sample(see below).

4.1.1.4. Mount Weber granophyre (GP99-82). Six sin-gle grain and three multigrain (5 grain) fractionswere analyzed from this sample. Eight of these anal-yses yielded compositions that were concordant tomoderately discordant (up to 15%) and plot along a well-defined regression line (MSWD = 1.19) with an upperintercept age of 1106.2 ± 3.6 Ma (Fig. 2(d)). One of thesingle grain analyses from this sample had a consider-ably older age (e.g., 207Pb/206Pb age of 1296 ± 8 Ma) andwas not included in the regression because of this oldercomponent. Although inheritance is rare in the MRS rhy-olites and granophyres (Davis and Green, 1997), thereis firm isotopic evidence for the incorporation of oldercrust in their genesis (Vervoort and Green, 1997), so itseems likely that this grain has a xenocrystic componentincorporated from older Proterozoic or Archean crust inthe basement to the MRS rift magmas.

4.1.2. Main Stage granophyres4.1.2.1. Pine Mountain granophyre (GP99-08).Results from one multigrain (two grain) and six singlegrain analyses range from concordant to highly (34%)discordant (Fig. 2(e)). A regression line fit through allpoints yields an upper intercept age of 1095.3 ± 3.8 Ma(MSWD = 0.83). This age is consistent with the singleconcordant analysis from this sample with a 207Pb/206Pbage of 1093 ± 5 Ma.

4.1.2.2. Eagle Mountain granophyre (GP99-05). Anal-yses of six single grain and two multigrain (fivegrain) fractions range from concordant to moderately

esearch 157 (2007) 235–268

(17%) discordant (Fig. 2(f)). A line fit through all thedata yields an upper intercept age of 1098.6 ± 3.6 Ma(MSWD = 0.95). The four most concordant analysesfrom this sample have 207Pb/206Pb ages that range from1095 ± 11 to 1096 ± 8 Ma (Table 1) consistent with theupper intercept age.

4.1.2.3. Finland granophyre (GP99-80). Analyses ofeight single grain and four multigrain (five grain)fractions range from slightly to highly (41%) discor-dant (Fig. 2(g)). A regression line fit through all datapoints yields an upper intercept age of 1098.2 ± 5.5 Ma(MSWD = 0.7).

4.1.3. Age populationsA statistical test was performed on the groups of gra-

nophyre ages to determine whether the Early Stage andMain Stage granophyres represent two distinct age pop-ulations. Using the ages and errors reported in Table 1and Fig. 2, a test of the null-hypothesis that the two sam-ple means are the same yields a two-tailed p-value of0.0024 meaning that there is a low probability that thegranophyre ages represent a single age population.

4.2. Major and trace-element data

Major and trace-element compositions of representa-tive samples of the granophyre complexes, transitionalborder zones, and selected adjacent country rocks arepresented in Tables 2 and 3; only those samples fromthe complexes and the transitional zones are plotted ingeochemical diagrams. Perhaps the most notable featureof the whole rock and trace-element geochemistry is thesimilar composition of samples from the early and mainstage granophyre complexes. Samples of granophyrecomplexes are classified as monzodiorite, quartz mon-zodiorite, quartz monzonite, and granite according totheir CIPW normative mineral compositions (Fig. 4(a)),but the majority of rocks are granite to quartz mon-zonite in composition. Less abundant monzodiorite andquartz monzodiorite occur predominately in the lowerportions of the complexes. The granitic rocks are metalu-minous to weakly peraluminous based on their aluminasaturation indices and plot along continuous trends onvariation diagrams (Figs. 4(c and d)). In particular, themajor elements TiO2, Fe2O3

*, MgO, CaO decrease, andNa2O and K2O increase, with increasing SiO2. In addi-tion, compatible trace elements (e.g., Sc, Cr, Ni, Co, V,
Zr) decrease and the incompatible elements (REE, Rb,Ba, Y, Zr, Nb, Hf, Pb, Th, U) increase with increasingsilica. In spite of the compositional variations exhib-ited by the granophyres, the major and trace-element


Table 2Major element compositions of Midcontinent rift granophyres and related rocks

Sample Rock type SiO2 TiO2 Al2O3 Fe2O3a MnO MgO CaO Na2O K2O P2O5 LOI Total

Early Stage granophyre complexes (reversed paleomagnetic polarity)Cucumber Lake granophyre (1107 Ma)

GP-99-11 Granite 68.90 0.59 12.70 6.66 0.12 0.45 1.79 3.53 5.06 0.10 0.21 100.11GP-99-12 Granite 71.27 0.44 12.76 4.88 0.09 0.24 1.26 3.47 5.31 0.05 0.31 100.08GP-99-26A Granite 71.16 0.45 12.47 5.08 0.08 0.26 1.22 3.62 5.23 0.06 0.07 99.70GP-99-27 Granite 71.43 0.44 12.48 4.84 0.08 0.15 1.35 3.84 5.08 0.05 0.12 99.86GP-99-28 Granite 71.23 0.50 12.83 5.21 0.09 0.46 1.26 3.70 5.14 0.07 0.25 100.74GP-99-29 Granite 70.70 0.50 12.75 5.14 0.08 0.56 1.17 3.35 5.52 0.07 0.20 100.04GP-99-30 Granite 70.49 0.52 12.82 5.22 0.08 0.60 1.09 3.42 5.46 0.07 0.21 99.98GP-99-31 Granite 68.47 0.63 13.16 6.50 0.12 0.46 1.82 3.87 4.94 0.10 0.18 100.25GP-99-32 Quartz Monzonite 56.58 2.03 14.53 12.88 0.18 2.14 4.71 4.05 2.40 0.53 0.33 100.36GP-99-33 Quartz Diorite 52.35 2.12 12.36 18.76 0.29 1.25 6.59 3.46 1.87 0.65 0.00 99.70GP-99-34 Monzodiorite 51.91 1.65 21.07 8.01 0.10 2.21 10.15 3.75 0.92 0.31 0.09 100.17GP-99-35 Quartz Monzonite 53.41 2.55 12.13 15.66 0.21 2.30 6.18 3.27 2.59 1.11 0.12 99.53GP-99-36A Quartz Monzonite 66.27 0.79 13.34 7.72 0.13 0.66 2.11 3.84 4.81 0.13 0.15 99.95GP-99-36B Quartz Diorite 53.21 2.85 12.67 15.12 0.20 3.02 6.48 3.65 1.44 1.20 0.19 100.03GP-99-37A Monzonite 53.27 2.79 13.13 13.81 0.18 3.08 6.73 4.02 2.04 0.87 0.05 99.97GP-99-37B Quartz Monzonite 58.05 1.71 12.19 16.88 0.30 3.05 5.47 3.65 2.56 0.20 0.00 104.06GP-99-38 Granite 71.27 0.48 12.95 4.52 0.06 0.46 0.97 2.87 5.76 0.09 0.16 99.59GP-99-39 Granite 71.37 0.49 13.08 4.61 0.07 0.50 1.01 2.94 5.84 0.09 0.17 100.17GP-99-40 Granite 71.62 0.47 12.76 4.92 0.09 0.17 1.02 3.70 4.96 0.05 0.19 99.95GP-00-133B Monzodiorite 51.77 2.96 12.75 15.20 0.19 3.64 7.61 3.27 1.56 0.87 0.01 99.83GP-00-133C Monzodiorite 51.96 2.35 16.27 12.21 0.16 2.30 8.31 3.37 1.53 0.81 0.12 99.39GP-00-134 Monzodiorite 47.07 3.46 13.98 16.65 0.19 4.90 8.08 2.60 1.20 0.45 0.83 99.41

Hovland lavas (∼1108 Ma; upper contact of Cucumber Lake granophyre)GP-99-24 Rhyolite 74.27 0.32 11.53 4.31 0.03 0.12 0.33 3.11 5.02 0.06 0.08 99.18

Misquah Hills (1106 Ma)GP-99-57 Granite 70.11 0.52 12.85 5.62 0.09 0.74 1.19 3.89 4.98 0.10 0.12 100.21GP-99-58 Granite 70.36 0.52 12.81 5.66 0.08 0.69 1.38 4.35 4.46 0.10 0.11 100.52GP-99-59 Quartz Monzonite 56.76 2.39 12.30 13.63 0.20 2.23 4.84 3.56 2.94 0.89 0.16 99.90GP-99-60 Granite 71.08 0.61 12.73 5.77 0.10 0.48 1.59 3.92 4.24 0.13 0.00 100.65GP-99-61 Granite 70.65 0.62 12.77 5.86 0.09 0.55 1.48 3.51 4.83 0.14 0.10 100.60GP-99-62 Monzodiorite 47.32 3.47 14.78 15.98 0.19 5.26 8.14 2.83 1.23 0.48 0.14 99.82GP-99-63 Monzodiorite 50.93 2.56 16.43 12.92 0.16 3.02 8.86 3.46 1.37 0.39 0.09 100.19GP-99-64 Monzodiorite 54.34 1.49 19.25 8.85 0.12 2.06 8.76 3.77 1.25 0.41 0.10 100.40

Undifferentiated lava flows (age uncertain; upper contact of Misquah Hills granophyre)GP-99-53 Rhyolite 74.54 0.45 11.77 4.01 0.03 0.33 0.93 2.15 5.31 0.06 0.09 99.67GP-99-54A Rhyolite 73.29 0.49 12.65 4.25 0.05 0.42 0.92 3.33 4.78 0.09 0.14 100.41GP-99-55 Rhyolite 74.01 0.47 12.52 4.34 0.04 0.33 0.43 3.12 5.46 0.08 0.13 100.93GP-99-56 Rhyolite 75.55 0.35 11.79 3.73 0.03 0.27 0.61 3.48 4.85 0.06 0.06 100.78

Mt. Weber granophyre (1106 Ma)GP-99-14A Granite 75.62 0.37 12.29 3.74 0.06 0.14 0.83 3.85 4.85 0.04 0.00 101.79

Whitefish Lake granophyre (1109 Ma)GP-99-46 Granite 75.98 0.26 12.01 2.81 0.04 0.18 0.34 3.05 5.28 0.03 0.11 100.09

Main Stage granophyre complexes (normal paleomagnetic polarity)Eagle Mountain granophyre (1099 Ma)

GP-99-01 Granite 72.74 0.51 12.27 4.70 0.05 0.80 0.46 2.77 5.94 0.08 0.20 100.52GP-99-02 Quartz Monzonite 56.97 1.94 12.49 14.08 0.22 1.69 4.96 4.04 2.30 0.75 0.14 99.58GP-99-03 Granite 72.61 0.45 12.55 4.10 0.06 0.41 1.13 3.65 4.82 0.06 0.17 100.01GP-99-05 Granite 72.15 0.52 12.63 4.68 0.08 0.36 1.59 4.00 4.27 0.08 0.05 100.41GP-99-65 Quartz Monzonite 61.59 1.49 12.79 11.77 0.18 1.26 3.66 3.76 3.30 0.40 0.14 100.34GP-99-66 Granite 69.17 0.79 12.65 6.78 0.12 0.71 1.53 3.65 4.51 0.16 0.13 100.20GP-99-67 Granite 68.95 0.82 12.65 6.89 0.11 0.64 2.16 3.81 4.09 0.16 0.09 100.37GP-99-68 Quartz Monzonite 54.39 2.67 11.86 16.56 0.23 2.30 5.80 3.11 2.14 0.75 0.09 99.90GP-99-69 Quartz Monzonite 54.63 2.32 11.38 17.73 0.24 2.26 5.36 3.01 2.12 0.70 0.18 99.93GP-99-70A Granite 68.62 0.80 12.77 6.35 0.10 1.06 2.65 3.33 4.24 0.16 0.05 100.13GP-99-70B Monzodiorite 53.76 0.75 21.42 7.25 0.10 3.00 9.17 3.36 1.27 0.19 0.20 100.47GP-99-70C Monzodiorite 54.14 0.78 21.56 6.64 0.11 2.62 8.95 3.29 1.76 0.18 0.24 100.27GP-99-70D Granite 69.31 0.74 12.80 6.16 0.10 1.06 2.45 3.32 4.24 0.16 0.06 100.40


Table 2 (Continued )

Sample Rock type SiO2 TiO2 Al2O3 Fe2O3a MnO MgO CaO Na2O K2O P2O5 LOI Total

GP-99-71 Granite 73.70 0.33 12.23 4.24 0.06 0.51 0.65 3.62 4.60 0.04 0.19 100.17GP-99-72 Quartz Monzonite 62.94 1.08 12.08 11.41 0.20 0.86 3.12 3.50 3.57 0.25 1.39 100.40GP-99-75 Granite 71.21 0.59 12.33 5.30 0.09 0.41 1.79 3.76 4.31 0.11 0.45 100.35GP-99-78 Granite 70.04 0.59 12.34 5.53 0.07 0.67 1.13 3.46 4.30 0.13 1.64 99.90

Undifferentiated lava flows (age uncertain; upper contact of Eagle Mtn. granophyre)GP-99-77B Rhyolite 65.85 1.09 12.68 8.15 0.10 1.29 1.45 3.99 4.41 0.28 1.20 100.49

Pine Mountain granophyre (1095 Ma)GP-00-115 Granite 67.00 1.02 11.95 7.29 0.09 1.62 1.28 3.41 3.93 0.15 2.38 100.12GP-00-117 Granite 65.87 0.83 12.27 8.13 0.15 1.14 1.81 3.38 3.80 0.18 2.50 100.06GP-00-118 Quartz Monzonite 56.96 1.93 11.79 14.60 0.21 1.26 5.07 3.07 2.56 0.64 1.80 99.89GP-99-06 Quartz Monzonite 65.95 1.16 12.81 8.73 0.13 1.30 2.06 4.00 3.75 0.28 0.26 100.43GP-99-07 Granite 68.59 0.82 12.56 6.61 0.10 1.07 1.38 3.44 4.95 0.17 0.29 99.98GP-99-08 Granite 69.01 0.84 12.55 6.83 0.11 1.01 1.31 3.37 5.13 0.18 0.25 100.59GP-99-09 Granite 73.26 0.49 12.27 4.19 0.05 0.82 0.51 3.07 5.71 0.09 0.17 100.63GP-99-10 Granite 73.03 0.50 12.36 3.98 0.05 0.67 0.66 3.47 5.23 0.08 0.25 100.28GP-99-41 Granite 71.04 0.52 12.20 4.57 0.06 0.62 0.55 3.11 4.84 0.10 2.05 99.66GP-99-43 Quartz Monzonite 57.57 2.13 12.22 14.75 0.22 2.02 4.33 3.78 2.62 0.54 0.28 100.46GP-99-44 Granite 67.52 0.92 12.87 7.67 0.10 1.78 1.30 3.19 4.57 0.19 0.27 100.38GP-99-45 Granite 70.37 0.63 12.84 5.74 0.09 1.14 0.97 3.24 5.03 0.11 0.32 100.48GP-99-52 Granite 66.63 0.95 12.08 7.05 0.08 1.63 0.95 3.03 5.14 0.16 1.84 99.54

Brule Lake Hovland Gabbro-Diabase (age uncertain in; basal contact of Pine Mountain granophre)GP-99-42 Monzodiorite 46.49 4.48 11.91 20.01 0.22 4.07 8.52 2.54 1.03 0.29 0.26 99.82GP-00-112 Gabbro 45.39 3.03 15.88 18.03 0.17 5.01 7.96 2.59 0.71 0.17 0.62 99.56GP-00-123 Gabbro 51.52 3.27 12.04 15.97 0.22 3.52 8.17 2.76 1.38 0.34 0.96 100.15GP-00-127 Gabbro 53.81 3.69 11.75 16.33 0.22 1.88 6.10 3.00 2.10 0.55 1.47 100.90

Finland granophyre (1098 Ma)GP-99-81 Granite 76.09 0.26 12.28 2.68 0.04 0.00 0.40 3.94 4.76 0.03 0.10 100.58GP-AC1-016 Granite 75.05 0.37 12.94 3.78 0.03 0.24 0.35 3.04 6.20 0.04 0.25 102.29GP-AC1-029 Granite 71.69 0.43 12.45 4.87 0.09 0.97 1.30 3.32 5.28 0.05 0.33 100.78GP-AC1-041 Granite 72.06 0.44 12.39 4.68 0.06 0.47 1.41 3.79 4.52 0.05 0.29 100.16GP-AC1-056 Granite 70.35 0.51 12.33 5.96 0.06 1.02 0.96 3.17 5.47 0.06 0.27 100.16GP-AC 1-079 Granite 69.93 0.56 12.50 6.77 0.07 0.94 0.88 3.60 4.69 0.07 0.27 100.28GP-AC1-100 Granite 68.01 0.73 12.79 8.04 0.12 1.34 1.04 3.74 4.32 0.10 0.57 100.80GP-AC1-120 Granite 67.43 0.71 12.00 8.36 0.14 0.90 1.65 3.47 4.43 0.10 1.33 100.52GP-AC1-134 Quartz Monzonite 62.68 1.06 11.94 12.76 0.22 0.98 3.01 3.96 3.16 0.19 0.29 100.25GP-AC1-156 Quartz Monzonite 62.35 1.06 11.61 13.59 0.25 0.55 3.55 3.61 3.07 0.21 0.22 100.07GP-AC1-178 Granite 67.12 0.71 12.21 9.25 0.16 0.57 2.43 4.04 3.81 0.10 0.17 100.57GP-AC1-197 Quartz Monzonite 62.17 1.23 11.96 13.09 0.25 0.64 3.80 3.69 2.99 0.27 0.17 100.26GP-AC 1-222 Quartz Monzodiorite 59.57 1.08 11.54 16.21 0.26 1.10 3.76 3.28 2.86 0.31 0.24 100.21GP-AC 1-225 Granite 74.62 0.29 12.03 3.53 0.04 0.29 0.60 3.46 5.29 0.03 0.17 100.35GP-AC1-253 Granite 72.96 0.34 12.12 4.16 0.05 0.58 1.03 3.45 5.23 0.04 0.25 100.21GP-AC 1-329 Granite 62.48 0.38 10.77 4.56 0.05 0.62 0.66 3.16 4.38 0.04 0.29 87.39GP-AC1-363 Granite 72.23 0.40 12.61 4.67 0.06 0.62 0.98 3.80 4.77 0.05 0.26 100.45GP-AC1-396 Granite 68.22 0.64 12.05 7.52 0.12 0.88 1.34 3.75 4.32 0.08 0.26 99.18

Standardsa

W-2 Average 52.41 1.06 15.43 10.91 0.17 6.43 10.93 2.26 0.60 0.13 100.30%Standard deviation 0.09 0.35 0.17 0.10 0.64 0.22 0.22 0.85 0.77 1.39Published values 52.73 1.07 15.43 10.80 0.16 6.41 10.93 2.15 0.63 0.13 100.44%Diff. publ. values −0.61 −0.59 −0.06 1.04 2.79 0.33 −0.03 4.91 −4.43 −4.53

AC-E Average 70.30 0.10 14.85 2.53 0.056 bd 0.34 6.65 4.53 0.009 99.36%Standard deviation 0.16 2.08 0.30 0.00 1.34 1.53 0.91 0.41 9.04Published values 70.56 0.11 14.74 2.54 0.06 0.03 0.34 6.56 4.50 0.01 99.45%Diff. publ. values −0.37 −6.79 0.70 −0.29 −3.44 −1.27 1.44 0.52 −10.13

LDMb 0.4 0.02 0.1 0.1 0.002 0.1 0.1 0.04 0.02 0.005

a Published Values from Govindaraju (1994) recalculated as anhydrous.b LDM: Limit of detection of method for XRF data = the minimum concentration that can be determined at the 95.4% level of confidence; calculated

as recommended by Rousseau (2001).


Table 3Trace-element compositions of Midcontinent rift granophyres and related rocks

Sample Rock type Co Cr Ni Sc V Zn Ga Ba Rb Sr

Early granophyre complexes (reversed paleomagnetic polarity)Cucumber Lake granophyre (1107 Ma)

GP-99-11 Granite 18 14 9 5 12 180 28 1090 188 135GP-99-12 Granite 13 16 8 1 bd 153 28 1127 204 108GP-99-26A Granite 13 14 8 4 7 104 27 982 212 106GP-99-27 Granite 13 16 8 3 5 166 28 945 196 90GP-99-28 Granite 14 16 8 3 4 155 27 980 186 106GP-99-29 Granite 14 16 8 3 7 161 27 988 187 130GP-99-30 Granite 14 14 8 3 5 148 26 1068 198 102GP-99-31 Granite 17 14 8 5 8 171 27 1329 171 143GP-99-32 Quartz Monzonite 37 5 6 22 33 154 25 595 85 286GP-99-33 Quartz Diorite 49 4 5 61 11 196 26 536 63 302GP-99-34 Monzodiorite 26 89 36 16 161 75 26 303 27 467GP-99-35 Quartz Monzonite 42 1 10 27 76 201 27 700 93 282GP-99-36A Quartz Monzonite 20 13 8 8 11 191 27 1469 156 170GP-99-36B Quartz Diorite 41 4 14 27 129 196 27 625 37 293GP-99-37A Monzonite 38 8 5 27 209 273 30 799 96 434GP-99-37B Quartz Monzonite 38 11 26 13 10 160 26 597 72 291GP-99-38 Granite 13 15 9 2 19 108 25 947 225 91GP-99-39 Granite 14 15 9 2 15 112 24 929 207 93GP-99-40 Granite 13 18 8 2 bd 168 28 962 187 92GP-00-133B Monzodiorite bd 18 bd 13 220 175 24 446 63 284GP-00-133C Monzodiorite bd bd bd 6 134 159 26 430 49 369GP-00-134 Monzodiorite bd 56 bd 12 264 161 25 328 39 246

Hovland lavas (∼1108 Ma; upper contact of Cucumber Lake granophyre)GP-99-24 Rhyolite 12 14 7 bd 2 54 20 865 126 79

Misquah Hills (1106 Ma)GP-99-57 Granite 15 14 8 5 5 125 26 962 172 108GP-99-58 Granite 15 17 8 5 5 143 27 828 141 80GP-99-59 Quartz Monzonite 38 bd 7 22 49 180 26 749 93 234GP-99-60 Granite 15 17 9 10 11 133 25 888 147 162GP-99-61 Granite 16 17 10 10 10 131 24 831 147 165GP-99-62 Monzodiorite 47 93 110 25 231 156 25 365 56 231GP-99-63 Monzodiorite 38 8 30 25 253 137 25 358 48 364GP-99-64 Monzodiorite 26 26 18 17 96 113 26 353 29 354

Undifferentiated lava flows (age uncertain; upper contact of Misquah Hills granophyre)GP-99-53 Rhyolite 11 12 8 6 14 52 22 864 178 112GP-99-54A Rhyolite 12 15 10 7 10 88 24 685 132 90GP-99-55 Rhyolite 13 15 11 6 10 116 21 932 175 137GP-99-56 Rhyolite 11 15 8 3 bd 106 20 884 132 123

Mt. Weber granophyre (1106 Ma)GP-99-14A Granite 11 19 8 2 bd 108 22 1225 102 78

Whitefish Lake granophyre (1109 Ma)GP-99-46 Granite 9 12 7 1 bd 100 22 874 150 27

Late granophyre complexes (normal paleomagnetic polarity)Eagle Mountain granophyre (1099 Ma)

GP-99-01 Granite 14 13 8 4 5 75 21 990 163 68GP-99-02 Quartz Monzonite 41 4 10 23 35 190 26 705 70 259GP-99-03 Granite 13 21 9 5 8 91 23 926 161 95GP-99-05 Granite 14 13 10 4 14 96 22 990 160 133GP-99-65 Quartz Monzonite 32 5 7 16 14 151 26 791 87 202GP-99-66 Granite 19 7 8 10 12 130 24 895 126 123GP-99-67 Granite 18 9 8 10 8 157 24 855 126 126GP-99-68 Quartz Monzonite 49 bd 7 25 42 206 26 624 79 221GP-99-69 Quartz Monzonite 54 bd 6 25 20 226 25 655 83 192




GP-99-70A Granite 18 10 10 9 36 105 23 815 134 127GP-99-70B Monzodiorite 29 52 70 10 33 90 22 268 40 303GP-99-70C Monzodiorite 25 46 42 11 43 86 22 458 55 344GP-99-70D Granite 19 10 10 9 34 117 23 861 130 124GP-99-71 Granite 12 17 8 bd bd 123 27 1030 119 63GP-99-72 Quartz Monzonite bd 8 bd 18 18 212 28 932 102 176GP-99-75 Granite bd 15 bd 8 12 131 23 869 152 130GP-99-78 Granite bd 6 bd 0 15 123 23 919 146 104

Undifferentiated lava flows (age uncertain; upper contact of Eagle Mtn. granophyre)GP-99-77B Rhyolite bd 9 bd 13 16 130 23 904 132 74

Pine Mountain granophyre (1095 Ma)GP-00-115 Granite bd bd bd bd 93 143 22 822 115 102GP-00-117 Granite bd bd bd bd 12 187 25 906 125 106GP-00-118 Quartz Monzonite bd bd bd 9 16 220 27 786 85 218GP-99-06 Quartz Monzonite 26 9 7 15 7 141 24 916 115 134GP-99-07 Granite 19 12 7 10 15 114 23 682 133 106GP-99-08 Granite 19 9 8 10 11 129 24 850 136 98GP-99-09 Granite 13 12 9 3 3 66 22 1005 167 52GP-99-10 Granite 12 18 8 6 7 69 21 1095 147 59GP-99-41 Granite bd 9 bd 6 28 97 22 1006 167 87GP-99-43 Quartz Monzonite 42 7 7 23 25 196 27 694 79 158GP-99-44 Granite 23 14 9 10 14 139 25 1041 137 185GP-99-45 Granite 17 20 9 6 12 112 24 1297 179 77GP-99-52 Granite bd 11 bd 10 54 114 23 848 141 97

Brule Lake Hovland Gabbro-Diabase (age uncertain; basal contact of Pine Mountain granophre)GP-99-42 Monzodiorite 35 bd 31 43 638 210 27 311 30 221GP-00-112 Gabbro bd 20 bd 4 594 136 27 227 18 285GP-00-123 Gabbro bd bd bd 24 113 176 25 430 47 241GP-00-127 Gabbro bd bd bd 11 49 196 28 599 73 212

Finland granophyre (1098 Ma)GP-99-81 Granite 9 16 7 bd bd 96 28 839 155 35GP-AC1-016 Granite 11 19 8 bd bd 30 29 1014 120 48GP-AC1-029 Granite 14 17 7 bd bd 40 27 936 100 52GP-AC1-041 Granite 13 19 7 bd bd 62 29 865 90 67GP-AC1-056 Granite 15 14 7 bd bd 49 28 1034 106 64GP-AC1-079 Granite 17 11 7 bd 4 51 29 1047 106 81GP-AC1-100 Granite 21 12 6 3 5 186 26 1067 117 166GP-AC1-120 Granite 22 12 6 2 6 229 30 988 116 139GP-AC1-134 Quartz Monzonite 33 9 5 12 8 254 30 864 87 207GP-AC1-156 Quartz Monzonite 37 13 5 12 10 311 31 743 85 196GP-AC1-178 Granite 23 16 6 8 4 190 33 831 93 158GP-AC1-197 Quartz Monzonite 35 19 5 18 6 211 31 798 74 219GP-AC1-222 Quartz Monzodiorite 47 10 5 17 9 269 31 730 86 203GP-AC1-225 Granite 10 19 7 bd bd 301 29 849 115 45GP-AC1-253 Granite 11 24 7 bd bd 34 29 987 123 53GP-AC1-329 Granite 12 7 7 bd bd 151 31 999 123 76GP-AC1-363 Granite 12 12 7 bd 2 63 31 1052 114 71GP-AC1-396 Granite 19 12 7 3 5 195 31 933 108 112

Standardsa

W-2 Average 42 94 66 37 267 78 19 186 21 198%Standard deviation 2.42 0.47 1.22 0.97 0.73 0.47 2.60 0.99 0.76 0.19Published valuesa 44 93 70 35 262 77 20 182 20 194%Diff. publ. valuesa −5.38 0.90 −5.69 6.19 1.98 0.89 −6.17 2.14 6.08 1.99

AC-E Average bd bd bd bd bd 223 41 57 151 bd%Standard deviation 0.26 0.98 0.85 0.27Published valuesa 0.2 3.4 1.5 0.1 3.0 224 39 55 152 3.0%Diff. publ. values −0.33 4.49 4.33 −0.34




LDMb 3.4 3.5 1.7 2.4 6.5 1.4 1.3 6.1 0.6 3.7

Sample Rock type Y Zr Nb La Ce Pr Nd Sm Eu Gd


GP-99-11 Granite 115 1121 99 125.7 270 34.5 133.5 26.1 5.01 24.6GP-99-12 Granite 111 1006 89 284GP-99-26A Granite 101 1006 85 117.8 249 31.2 119.1 22.9 3.66 21.5GP-99-27 Granite 105 960 91 121.6 256.9 32.2 122.7 23.8 3.91 22.3GP-99-28 Granite 107 963 88 117.2 249.5 31.4 121.2 23.6 4.07 22.2GP-99-29 Granite 107 942 91 110.6 242.5 31.3 122.1 24.4 4.24 22.8GP-99-30 Granite 110 933 89 270GP-99-31 Granite 101 1000 80 260GP-99-32 Quartz Monzonite 50 409 45 59.5 129.9 17.2 70.3 13.8 4.00 13.3GP-99-33 Quartz Diorite 66 378 38 60.1 135.6 18.5 77.9 15.8 5.06 15.3GP-99-34 Monzodiorite 21 149 15 24.0 53.0 7.1 29.9 6.19 2.41 6.12GP-99-35 Quartz Monzonite 82 581 59 185GP-99-36A Quartz Monzonite 92 1194 76 220GP-99-36B Quartz Diorite 83 528 54 78.5 174.3 23.3 97.3 19.7 4.93 19.3GP-99-37A Monzonite 97 627 80 68.7 149.0 19.2 78.0 15.7 3.71 15.0GP-99-37B Quartz Monzonite 58 450 37 146GP-99-38 Granite 74 831 80 72.5 168.6 19.7 75.0 14.6 2.37 13.8GP-99-39 Granite 81 829 79 245GP-99-40 Granite 110 1024 90 131.4 270.4 34.5 131.6 25.2 4.16 23.9GP-00-133B Monzodiorite 65 416 54 61.5 138 18.6 78.1 16.1 4.00 15.8GP-00-133C Monzodiorite 56 338 42 52.9 117 15.7 66.6 13.7 3.82 13.0GP-00-134 Monzodiorite 51 285 28 30.9 70.3 9.7 42.4 9.65 2.78 10.4

Hovland lavas (∼1108 Ma; upper contact of Cucumber Lake granophyre)GP-99-24 Rhyolite 36 569 48 92.2 172 23.2 85.8 14.2 2.60 11.5

Misquah Hills (1106 Ma)GP-99-57 Granite 101 1012 90 111.7 239 30.6 117.2 22.8 4.24 21.2GP-99-58 Granite 100 936 87 118.2 247 31.3 122.5 23.6 4.35 22.2GP-99-59 Quartz Monzonite 73 505 46 162GP-99-60 Granite 71 695 59 215GP-99-61 Granite 70 694 58 85.5 184 23.7 93.1 18.0 3.48 16.4GP-99-62 Monzodiorite 42 284 26 69.0GP-99-63 Monzodiorite 34 190 13 28.2 61.9 8.3 35.3 7.4 2.35 7.43GP-99-64 Monzodiorite 36 221 20 32.2 71.0 9.5 39.7 8.3 2.56 8.16

Undifferentiated lava flows (age uncertain; upper contact of Misquah Hills granophyre)GP-99-53 Rhyolite 71 863 57 172GP-99-54A Rhyolite 44 521 43 57.9 137 16.3 63.2 12.4 2.05 11.2GP-99-55 Rhyolite 47 522 41 64.0 134 17.5 68.1 13.3 2.21 12.0GP-99-56 Rhyolite 37 568 49 83.2 174 21.3 79.9 13.6 2.49 11.2

Mt. Weber granophyre (1106 Ma)GP-99-14A Granite 72 577 35 125

Whitefish Lake granophyre (1109 Ma)GP-99-46 Granite 76 584 48 84.6 174 22.4 86.3 16.9 2.26 16.1


GP-99-01 Granite 81 676 48 123 192GP-99-02 Quartz Monzonite 75 488 28 65.5 141 18.7 77.7 15.9 4.12 15.6GP-99-03 Granite 77 612 42 196GP-99-05 Granite 52 493 31 72.6 151 18.6 70.3 12.9 2.20 11.6GP-99-65 Quartz Monzonite 70 489 37 151GP-99-66 Granite 75 622 43 78.7 166 21.1 82.5 16.5 3.13 15.9




GP-99-67 Granite 78 587 42 78.1 163 20.9 81.8 16.3 3.09 15.7GP-99-68 Quartz Monzonite 60 435 29 126GP-99-69 Quartz Monzonite 66 489 30 139GP-99-70A Granite 69 541 38 157GP-99-70B Monzodiorite 29 216 18 27.9 59.3 7.7 31.3 6.42 1.61 6.29GP-99-70C Monzodiorite 25 217 17 55.0GP-99-70D Granite 71 493 38 160GP-99-71 Granite 114 911 62 110.3 222 28.5 109.9 21.9 3.62 21.9GP-99-72 Quartz Monzonite 95 731 50 76.4 163 21.7 88.8 18.7 4.51 18.6GP-99-75 Granite 84 581 47 79.9 169 21.3 82.1 16.4 2.73 15.7GP-99-78 Granite 70 570 39 78.4 160 20.4 78.8 15.2 2.75 14.2

Undifferentiated lava flows (age uncertain; upper contact of Eagle Mtn. granophyre)GP-99-77B Rhyolite 87 588 45 73.4 141 20.1 80.3 16.3 3.33 16.3

Pine Mountain granophyre (1095 Ma)GP-00-115 Granite 66 507 34 72.6 143 19.2 74.1 14.2 2.48 13.5GP-00-117 Granite 94 734 51 85.8 183 23.7 94.9 19.5 3.83 19.0GP-00-118 Quartz Monzonite 87 513 41 160GP-99-06 Quartz Monzonite 68 497 37 164GP-99-07 Granite 71 535 35 138GP-99-08 Granite 72 568 41 77.4 164 20.8 80.9 16.0 3.03 15.5GP-99-09 Granite 63 550 36 71.4 155 18.9 72.8 14.0 2.37 13.2GP-99-10 Granite 65 549 37 75.8 159 19.8 76.0 14.5 2.47 13.6GP-99-41 Granite 65 587 37 49.5 125.5 13.9 54.7 11.6 2.09 11.4GP-99-43 Quartz Monzonite 71 504 32 63.1 138 18.4 76.5 16.1 3.78 16.5GP-99-44 Granite 70 576 42 163GP-99-45 Granite 78 607 42 81.6 167 21.0 80.5 15.4 2.69 14.5GP-99-52 Granite 75 640 46 75.7 155 20.7 81.2 16.0 2.93 15.0

Brule Lake Hovland Gabbro-Diabase (age uncertain; basal contact of Pine Mountain granophre)GP-99-42 Monzodiorite 42 237 22 28.0 62.2 8.4 35.9 8.00 2.29 8.52GP-00-112 Gabbro 26 135 13 16.9 37.0 5.0 21.6 4.80 1.59 5.14GP-00-123 Gabbro 55 307 29 86GP-00-127 Gabbro 81 456 37 55.2 121 16.2 67.7 14.8 3.77 15.1

Finland granophyre (1098 Ma)GP-99-81 Granite 121 725 57 99.7 207 25.8 100.1 20.7 2.77 21.3GP-AC1-016 Granite 124 915 55 218GP-AC1-029 Granite 114 1012 59 252GP-AC1-041 Granite 107 1163 60 189GP-AC1-056 Granite 105 1156 62 200GP-AC1-079 Granite 96 1228 63 191GP-AC1-100 Granite 63 767 47 145GP-AC1-120 Granite 116 1285 63 201GP-AC1-134 Quartz Monzonite 106 731 51 169GP-AC1-156 Quartz Monzonite 108 793 50 170GP-AC1-178 Granite 104 637 49 189GP-AC1-197 Quartz Monzonite 97 765 44 162GP-AC1-222 Quartz Monzodiorite 101 326 43 163GP-AC1-225 Granite 129 901 60 229GP-AC1-253 Granite 125 1028 59 249GP-AC1-329 Granite 105 935 54 193GP-AC1-363 Granite 126 1074 57 233GP-AC1-396 Granite 128 1710 57 219

Standardsa

W-2 Average 22 90 8.5 10.1 22.5 2.99 13.2 3.22 1.08 3.62%Standard deviation 0.77 0.42 8.24 1.87 1.13 0.96 2.08 1.88 1.41 1.98Published valuesa 24 94 7.9 10 23.5 3.2 14 3.4 1.15 3.6%Diff. publ. valuesa −8.33 −4.13 7.17 −1.25 4.34 6.45 6.05 5.28 6.31 −0.59




AC-E

Average 183 796 109 57.8 150 21.6 91.5 24.1 1.87 26.2%Standard deviation 0.26 0.20 0.95 1.11 1.06 1.73 1.39 1.11 3.05 1.35Published valuesa 184 780 110 59 154 22.2 92 24.2 2 26%Diff. publ. values −0.81 2.09 −0.58 2.08 2.30 2.83 0.58 0.60 6.53 −0.59

LDMb 1.1 2.2 1.3

Sample Rock type Tb Dy Ho Er Tm Yb Lu Hf Pb Th U


GP-99-11 Granite 3.91 21.7 4.67 12.33 1.83 11.75 1.72 33.0 22.4 21.2 6.07GP-99-12 Granite 29 23 8GP-99-26A Granite 3.44 19.4 4.17 11.07 1.58 10.38 1.51 27.6 19.0 23.4 7.02GP-99-27 Granite 3.57 20.0 4.25 11.40 1.68 10.69 1.55 30.4 28.1 21.9 6.55GP-99-28 Granite 3.53 19.9 4.30 11.42 1.71 10.85 1.58 30.5 21.4 21.3 5.48GP-99-29 Granite 3.67 20.6 4.39 11.57 1.72 10.93 1.58 29.6 22.1 21.0 6.55GP-99-30 Granite 18 22 5GP-99-31 Granite 23 19 5GP-99-32 Quartz Monzonite 2.00 10.6 2.24 5.67 0.78 5.07 0.76 11.3 14.0 10.2 2.64GP-99-33 Quartz Diorite 2.21 12.1 2.46 6.25 0.90 5.38 0.75 11.6 11.6 7.51 2.16GP-99-34 Monzodiorite 0.92 5.00 1.06 2.70 0.37 2.37 0.35 4.30 5.00 2.94 0.87GP-99-35 Quartz Monzonite 16 11 3GP-99-36A Quartz Monzonite 15 17 5GP-99-36B Quartz Diorite 2.84 15.4 3.19 8.27 1.21 7.04 1.01 1.72 17.0 10.5 3.24GP-99-37A Monzonite 2.26 12.1 2.54 6.53 0.97 5.74 0.85 3.72 14.3 7.6 2.35GP-99-37B Quartz Monzonite 12 8 2GP-99-38 Granite 2.25 13.4 3.01 8.39 1.26 8.60 1.28 24.6 10.0 21.6 5.89GP-99-39 Granite 10 22 6GP-99-40 Granite 3.69 20.3 4.38 11.62 1.74 10.61 1.53 30.6 34.5 22.8 6.68GP-00-133B Monzodiorite 2.29 12.4 2.63 6.68 0.92 5.93 0.88 11.7 13.0 8.06 2.36GP-00-133C Monzodiorite 1.94 10.5 2.20 5.48 0.80 4.92 0.71 2.34 10.4 5.79 1.72GP-00-134 Monzodiorite 1.59 8.97 1.98 5.14 0.73 4.75 0.71 7.90 8.00 4.36 1.35

Hovland lavas (∼1108 Ma; upper contact of Cucumber Lake granophyre)GP-99-24 Rhyolite 1.53 7.89 1.45 3.80 0.55 3.34 0.49 16.1 12.7 13.2 2.44

Misquah Hills (1106 Ma)GP-99-57 Granite 3.38 18.8 4.08 10.68 1.63 10.17 1.53 28.3 12.1 19.4 5.79GP-99-58 Granite 3.46 19.3 4.06 10.86 1.61 9.94 1.46 28.1 19.9 19.2 5.81GP-99-59 Quartz Monzonite 15 10 3GP-99-60 Granite 17 15 5GP-99-61 Granite 2.46 13.2 2.75 7.06 1.05 6.40 0.88 20.2 16.9 14.6 4.62GP-99-62 Monzodiorite 8 4 1GP-99-63 Monzodiorite 1.13 6.33 1.35 3.48 0.53 3.09 0.46 6.22 9.40 3.72 0.97GP-99-64 Monzodiorite 1.28 7.05 1.50 3.87 0.57 3.45 0.52 7.50 10.15 4.01 1.12

Undifferentiated lava flows (age uncertain; upper contact of Misquah Hills granophyre)GP-99-53 Rhyolite 17 14 6GP-99-54A Rhyolite 1.66 9.10 1.79 4.86 0.70 4.26 0.62 16.0 9.95 12.7 4.09GP-99-55 Rhyolite 1.73 9.49 1.79 4.71 0.70 4.05 0.57 16.0 9.58 13.2 3.79GP-99-56 Rhyolite 1.46 7.72 1.46 3.77 0.55 3.17 0.45 16.1 10.4 13.5 2.81

Mt. Weber granophyre (1106 Ma)GP-99-14A Granite 18 8 3

Whitefish Lake granophyre (1109 Ma)GP-99-46 Granite 2.54 14.3 3.07 8.23 1.25 7.54 1.09 17.9 11.4 17.2 4.06





GP-99-01 Granite 16 17 4GP-99-02 Quartz Monzonite 2.41 13.4 2.85 7.40 1.04 6.77 0.96 10.7 16.1 7.9 1.85GP-99-03 Granite 21.0 17.0 5.00GP-99-05 Granite 1.81 9.99 2.15 5.64 0.81 5.38 0.79 13.6 20.0 13.8 3.60GP-99-65 Quartz Monzonite 12 11 3GP-99-66 Granite 2.49 14.4 3.10 8.30 1.26 7.61 1.11 18.9 12.9 13.3 3.52GP-99-67 Granite 2.42 13.6 2.94 7.74 1.16 6.94 0.99 18.2 24.3 13.0 3.51GP-99-68 Quartz Monzonite 11 7 2GP-99-69 Quartz Monzonite 12 7 2GP-99-70A Granite 17.0 13.0 4.00GP-99-70B Monzodiorite 1.00 5.57 1.20 3.09 0.46 2.89 0.43 6.76 8.87 4.16 1.23GP-99-70C Monzodiorite 7 4 2GP-99-70D Granite 14 12 4GP-99-71 Granite 3.63 20.7 4.59 12.42 1.88 11.37 1.70 27.3 10.32 15.9 3.95GP-99-72 Quartz Monzonite 3.00 17.2 3.71 10.00 1.56 9.78 1.45 21.8 13.3 9.7 2.44GP-99-75 Granite 2.48 14.4 3.06 8.27 1.25 7.51 1.10 19.0 22.6 14.1 3.99GP-99-78 Granite 2.21 12.5 2.68 7.17 1.07 6.32 0.98 17.4 12.9 13.5 3.83

Undifferentiated lava flows (age uncertain; upper contact of Eagle Mtn. granophyre)GP-99-77B Rhyolite 2.56 14.7 3.14 8.43 1.22 8.00 1.12 17.1 7.74 12.1 3.29

Pine Mountain granophyre (1095 Ma)GP-00-115 Granite 2.15 12.0 2.63 6.99 1.07 6.80 0.95 15.5 8.94 10.21 2.59GP-00-117 Granite 3.08 17.3 3.72 9.88 1.47 9.06 1.39 22.3 18.4 13.3 3.76GP-00-118 Quartz Monzonite 12 9 2GP-99-06 Quartz Monzonite 20 12 3GP-99-07 Granite 13 14 4GP-99-08 Granite 2.37 13.6 2.87 7.71 1.15 7.03 1.01 18.4 15.2 13.1 3.45GP-99-09 Granite 2.12 11.8 2.58 6.87 1.00 6.65 0.98 15.6 7.00 13.8 3.51GP-99-10 Granite 2.02 11.83 2.45 6.84 1.01 6.64 0.85 16.9 10 12 3GP-99-41 Granite 1.84 11.0 2.38 6.56 0.97 5.88 0.89 17.4 15.2 13.6 3.73GP-99-43 Quartz Monzonite 2.55 14.3 3.12 8.14 1.15 7.50 1.10 13.8 14.0 9.10 2.51GP-99-44 Granite 12 12 3GP-99-45 Granite 2.12 12.1 2.55 6.91 1.04 6.35 0.98 17.9 16.0 13.7 3.92GP-99-52 Granite 2.36 13.5 2.93 7.96 1.22 7.34 1.13 18.3 7.04 11.9 3.17

Brule Lake Hovland Gabbro-Diabase (age uncertain; basal contact of Pine Mountain granophre)GP-99-42 Monzodiorite 1.33 7.62 1.67 4.35 0.61 3.93 0.58 7.46 8.08 3.27 0.94GP-00-112 Gabbro 0.80 4.53 0.99 2.56 0.36 2.34 0.35 3.90 4.00 2.08 0.56GP-00-123 Gabbro 14 4 2GP-00-127 Gabbro 2.38 13.5 2.84 7.42 1.12 6.89 1.05 12.6 11.8 6.90 1.87

Finland granophyre (1098 Ma)GP-99-81 Granite 3.63 21.2 4.75 12.64 1.93 11.41 1.68 23.7 11.3 19.2 4.62GP-AC1-016 Granite 5 18 4GP-AC1-029 Granite 3 16 4GP-AC1-041 Granite 5 18 5GP-AC1-056 Granite 5 15 4GP-AC1-079 Granite 7 15 4GP-AC1-100 Granite 10 14 3GP-AC1-120 Granite 29 14 3GP-AC1-134 Quartz Monzonite 21 12 2GP-AC1-156 Quartz Monzonite 18 11 3GP-AC1-178 Granite 15 13 3GP-AC1-197 Quartz Monzonite 12 11 3GP-AC1-222 Quartz Monzodiorite 18 9 2GP-AC1-225 Granite 10 19 5GP-AC1-253 Granite 7 19 5GP-AC1-329 Granite 12 18 4GP-AC1-363 Granite 13 18 4GP-AC1-396 Granite 12 17 4




Standardsa

W-2 Average 0.617 3.74 0.82 2.22 0.302 2.09 0.300 2.65 8.47 1.99 0.476%Standard deviation 3.14 1.66 3.23 1.57 5.29 1.80 8.42 3.66 6.33 3.82 5.13Published valuesa 0.63 3.8 0.84 2.4 0.328 2.08 0.32 2.4 7.6 2.2 0.48%Diff. publ. valuesa 2.02 1.69 1.99 7.52 7.90 −0.25 6.25 −10.59 −11.39 9.62 0.74

AC-E Average 4.92 29.4 6.57 17.9 2.66 17.7 2.54 31.5 37.2 17.5 4.40%Standard deviation 2.92 1.49 2.67 1.52 2.60 2.70 4.83 0.96 2.31 1.69 2.74Published valuesa 4.8 29 6.5 17.7 2.6 17.4 2.45 27.9 39 18.5 4.6%Diff. publ. values −2.51 −1.38 −1.07 −1.22 −2.46 −1.60 −3.84 −12.8 4.63 5.37 4.24

LDMb

Ce, Pb, Th, and U by XRF unless values for all REE present, in which case reported values are from ICP-MS).a Published values for Ce, Cr, Ni, Sc, V, Zn, Ga, Bam, Rb, Sr, Y, Zr, and Nb are from Govindaraju (1994); published values for REE, Hf, Pb, Th,

and U from Eggins et al. (1997).b LDM: Limit of detection of method for XRF data = the minimum concentration that can be determined at the 95.4% level of confidence; calculated

as recommended by Rousseau (2001).

Fig. 4. Variation of (a) Quartz-Alkali Feldspar-Plagioclase CIPW normative mineral compositions; (b) SiO2 vs. Na2O + K2O; (c) SiO2 vs. TiO2; (d)SiO2 vs. Nb/Th. QAP classification diagram from Streckeisen (1976) and LeMaitre (1989). CIPW normative mineral compositions were calculatedwith iron recast following the method of Irvine and Baragar (1971). A 25% of normative albite is apportioned to normative orthoclase to accountfor the high Na contents of alkali feldspars in granophyric rocks (Miller et al., 2002). Filled symbols are early stage granophyre complexes (circles,Cucumber Lake; squares, Misquah Hills; triangles, Whitefish Lake; diamonds, Mt. Weber). Open symbols are main stage granophyre complexes(circles, Pine Mtn.; squares, Eagle Mtn.; triangles, Finland).

brian R
compositions of the early and main stage granophyrecomplexes are nearly indistinguishable. Subtle excep-tions to this are shown in the plots that include Nb(e.g., SiO2 versus Nb/Zr, Nb/Th, or Nb/La) where sam-ples from the early stage granophyre complexes haverelatively higher Nb/Th contents (Table 3; Fig. 4(d)).The early and main stage granophyre complexes alsohave similar rare earth element patterns (Fig. 5(a andb)) with light-REE enrichment, negative Eu anoma-lies, and gently sloping heavy-REE patterns. Sampleswith higher silica contents generally have slightly largerEu-anomalies, higher La/Sm ratios, and higher REEabundances. Rare earth element patterns of the morefelsic rocks from the granophyre complexes are indis-
tinguishable from rhyolites of the North Shore VolcanicGroup (Fig. 5(b); Vervoort and Green, 1997). The Rb,Nb, and Y contents of the granitic samples from the
Fig. 5. Rare earth element patterns of samples from (a) early stage and(b) main stage granophyre complexes. For comparison, the range ofREE patterns of MRS rhyolites are indicated with shading in 5b. REEconcentrations are normalized to chondrite values of McDonough andSun (1995). Symbols as in Fig. 4.

esearch 157 (2007) 235–268

granophyre complexes are similar to those of “within-plate” granites as defined by Pearce et al. (1984) andhave many similarities to A-type granites such as highSiO2, Na2O + K2O, Fe/Mg, Zr, Nb, Ga, Ga/Al, Y, REE,and low Al, Mg, Ca, V, Ba, and Sr (Loiselle and Wones,1979; Collins et al., 1986; Whalen et al., 1987).

4.3. Nd isotope data

The Sm–Nd isotopic data for the 51 granophyre sam-ples and eight related volcanic and hypabyssal rocksare presented in Table 4. All samples were spiked witha mixed 149Sm–150Nd tracer, which allow for precisecalculation of 147Sm/144Nd ratios and Nd isotopic com-positions on the same sample dissolution. These data,in conjunction with the U–Pb age data for these samples(reported herein), allow the calculation of accurate initialNd isotopic compositions.

The granophyres have a fairly broad range of iso-topic compositions with initial εNd values (calculatedat their crystallization ages) that range from near chon-dritic (a high of εNd = 0.2) to strongly negative (a low ofεNd = −7.6). The volcanic and hypabyssal rocks spatiallyassociated with the granophyres also have initial Nd iso-topic compositions in this range. The 147Sm/144Nd ratiosof all samples are relatively low, consistent with theirevolved, LREE-enriched compositions, and range froma high of 0.1415 for a monzodiorite from the Cucum-ber Lake granophyre to a low of 0.1076 from a rhyolitefrom near the upper contact with the Misquah Hills gra-nophyre. The present-day �Nd values for these samplesrange from −8.1 to −19.0.

In general, the granophyres fall into two compo-sitional groups that correlate with their crystallizationages. The older granophyres have more radiogenic Ndisotopic compositions (higher 143Nd/144Nd) whereas theyounger granophyres have less radiogenic Nd isotopiccompositions (Figs. 6–8). The details of these differ-ences, and their implications, are discussed below.

5. Discussion

5.1. Chronology of silicic magmatism in theMidcontinent Rift

The U–Pb ages of MRS felsic plutonic rocks indicatetwo discrete episodes of plutonism. Granophyric rocksfrom the Cucumber Lake, Misquah Hills, Whitefish
Lake, and Mt. Weber bodies have reversed paleo-magnetic polarities and range in age from 1109 to1106 Ma with a weighted mean age of 1106.9 ± 1.8 Ma(Fig. 3). These are distinctly older than the Eagle Moun-


Table 4Nd isotope compositions of Midcontinent rift granophyres and related rocks

Sample Sma (ppm) Nda (ppm) 143Nd/144Ndb 147Sm/144Nda 143Nd/144Ndinitialc Initial εNd

c Present εNdc

Early Stage granophyre complexes (reversed paleomagnetic polarity)Cucumber Lake granophyre (1107 Ma)

GP-99-11 25.7 128 0.512057 0.1213 0.511176 −0.7 −11.3GP-99-26A 22.4 113 0.512056 0.1201 0.511184 −0.5 −11.4GP-99-27 23.3 117 0.512053 0.1202 0.511180 −0.6 −11.4GP-99-28 23.2 116 0.512053 0.1212 0.511173 −0.7 −11.4GP-99-29 24.0 118 0.512020 0.1235 0.511123 −1.7 −12.1GP-99-31 23.4 116 0.512027 0.1219 0.511141 −1.3 −11.9GP-99-32 13.5 65.8 0.512051 0.1235 0.511154 −1.1 −11.5GP-99-33 16.0 76.4 0.512096 0.1268 0.511175 −0.7 −10.6GP-99-34 5.82 27.7 0.512105 0.1272 0.511181 −0.6 −10.4GP-99-36A 21.5 104 0.512061 0.1241 0.511159 −1.0 −11.3GP-99-36B 19.2 91.9 0.512091 0.1261 0.511175 −0.7 −10.7GP-99-38 14.0 70.0 0.512011 0.1209 0.511133 −1.5 −12.2GP-99-40 25.0 126 0.512034 0.1196 0.511165 −0.9 −11.8GP-00-133B 15.2 72.2 0.512109 0.1273 0.511184 −0.5 −10.3GP-00-133C 13.1 62.0 0.512077 0.1280 0.511147 −1.2 −10.9GP-00-134 9.11 38.9 0.512205 0.1415 0.511177 −0.6 −8.5

Hovland lavas (∼1108 Ma; upper contact of Cucumber Lake granophyre)GP-99-24 13.9 81.7 0.511747 0.1025 0.511002 −4.1 −17.4

Misquah Hills (1106 Ma)GP-99-57 22.9 115 0.512040 0.1204 0.511166 −0.9 −11.7GP-99-58 23.2 116 0.512058 0.1209 0.511181 −0.6 −11.3GP-99-59 17.8 84.0 0.512056 0.1281 0.511126 −1.7 −11.4GP-99-60 18.7 93.8 0.511962 0.1204 0.511088 −2.4 −13.2GP-99-61 17.9 89.8 0.511993 0.1207 0.511117 −1.9 −12.6GP-99-62 8.90 39.0 0.512221 0.1381 0.511219 0.2 −8.1GP-99-63 7.33 33.4 0.512105 0.1325 0.511143 −1.3 −10.4GP-99-64 8.13 38.0 0.512072 0.1293 0.511133 −1.5 −11.0

Undifferentiated lava flows (age uncertain; upper contact of Misquah Hills granophyre)GP-99-54A 12.6 62.1 0.511946 0.1224 0.511057 −3.0 −13.5GP-99-55 13.3 66.7 0.511932 0.1203 0.511058 −3.0 −13.8GP-99-56 13.6 76.4 0.511758 0.1076 0.510976 −4.6 −17.2

Mt. Weber granophyre (1106 Ma)GP-99-14A 14.9 68.7 0.511970 0.1308 0.511021 −3.8 −13.0

Whitefish Lake granophyre (1109 Ma)GP-99-46 16.7 82.9 0.511925 0.1217 0.511039 −3.3 −13.9

Main Stage granophyre complexes (normal paleomagnetic polarity)Eagle Mountain granophyre (1099 Ma)

GP-99-02 15.7 74.4 0.511856 0.1279 0.510934 −5.6 −15.3GP-99-05 12.6 66.2 0.511664 0.1153 0.510833 −7.6 −19.0GP-99-65 15.8 74.6 0.511932 0.1278 0.511011 −4.1 −13.8GP-99-66 15.6 74.6 0.511892 0.1267 0.510979 −4.7 −14.6GP-99-67 16.4 80.0 0.511868 0.1237 0.510976 −4.8 −15.0GP-99-70A 14.4 70.1 0.511863 0.1242 0.510967 −5.0 −15.1GP-99-70B 5.65 26.4 0.511981 0.1292 0.511049 −3.4 −12.8GP-99-71 21.8 107 0.511879 0.1238 0.510986 −4.6 −14.8GP-99-72 18.8 86.7 0.512009 0.1311 0.511064 −3.1 −12.3GP-99-75 16.6 81.1 0.511879 0.1234 0.510990 −4.5 −14.8GP-99-78 14.9 75.4 0.511808 0.1199 0.510944 −5.4 −16.2

Undifferentiated lava flows (age uncertain; upper contact of Eagle Mtn. granophyre)GP-99-77B 16.2 77.1 0.511964 0.1274 0.511045 −3.4 −13.2



Sample Sma (ppm) Nda (ppm) 143Nd/144Ndb 147Sm/144Nda 143Nd/144Ndinitialc Initial εNd

c Present εNdc

Pine Mountain granophyre (1095 Ma)GP-00-115 13.9 69.9 0.511787 0.1204 0.510921 −5.9 −16.6GP-00-117 18.4 87.5 0.511970 0.1274 0.511054 −3.3 −13.0GP-00-118 17.9 82.9 0.511922 0.1305 0.510984 −4.7 −14.0GP-99-08 15.6 76.5 0.511857 0.1230 0.510973 −4.9 −15.2GP-99-09 13.8 69.1 0.511786 0.1209 0.510917 −6.0 −16.6GP-99-10 12.5 62.5 0.511718 0.1208 0.510850 −7.3 −18.0GP-99-41 11.6 53.1 0.511800 0.1319 0.510852 −7.3 −16.4GP-99-43 15.3 69.3 0.511999 0.1331 0.511042 −3.6 −12.5GP-99-44 14.7 70.2 0.511824 0.1264 0.510916 −6.1 −15.9GP-99-45 14.4 71.6 0.511810 0.1213 0.510938 −5.6 −16.2GP-99-52 16.2 80.4 0.511903 0.1219 0.511027 −3.9 −14.3

Brule Lake Hovland Gabbro-Diabase (age uncertain; basal contact of Pine Mountain granophre)GP-99-42 7.72 33.4 0.512092 0.1397 0.511087 −2.7 −10.7GP-00-112 4.62 20.1 0.512126 0.1386 0.511129 −1.9 −10.0GP-00-123 10.1 11.1 0.512081 0.1387 0.511084 −2.6 −10.9GP-00-127 14.7 66.2 0.512069 0.1344 0.511103 −2.4 −11.1

Finland granophyre (1098 Ma)GP-99-81 14.9 68.9 0.511980 0.1306 0.511039 −3.6 −12.8GP-AC1-079 17.3 78.0 0.512035 0.1341 0.511068 −3.0 −11.8

a 2σ errors for Sm and Nd concentrations and 147Sm/144Nd are <0.5%.rs) aver

638, 14

nophyres were emplaced at or near the unconformitybetween Archean and Paleoproterozoic basement rocksand Mesoproterozoic flows of the MRS (Miller et al.,

b Ratios normalized to 146Nd/144Nd = 0.7219. The long-term (5 yeacourse of this work was 143Nd/144Nd of 0.511869 ± 15, n = 111.

c For calculation of εNd values we used 143Nd/144NdCHUR(0) = 0.512

tain, Pine Mountain, and Finland granophyre bodies(1099–1095 Ma), which have normal magnetic polar-ity and a weighted mean age of 1097.3 ± 2.3 Ma. Thesetwo discrete age groups are consistent with the generalchronologic framework of MRS magmatism from pre-
vious studies (e.g., see Miller et al., 2002, for a recentcompilation).
The early stage granophyre bodies occupy a struc-tural position in the rift between overlying volcanic rocks

Fig. 6. Histogram plot of initial εNd values in samples from MRSgranophyre complexes. Early granophyre complexes have less negativeinitial εNd values than main stage granophyre complexes.

age of the LaJolla Nd standard at the University of Arizona over the

7Sm/144NdCHUR(0) = 0.1966, and λ147Sm = 6.54 × 10−12.

and underlying plutonic rocks of the Duluth Complex(Figs. 1 and 2(h)). At the time of intrusion, these gra-

2002). The main stage granophyre complexes, in con-

Fig. 7. SiO2 vs. initial εNd of MRS granophyre samples. Initial εNd val-ues decrease slightly with increasing SiO2 in both early and main stagegranophyre complexes. Also shown for comparison are fields of NorthShore Volcanic Group tholeiitic basalts and felsic rocks (Vervoort andGreen, 1997). Symbols as in Fig. 4.

J.D. Vervoort et al. / Precambrian R

FpmS

tiawn(wssGngee

uioRcotcifir(e(11bD

ig. 8. Initial εNd vs. Nb/Th in samples from MRS granophyre com-lexes. Compared with samples from early granophyre complexes,ain granophyre complexes have lower Nb/Th and initial εNd values.ymbols as in Fig. 4.

rast, occupy higher structural and stratigraphic levelsn the Midcontinent Rift System. The Pine Mountainnd Eagle Mountain granophyre bodies were emplacedithin volcanic flow sequences above the early stage gra-ophyre bodies in the roof zone of the Duluth ComplexFigs. 1 and 2(h)). The Finland granophyre was emplacedithin the Beaver Bay Complex, which consists of a

uite of troctolitic, gabbroic, and felsic hypabyssal intru-ions within the central part of the North Shore Volcanicroup. The main stage granophyre complexes also haveormal paleomagnetic polarities, similar to the strati-raphically higher NSVG lavas, and are consistent withmplacement during the main magmatic stage of MRSvolution.

The ages of the early granophyre bodies, and in partic-lar the Whitefish Lake Granophyre (1108.5 ± 5.6 Ma),ndicate these are among the earliest intrusive rocks rec-gnized in the Minnesota segment of the Midcontinentift System. The basal contacts of early granophyresommonly appear gradational with underlying rocksf the early gabbro series. With the anorthositic serieshe intrusive contacts are often sharp with distincthill zones on the underlying leucogabbros. This mayndicate that the felsic plutonic rocks were emplacedrst, and were later intruded and partially melted byocks of the early gabbroic and anorthositic seriesMiller et al., 2002). In northeastern Minnesota, the old-st intrusive units include the Poplar Lake IntrusionU–Pb zircon age, 1106.9 ± 0.6 Ma; Paces and Miller,
993), the Crocodile Lake Gabbro (U–Pb zircon age,107.0 ± 1.1 Ma; Davis and Green, 1997), and the dia-asic Logan Sills (U–Pb zircon age, 1108.8+4/−2 Ma;avis and Sutcliffe, 1985).
esearch 157 (2007) 235–268 259

The main stage granophyres have ages that areapproximately 10 million years younger, on average,than the early stage granophyres. These ages are coinci-dent with the ages determined from the upper normallypolarized section of the NSVG (Davis and Green, 1997)and most certainly were produced during the samestage of MRS magmatism. This stage of rift evolutionwas characterized by voluminous magmatism, includingthick volcanic sequences and large intrusive complexes(e.g., the Beaver Bay Complex and the layered andanorthositic series intrusions of the Duluth Complex).In the northeast part of the NSVG, the Pine Mtn. andEagle Mtn. granophyre bodies were intruded along,or slightly above, the contact between older, reversedvolcanic rocks and younger, normal polarity volcanicrocks. In the southwest region of the NSVG the mainstage granophyres were emplaced within the normalpolarity volcanic (NSVG) and plutonic (Beaver BayComplex) rocks that were formed during the main stage(1100–1094 Ma) of rifting.

5.2. Isotopic composition of granophyres and theirrelationship to MRS chronology

The most striking aspect of the Nd isotopic datapresented here is that the early and main stage gra-nophyres, despite having indistinguishable major andtrace-element compositions, have isotopic compositionsthat are distinct from each other. This is shown clearlyby a simple histogram presentation of the initial Ndisotopic composition of the early and main stage gra-nophyres (Fig. 6), which illustrates the difference inisotopic composition between the two stages. The earlystage granophyres have near chondritic Nd isotopic com-positions similar to the composition of the mantle plume(εNd(i) ∼ 0) proposed to be responsible for the MRS mag-matism (Nicholson and Shirey, 1990; Hutchinson et al.,1990). In contrast, the main stage granophyres have sig-nificantly more negative εNd(i) compositions similar tothe composition of Early Proterozoic and Archean crustthat exists in the region.

The isotopic distinction between the two generationsof granophyres is also shown in Fig. 7, which plots ini-tial εNd(i) versus SiO2 contents. As can be seen fromthis diagram, both generations of granophyres have over-lapping ranges in SiO2 contents from ∼47 to ∼76%,consistent with variation in the composition of theserocks from monzodiorite to granite. In spite of this
large range in major-element composition, there is anarrow range of isotopic compositions within the indi-vidual early and main stage granophyre groups. The earlystage granophyres, in total, range from εNd(i) = −3.7 to

brian R
−0.5 with only a slight decrease in εNd(i) values withincreasing SiO2. A similar narrow range in Nd isotopiccompositions exists within the main stage granophyres,although these exhibit a greater decrease in εNd(i) valueswith increasing SiO2. Juxtaposed against this relativehomogeneity within each generation of granophyre isthe distinct isotopic difference between the two gen-erations. The Nd isotopic difference between the twogenerations of granophyres indicates they were derivedfrom different combinations of crust and mantle sourcematerials. Furthermore, the Nd isotopic compositions ofthe low- and high-silica end members of both groups ofgranophyres also overlap with the mafic and felsic vol-canic rocks of the NSVG (Fig. 7) suggesting that theyshare common origins.

These age and isotopic relationships are most evidentin the four largest granophyre complexes in northeastMinnesota north of the town of Grand Marais. The twonorthernmost granophyres (Misquah Hills and Cucum-ber Lake) occur near the top of the Duluth Complexand the base of the lowermost lavas of the NSVG(Fig. 1). These granophyres have identical ages of1106.8 ± 2.8 Ma (Cucumber Lake) and 1106.0 ± 4.8 Ma(Misquah Hills) and appear, based on their present posi-tion, to be at the same stratigraphic level. Each of thesegranophyres has narrow and overlapping ranges of Ndisotopic compositions. Misquah Hills has εNd(i) valuesthat range from −2.4 to +0.2 (n = 8) and CucumberLake has a remarkably narrow range of εNd(i) valuesof −1.7 to −0.5 (n = 16). The very narrow range ofNd isotopic compositions of the Cucumber Lake gra-nophyre (a total variation of only 1.2εNd(i) units, or justslightly more than analytical uncertainty) is exceptionalbecause the 16 samples analyzed were collected through-out the exposed length and thickness of the granophyrefor the purpose of determining the isotopic heterogeneitywithin this large intrusive body. One of the goals of thisstudy was to determine if petrological variations withinthe granophyres are reflected in changes in isotopiccomposition. The samples from the Cucumber Lake gra-nophyre vary widely in composition from monzodioriteand quartz diorite with SiO2 contents as low as 47%, togranite with SiO2 contents as high as 72%. It is appar-ent, at least based on the early stage Cucumber Lakegranophyre, that there is only minor isotopic variationwith major element and petrological composition. Thisisotopic homogeneity contrasts with the major-elementheterogeneity and may indicate that these granophyres
were either (1) derived from an isotopically homoge-neous source or sources and/or (2) that if significantassimilation or magma mixing resulted in the major-element compositional variation within the granophyres,
esearch 157 (2007) 235–268

these components must have had a very similar Nd iso-topic composition to the parental granophyre magmas.It is also possible that the major element compositionalvariations of this granophyre and its border phases maybe due to partial melting and mixing of the granophyreswith later mafic magmas of similar Nd isotopic compo-sition.

The other large granophyres exposed north of GrandMarais are the main stage granophyres, Eagle Moun-tain and Pine Mountain. Both of these granophyres aresignificant ridge formers in the region (in fact, EagleMountain, which is held up by the Eagle Mountain gra-nophyre, is the highest point of land in Minnesota) andare well exposed. These granophyres have identical agesof 1098.6 ± 3.6 (Eagle Mtn.) and 1095.3 ± 3.8 (PineMtn.) and also appear to be stratigraphically equivalentbased on their present exposure. They have more nega-tive εNd(i) values than the early stage Misquah Hills andCucumber Lake granophyres, but still have narrow andoverlapping ranges of Nd isotopic compositions (EagleMtn. εNd(i) = −7.6 to −3.1; Pine Mtn. εNd(i) = −7.3 to−3.8). There is slightly more variation in Nd isotopiccompositions for the main stage granophyres and asomewhat stronger correlation of Nd isotopic compo-sition with SiO2 (Fig. 7) and Nb/Th ratios (Fig. 8). Ndisotopic compositions of the main stage granophyres rep-resent increased contribution from older evolved crustcompared to the early stage granophyres. The isotopicvariations that exist between all granophyres, however,are not very large and are neither as varied nor as negativeas the NSVG rhyolites (as low as εNd(i) = −15; Vervoortand Green, 1997). What causes the isotopic differencesbetween the two generations of granophyres? We arguebelow that these granophyres have been derived, at leastin part, by partial melting (or assimilation) of differentcrustal sources, including the mantle and perhaps occur-ring at different levels in the crust. We suggest that theearly stage granophyres, with their more radiogenic Ndisotopic compositions, have contributions from primi-tive (higher Sm/Nd) sources in the mantle and perhapsin the lower crust. The main stage granophyres, withtheir unradiogenic Nd isotopic compositions, were likelyderived from more evolved (lower Sm/Nd) sources athigher crustal levels.

5.3. Relationship between major element and Ndisotopic variations in the granophyres

The granophyre bodies typically range in composi-tion from monzodiorite through leucogranite, althoughgranite is by far the predominant rock type. The moremafic components of the granophyre sample suite typi-

rian R

criM1cmmgtpacdpltDt(S

vamspiedhrpalcmhmin

5g

snGSCn


ally occur in the lower parts of the bodies where felsicock types typically grade into the underlying gabbrontrusions (Nelson, 1991; Miller and Chandler, 1997;

iller et al., 2002). This has led some workers (Nelson,991; Miller et al., 2002) to conclude that the lowerontacts of the granophyres reflect partial melting andixing due to the intrusions of younger gabbroic mag-as. The trends observed on major and trace-element

eochemical plots are consistent with this interpreta-ion. The samples from the granophyre bodies generallylot along linear trends (Figs. 4(a–d) and 5(a and b))nd also commonly exhibit a ‘gap’ in the intermediateompositions, perhaps reflecting mixing of granite andiorite end-members. Other models that have been pro-osed for the origin of the MRS granophyres includeiquid immiscibility (Grout, 1918; Venzke, 1994), frac-ional crystallization (Schwartz and Sandberg, 1940;avidson, 1972; Nelson, 1991; Venzke, 1994), par-

ial melting (Davidson, 1972), contact metamorphismMudrey, 1973), and crustal assimilation (Taylor, 1964;tevenson, 1974).

The trends of granophyric rocks on geochemicalariation diagrams are identical to those of the basaltsnd rhyolites of the NSVG and could indicate that theafic and silicic magmas for both intrusive and extru-

ive suites could all have been derived from the samearental magma by fractional crystallization. The Ndsotopic data for the NSVG basalts and rhyolites, how-ver, show clearly that the rhyolites could not have beenerived from the same parent as the basalts because theyave strikingly different Nd isotopic compositions: thehyolites have much less radiogenic Nd isotopic com-ositions (εNd(i) = −15 to −2 with most <−10; Vervoortnd Green, 1997) than the NSVG basalts (εNd(i) ∼ 0 withittle variation; Dosso, 1984; Brannon, 1984). This indi-ates the rhyolites cannot be simple differentiates of theore voluminous NSVG basaltic magma but rather must

ave incorporated older crust in its genesis. Similar argu-ents can be made for the granophyres although the

sotopic differences between them and the gabbros areot as large.

.4. Comparisons with granophyres and relatedranites in other settings

The granites of the granophyre complexes share manyimilarities (e.g., metaluminous to weakly peralumi-ous; high SiO , Na O + K O, Fe/Mg, Zr, Nb, Ga,
2 2 2a/Al, Y and REE; and low Al, Mg, Ca, V, Ba, andr) with A-type granites (Loiselle and Wones, 1979;ollins et al., 1986; Whalen et al., 1987). The MRS gra-ophyres also share similarities with granophyric rocks
esearch 157 (2007) 235–268 261

in other within-plate and extensional settings. In the RedSea region, Miocene-Oligocene A-type granites wereemplaced along the rift flanks and represent some of theearliest magmatic activity associated with lithospherethinning (Coleman et al., 1992; Tommasini et al., 1994).These felsic bodies were typically emplaced into vol-canic sequences at shallow-levels (<2 km) and typicallycomprise 25–50% of the exposed volcanic and plutonicsection (Coleman et al., 1992). The whole-rock traceelement and isotope compositions of the Red Sea gra-nophyres have been interpreted to indicate generationby fractional crystallization of a tholeiitic parent com-bined with partial melting of late Precambrian tonaliticcrust (Coleman et al., 1992). In an alternative model,Tommasini et al. (1994) argue that the granophyres couldnot have originated solely by partial melting of upper orlower crustal rocks, or underplated basalts. Instead, thehigh Nb/Y granites are interpreted to be the result ofcombined fractional crystallization and assimilation inmulti-level magma chambers.

The Skaergaard intrusion in east Greenland (Wagerand Deer, 1939) occupies a similar geological settingas the Duluth Complex. The upper portions the Skaer-gaard intrusion intrude early Tertiary basalts and thelower parts of the intrusion intrude Archean gneisses.Granophyric rocks associated with the Skaergaard intru-sion have been divided into five different groups (Wagerand Brown, 1968). The ‘transgressive granophyres’ con-sist of ferrodiorite to leucogranite, crosscut gabbros ofthe layered series, and have radiogenic initial 87Sr/86Srvalues indicative of assimilation of older evolved crust.The volume of these granophyres is relatively smallcompared with the large granophyre bodies of theMRS. Hirschmann (1992) argues that the transgres-sive granophyres originated by reaction of differentiatedmafic melts with older evolved crust to form hybridmagmas.

In the St. Francois Mountains of Missouri, Meso-proterozoic A-type rhyolite and granite have beeninterpreted to be partial melts of juvenile calc-alkalinecrust (Menuge et al., 2002). The large volume metalu-minous rhyolites and granites are not associated withextensive mafic magmatism, so are unlikely to haveformed from a mantle plume. Rather, the positive ini-tial εNd values (+3 to +5) of these rocks are interpretedto record partial melting of juvenile calc-alkaline materi-als during extension in a back-arc setting (Menuge et al.,2002). In this model, melting is a result of heating due to
extension related magmatism. Similar models have beenproposed to explain A-type silicic magmatism in Patag-onia and West Antarctica (Kay et al., 1989; Pankhurst etal., 1998).

brian Research 157 (2007) 235–268

Fig. 9. Epsilon Nd evolution diagram of early and main stage gra-nophyre complexes, Midcontinent rift rhyolites and primitive basalts,


5.5. Model for Midcontinent rift evolution

The granophyres of the Midcontinent rift exposed innortheast Minnesota, as detailed earlier, have crystal-lization ages consistent with the intrusive rocks of theDuluth and Beaver Bay complexes as well as the extru-sive rocks of the North Shore Volcanic Group. Togetherthese data provide an increasingly clearer picture of themagmatic events that produced the diverse rocks of therift.

The early stage of MRS magmatism occurred fromapproximately 1109 to 1106 Ma. During this period thefirst large basalt flows on the northeast flank of the riftwere erupted to form the lowermost portion (reversedmagnetic polarity) of the NSVG. The earliest phases ofthe Duluth Complex (the early gabbro series and the fel-sic series) were emplaced coincident with the basalticvolcanism. The Nd isotopic composition of the major-ity of the basalts of the NSVG are near chondritic withεNd(i) values ∼ 0 with only minor variation and havebeen interpreted to represent the isotopic compositionof an enriched mantle plume (Nicholson and Shirey,1990). The most radiogenic Nd isotopic values (mostprimitive whole-rock compositions) of the early phasesof the Duluth Complex are similarly near chondritic(e.g., GP-99-33, 34, and 36; Tables 2 and 4). The sili-cic magmas produced during this period include minorrhyolites interlayered with the basalts of the NSVG aswell as the early stage granophyres. The early stagegranophyres have slightly negative to chondritic εNd(i)values (−3.7 to −0.5) that are slightly less radiogenicthan the mafic MRS magmas. The one rhyolite analyzedfrom this stage (Red Rock rhyolite; Vervoort and Green,1997) has a slightly more negative εNd(i) value (−5.2)but this value is less negative than the rhyolites from thenormally polarized upper part of the sequence. Thesedata suggest that although the silicic magmas have likelyincorporated crust in their genesis, these componentshave not greatly altered the Nd isotopic compositionof these rocks. Two options are most likely: partialmelting or assimilation of early-formed MRS productsor of high Sm/Nd Archean (e.g., mafic volcanic rocksof the Vermillion greenstone belt) or early Proterozoiccrust. These crustal components would have εNd val-ues ∼ 0 to slightly negative at 1.1 Ga (Fig. 9) and arenot greatly different from that of mantle-derived basalticmagma.

The main stage of MRS magmatism occurred from
ca. 1100 to 1094 Ma. During this period the voluminous,normally polarized basalts were erupted and formed theupper part of the NSVG. Coincident with this volcan-ism was the emplacement of the anorthositic and layered
and Archean and Proterozoic basement rocks. Modified from Vervoortand Green (1997).

series of the Duluth Complex and also the Beaver Baycomplex and related hypabyssal intrusions, includingthe Pine Mountain, Eagle Mountain, and Finland gra-nophyres. These later silicic magmas, in contrast withthe silicic magmas of the early magmatic stage, haveNd isotopic compositions significantly different fromthe mantle-derived mafic magmas. This is manifest inthe main stage granophyres with εNd(i) values of −7.6 to−3.1 and to an even greater extent in the rhyolites withεNd(i) values mostly in the range of −15 to −10 (Vervoortand Green, 1997). These compositions likely require sig-nificant contributions from older, evolved (low Sm/Nd)crustal sources. Based on the difference in Nd isotopiccompositions between the early and main stage gra-nophyres and rhyolites it appears likely that the sourcesfor the main stage silicic magmas, on balance, must befundamentally different than those incorporated by theearly stage magmas. The crustal sources for the mainstage magmas would need to have evolved to highlynegative εNd(i) values by 1.1 Ga in order to create thenegative εNd(i) values in the granophyres and rhyolites(Fig. 9). Two components are most likely: late Archeangranitic crust and early Proterozoic sediments. For exam-ple, the evolved compositions of late Archean crust in theregion typically have εNd(i) values (at 2.7 Ga) of +1 to +3but will evolve to highly negative εNd values of −20 to−13 by 1.1 Ga (e.g., the nearby Saganaga Tonalite hasεNd(1.1 Ga) value of −18; Vervoort, unpublished data).The early Proterozoic sediments of the nearby Virginia
and Rove formations have εNd values at 1.8 Ga of ∼−4to +4 (Hemming et al., 1995) that would evolve to εNdvalues of −10 to −6 by 1.1 Ga. Both the late Archeangranites and the early Proterozoic sediments are present

rian R

irtm

FdgbcG


n the basement to the MRS and could have been incorpo-ated, or partially melted, by mafic magmas as they makeheir way through the crust. The early Proterozoic sedi-

ents have comparable Nd isotopic compositions to the

ig. 10. Model for granophyre genesis in the Midcontinent Rift System. Duue to intrusion of mafic melts (dark shading) into the lower crust. These siliranophyre complexes (fine stipple) along the base of the supracrustal contacut continued crustal heating. The main magmatic stage is characterized by vharacterized by less radiogenic Nd isotopic compositions suggesting involvranophyre complexes formed during this stage were emplaced within the su

esearch 157 (2007) 235–268 263

main stage granophyres and could conceivably representthe crustal contaminant for these rocks but have moreradiogenic Nd (less negative εNd values) than most of themain stage rhyolites and, therefore, could not have been

ring the early magmatic stage, silicic melts are generated by heatingcic melts are erupted as rhyolite flows (unpatterned) and emplaced ast. The latent magmatic stage is charactaterized by minor magmatismoluminous magmatism. Silicic melts generated during this stage areement of evolved Archean and Proterozoic rocks in the upper crust.pracrustal section. Modified from Vervoort and Green (1997).

brian R
the crustal contaminant responsible for the Nd isotopicsignature in the rhyolites.

Separating these two felsic magmatic pulses is aperiod of apparent magmatic quiescence from ca. 1105to 1100 Ma where little magmatism is preserved. Thesethree distinct time periods are consistent with the early,latent, and main magmatic stages of MRS evolution assuggested by Miller and Vervoort (1996) and Miller andSeverson (2002).

Based on the existing age and isotopic data we sug-gest the following model for MRS magmatic evolutionfollowing Miller and Vervoort (1996) and Vervoort andGreen (1997). The initiation of Midcontinent magma-tism is widely attributed to be the result of an upwellingmantle plume and its interaction with the lithosphereof present-day North America (e.g., Hutchinson et al.,1990). The early magmatic stage (ca. 1109–1106 Ma)represents the initial pulse of magmatism in the MRS(Fig. 10). It is dominantly basaltic in composition andreflects the isotopic composition of an enriched plumesource (εNd(i) values ∼ 0). With few exceptions, thesemagmas appear to transit the crust without significantmodification. Some silicic magmas are produced duringthis stage but are relatively minor. The isotopic compo-sitions of these silicic magmas are produced by eitherpartial melting of earlier formed MRS rocks or higherSm/Nd Archean or Proterozoic crust. We speculate thatthis represents processes going on in the lower portionsof the crust, based on three lines of reasoning. First, ifmantle-derived magmas pond at or near the base of thecrust, thermal conditions during the early stages of riftingmay be favorable for the production of partial melts. Sec-ond, the early granophyres also have higher Nb/Th ratiosthan the later, main stage granophyres (Fig. 8), indicativeof partial melting of mantle-derived mafic componentsrather than evolved crust. Finally, there is little evi-dence in either the granophyres or rhyolites from theearly magmatic stage for contributions from sources withless radiogenic Nd isotopic compositions. Such evolvedsources, such as late Archean granites and supracrustalrocks and early Proterozoic sediments, are likely to bemore common components of the upper crust in the Mid-continent region. If these lithologies were present in thesource region for the early stage granophyres it seemslikely that they would be more readily melted than themore mafic components and their presence would bereadily recognizable in the Nd isotope systematics.

The early stage of magmatism is followed by a period
of apparent magmatic quiescence lasting from approx-imately 1105 to 1100 Ma. Miller and Vervoort (1996)and Miller and Severson (2002) refer to this hiatus inmagmatic activity as the “latent magmatic” stage and
esearch 157 (2007) 235–268

proposed that this represented a period of magmaticunderplating or ponding of magmas in the crust (Fig. 10).The ultimate cause of this hiatus is unknown, but onepossible consequence is widespread crustal heating thatmay ultimately contribute to partial melting of variouscomponents at higher levels in the crust.

The main stage of MRS evolution (∼1100–1094 Ma)represents an increased volume of magma migratingthrough the crust (Fig. 10). This results in voluminousbasaltic lava flows that form the bulk of the NSVG,and is also responsible for the genesis of large rhyolitesand the main stage granophyres. The increased magmamigration through the crust during this stage, perhapsin response to renewed extension or plume dynamics ordue to partial removal of the lithospheric mantle, marks asharp transition from the latent magmatic stage. Not onlydoes the beginning of this stage represent the initiation ofthe main pulse of MRS magmatism, but also representsa fundamental change in the Nd isotopic compositionof the silicic magmas. The granophyres and rhyolitesthat were emplaced and erupted during this stage havea distinctly more crustal character than the early stagesilicic magmas and most likely represent incorporationof a different crustal source in their genesis. We spec-ulate that this change represents the migration of theprimary location of melting and assimilation to higherlevels in the crust, perhaps due to the combination of pro-longed crustal heating during the latent magmatic stageand increased magma migration during the main mag-matic stage. In this model, the early stage silicic magmasrepresent partial melts of either younger MRS productsor older primitive sources perhaps lower in the crust andthe main stage silicic magmas represent melts of older,more evolved compositions that are more likely to bepresent at higher crustal levels.

Acknowledgements

We would like to thank George Gehrels for providingthe U–Pb zircon age data. U–Pb zircon geochronol-ogy and Nd isotope analyses at Arizona were supportedby NSF grant EAR-9903266 to Vervoort. Mineral andwhole-rock chemistry analyses at Macalester Collegewere supported by National Science Foundation (NSF)grant EAR-9601475, DUE-9651385, EAR-0520870,and EAR-9903003 to K.R. Wirth. Funding for stu-dent summer stipends and field expenses was providedthe Minnesota Space Grant Consortium. Trace-element
analyses at Colgate University were supported by an NSFgrant CHE-0006141 to Karen Harpp. This manuscriptwas greatly improved by constructive reviews by JimMiller and Mark Schmitz.

rian R

R

A

B

B

B

B

B

B

B

B

B

B

C

C

C

C


eferences

llen, D.J., Hinze, W.J., Dickas, A.B., Mudrey, M.G., Jr., 1997. Inte-grated geophysical modeling of the North American Midcontinentrift system: new interpretations for western Lake Superior, north-western Wisconsin, and eastern Minnesota. In: Ojakangas, R.W.,Dickas, A.B., Green, J.C. (Eds.), Middle Proterozoic to Cambrianrifting, central North America: Geol. Soc. Am. Spec. Paper 312.Geol. Soc. Am., Boulder, CO, pp. 47–72.

abcock, R.C., Jr., 1959. Petrogeny of the granophyre and intermediaterock in the Duluth Gabbro of northern Cook County, Minnesota.M.S. Thesis. University of Wisconsin, Madison, WI, 47 pp.

asaltic Volcanism Study Project, 1981. Pre-Tertiary continental floodbasalts. In: Basaltic Volcanism on the Terrestrial Planets. PergamonPress, New York, pp. 30–77.

ehrendt, J.C., Green, A.G., Cannon, W.F., Hutchinson, D.R., Lee,M.W., Milkereit, B., Agena, W.F., Spencer, C., 1988. Crustal struc-ture of the Midcontinent rift system: results from GLIMPCE deepseismic reflection profiles. Geology 16, 81–85.

ickford, M.E., Wooden, J.L., Bauer, R.L., 2006. SHRIMP study ofzircons from Early Archean rocks in the Minnesota River Valley:implications for the tectonic history of the Superior Province. Geol.Soc. Am. Bull. 118, 94–108.

oerboom, T.J., Miller, J.D., Jr., 1994. Bedrock geologic map ofthe Silver Island Lake, Wilson Lake, and western Toohey Lakequadrangles, Cook and Lake Counties, Minnesota. Minnesota Geo-logical Survey Miscellaneous Map M-81, scale 1:24,000.

onnichsen, B., 1971. Outcrop map of southern part of DuluthComplex and associated Keweenawan rocks, St. Louis and LakeCounties, Minnesota. Minnesota Geological Survey MiscellaneousMap M-11, scale 1:125,000.

onnichsen, B., Kauffman, D.F., 1987. Physical features of rhyolitelava flows in the Snake River Plain volcanic province, southwesternIdaho. In: Fink, J.H. (Ed.), The emplacement of silicic domes andlava flows. Geol. Soc. Am. Special Paper 212. Geological Societyof America, Boulder, CO, pp. 119–145.

rannon, J.C., 1984. Geochemistry of successive lava flows of theKeweenawan North Shore Volcanic Group. Ph.D. Thesis. Wash-ington University, St. Louis, MO, 312 pp.

ryan, S.E., Rile, T.R., Jerram, D.A., Stephens, C.J., Leat, P.T., 2002.Silicic volcanism: An undervalued component of large igneousprovinces and volcanic rifted margins. In: Menzies, M.A., Klem-perer, S.L., Ebinger, C.J., Baker, J. (Eds.), Volcanic Rifted Margins.Geol. Soc. Am. Spec. Paper 362. Geol. Soc. Am., Boulder, CO, pp.99–120.

urke, K., Dewey, K.L., 1973. Plume generated triple junctions: keyindicators in applying plate tectonics to old rocks. J. Geol. 81,406–433.

ard, K.D., 1990. A review of the Superior Province of the Cana-dian Shield, a product of Archean accretion. Precambrian Res. 48,99–156.

annon, W.F., Green, A.G., Hutchinson, D.R., Lee, M.W., Milk-ereit, B., Behrendt, J.C., Halls, H.C., Green, J.C., Dickas,A.B., Morey, G.B., Sutcliffe, R.H., Spencer, C., 1989. TheNorth American Midcontinent rift beneath Lake Superior fromGLIMPCE seismic reflection profiling. Tectonics 8, 305–332.

annon, W.F., 1992. The North American Midcontinent Rift beneaththe Lake Superior region with emphasis on its geodynamic evolu-tion. Tectonophysics 213, 41–48.

oleman, R.G., DeBari, S., Peterman, Z., 1992. A-type granite and theRed Sea opening. Tectonophysics 204, 27–40.

esearch 157 (2007) 235–268 265

Collins, W.J., Beams, S.D., White, A.J.R., 1986. Nature and origin ofA-type granites with particular reference to southeastern Australia.Contrib. Miner. Petrol. 80, 189–200.

Davidson, D.M., Jr., Burnell, J.R., Jr., 1977. Reconnaissance geo-logic map of the Brule Lake quadrangle, Cook County, Minnesota.Minnesota Geological Survey Miscellaneous Map M-29, scale1:24,000.

Davidson, D.M., Jr., 1977a. Reconnaissance geologic map of the AliceLake quadrangle, Lake County, Minnesota. Minnesota GeologicalSurvey Miscellaneous Map M-33, scale 1:24,000.

Davidson, D.M., Jr., 1977b. Reconnaissance geologic map of the BethLake quadrangle, Cook County, Minnesota. Minnesota GeologicalSurvey Miscellaneous Map M-26, scale 1:24,000.

Davidson, D.M., Jr., 1977c. Reconnaissance geologic map of theCherokee Lake quadrangle, Cook County, Minnesota. MinnesotaGeological Survey Miscellaneous Map M-30, scale 1:24,000.

Davidson, D.M., Jr., 1977d. Reconnaissance geologic map of the EagleMountain quadrangle, Cook County, Minnesota. Minnesota Geo-logical Survey Miscellaneous Map M-28, scale 1:24,000.

Davidson, D.M., Jr., 1977e. Reconnaissance geologic map of the KelsoMountain quadrangle, Cook County, Minnesota. Minnesota Geo-logical Survey Miscellaneous Map M-27, scale 1:24,000.

Davidson, D.M., Jr., 1972. Eastern part of Duluth Complex. In: Sims,P.K., Morey, G.B. (Eds.), Geology of Minnesota: a centennial vol-ume. Minnesota Geological Survey, pp. 354–360.

Davis, D.W., Green, J.C., 1997. Geochronology of the North AmericanMidcontinent rift in western Lake Superior and implications for itsgeodynamic evolution. Can. J. Earth Sci. 34, 476–488.

Davis, D.W., Paces, J.B., 1990. Time resolution of geologic events onthe Keweenawan Peninsula and implications for development ofthe Midcontinent Rift system. Earth Planet. Sci. Lett. 97, 54–64.

Davis, D.W., Sutcliffe, R.H., 1985. U–Pb ages from the Nipigonplate and northern Lake Superior. Geol. Soc. Am. Bull. 96, 1572–1579.

Dosso, L., 1984. The nature of the Precambrian subcontinental man-tle: Isotopic study (Sr, Pb, Nd) of the Keweenawan volcanism ofthe North Shore of Lake Superior. Ph.D. Thesis. University ofMinnesota, Minneapolis, MN, 220 pp.

Eggins, S.M., Woodhead, J.D., Kingsley, L.P.J., Mortimer, G.E.,Sylvester, P., McCullock, M.T., Hergt, J.M., Handler, M.R., 1997.A simple method for precise determination of 40 or more traceelements in geological samples by ICPMS using enriched isotopestandardization. Chem. Geol. 134, 311–326.

Fitz, T.J., 1988. Large felsic flows in the Keweenawan North Shore Vol-canic Group in Cook County, Minnesota. M.S. Thesis. Universityof Minnesota, Duluth, MN, 181 pp.

Gehrels, G.E., Boghossian, N.D., 2000. Reconnaissance geology andU–Pb geochronology of the west flank of the Coast Mountainsbetween Bella Coola and Prince Rupert, coastal British Columbia.In: Stowell, H.H., McClelland, W.C. (Eds.), Tectonics of the CoastMountains, southeastern Alaska and British Columbia. Geol. Soc.Am. Spec. Paper 343. Geol. Soc. Am., Boulder, CO, pp. 61–75.

Govindaraju, K., 1994. Compilation of working values and sampledescription for 383 geostandards. Geostandards Newsletter 18,Special Issue, 158 pp.

Green, J.C., 1983. Geologic and geochemical evidence for the natureand development of the middle Proterozoic (Keweenawan) Mid-
continent rift of North America. Tectonophysics 94, 413–437.
Green, J.C., 1982a. Geology of Keweenawan extrusive rocks. In: Wold,R.J., Hinze, W.J. (Eds.), Geology and tectonics of the Lake Superiorbasin. Geol. Soc. Am. Memoir 156. Geol. Soc. Am., Boulder, CO,pp. 47–55.

brian R
Green, J.C., 1982b. Geologic map of Minnesota: Two Harbors Sheet,Bedrock Geology. Minnesota Geological Survey, scale 1: 250,000.

Green, J.C., Fitz III, T.J., 1993. Extensive felsic lavas and rheoign-imbrites in the Keweenawan Midcontinent Rift plateau volcanics,Minnesota: petrographic and field recognition. J. Volcanol.Geotherm. Res. 54, 177–196.

Grout, F.F., 1918. A type of igneous differentiation. J. Geol. 26,626–658.

Grout, F.F., Schwartz, G.M., 1933. The geology of the Rove Formationand associated intrusives in northeastern Minnesota. MinnesotaGeol. Surv. Bull. 24, 99.

Grout, F.F., Sharp, R.P., Schwartz, G.M., 1959. The Geology of CookCounty, Minnesota. Minnesota Geol. Surv. Bull., 39.

Halls, H.C., Pesonen, L.J., 1982. Paleomagnetism of Keweenawanrocks. In: Wold, R.J., Hinze, W.J. (Eds.), Geology and tectonicsof the Lake Superior basin. Geol. Soc. Am. Memoir 156. Geol.Soc. Am., Boulder, CO, pp. 173–201.

Harpp, K.S., Fornari, D.J., Geist, D.J., Kurz, M.D., 2003. GenovesaSubmarine Ridge: a manifestation of plume-ridge interaction inthe northern Galapagos Islands. Geochem., Geophys., Geosyst. 4,doi:10.1029/2003GC000531.

Harris, E.C., Wirth, K.R., 2000. Petrogenesis of granitic rocks of theDunka River region, eastern Mesabi Range, Minnesota. In: 46thAnnual Meeting of the Institute on Lake Superior Geology, vol.46, Thunder Bay, Ontario, pp. 15–16.

Heaman, L.M., Machado, N., 1992. Timing and origin of Midconti-nent Rift alkaline magmatism, North America: Evidence from theColdwell Complex. Contrib. Miner. Petrol. 110, 289–303.

Hemming, S.R., McLennan, S.M., Hanson, G.N., 1995. Geochem-ical and Nd/Pb isotopic evidence for the provenance of theEarly Proterozoic Virginia Formation, Minnesota: Implications forthe tectonic setting of the Animikie Basin. J. Geol. 103, 147–168.

Hinze, W.J., Braile, L.W., Chandler, V.W., 1990. A geophysical profileof the southern margin of the Midcontinent rift system in westernLake Superior. Tectonics 9, 303–310.

Hirschmann, M., 1992. Origin of the Transgressive granophyres fromthe Layered Series of the Skaergaard intrusion, East Greenland. In:Geist, D.J., White, C.M. (Eds.), J. Volcanol. Geotherm. Res. 52,185–207 (Special Issue in Honour of Alexander R. McBirney).

Hooper, P.R., 1997. The Columbia River flood basalt provinces: cur-rent status. In: Mahoney, J.J., Coffin, M.F. (Eds.), Large igneousprovinces: continental, oceanic, and planetary flood volcanism.Am. Geophys. Union Geophys. Monogr. 100, 1–27.

Hutchinson, D.R., White, R.S., Cannon, W.F., Schulz, K.J., 1990.Keweenaw hot spot: Geophysical evidence for a 1.1 Ga mantleplume beneath the Midcontinent Rift system. J. Geophys. Res. 95,10869–10884.

Irvine, T.N., Baragar, W.R.A., 1971. A guide to the chemical classifi-cation of common volcanic rocks. Can. J. Earth Sci. 8, 523–548.

Jerde, E.A., 2003a. Layering in the Early Gabbroic Series of the Mid-continent rift system: the effect of magma density. Geol. Soc. Am.Abstr. 35, 325.

Jerde, E.A., 2003b. Gabbro/granophyre relations of the Crocodile LakeIntrusion; a possible vent for the Hovland Lavas? In: Inst. LakeSuperior Geol. 49th Ann. Mtg., Proceedings, vol. 49, pp. 35–36.

Kay, S.M., Ramos, V.A., Mpodozis, C., Sruoga, P., 1989. Late Pale-
ozoic to Jurassic silicic magmatism at the Gondwana margin;analogy of the middle Proterozoic in North America? Geology17, 324–328.
Kennedy, B.K., 2000. Petrogenesis of Midcontinent Rift granophyres:Evidence from geochemistry of Keweenewan granitic igneous

esearch 157 (2007) 235–268

complexes of northern Minnesota. B.A. Honors Thesis. MacalesterCollege, St. Paul, MN, 138 pp.

Kennedy, B.C., Sandland, T.O., Wirth, K.R., Vervoort, J.D., 2000. Pet-rogenesis of Midcontinent rift granophyres in northern Minnesota:geochemical evidence. Geol. Soc. Am. Abstr. 32, 398.

Klewin, K.W., Berg, J.H., 1991. Petrology of the KeweenawanMamainse Point lavas, Ontario: Petrogenesis and continental riftevolution. J. Geophys. Res. 96, 457–474.

King, E.R., Zietz, I., 1971. Aeromagnetic study of the midcontinentgravity high of central United States. Geol. Soc. Am. Bull. 82,2187–2208.

LeMaitre, R.W. (Ed.), 1989. A Classification of Igneous Rocks andGlossary of Terms: Recommendations of the International Unionof Geological Sciences Subcommission on the Systematics ofIgneous Rocks. Blackwell Scientific, Oxford, UK, p. 193.

Lightfoot, P.C., Sutcliffe, R.H., Doherty, W., 1991. Crustal contami-nation identified in Keweenawan Osler Group tholeiites, Ontario:a trace element perspective. J. Geol. 99, 739–760.

Loiselle, M.C., Wones, D.R., 1979. Characteristics and origin of anoro-genic granites. Geol. Soc. Am. Abstr. 11, 468.

Ludwig, K.R., 2003. Isoplot 3.00—a geochronological toolkit forMicrosoft Excel, Special publication 4, Berkeley GeochronologyCenter, Berkeley, CA, 70 pp.

McDonough, W.F., Sun, S., 1995. The composition of the Earth. Chem.Geol. 120, 223–253.

Menuge, J.F., Brewer, T.S., Seeger, C.M., 2002. Petrogenesis of met-aluminous A-type rhyolites from the St. Francois Mountains,Missouri and the Mesoproterozoic evolution of the southern Lau-rentian margin. Precambrian Res. 113, 269–291.

Miller, J.D., Jr., Chandler, V.W., 1997. Geology, petrology, and tectonicsignificance of the Beaver Bay Complex, northeastern Minnesota.In: Ojakangas, R.W., Dickas, A.B., Green, J.C. (Eds.), Middle Pro-terozoic to Cambrian Rifting, Central North America. Geol. Soc.Am. Special Paper 312. Geol. Soc. Am., Boulder, CO, pp. 73–96.

Miller, J.D., Jr., Green, J.C., Severson, M.J., Chandler, V.W., Hauck,S.A., Peterson, D.M., Wahl, T.E., 2002. Geology and MineralPotential of the Duluth Complex and Related Rocks of northeasternMinnesota. Minnesota Geological Survey Report of Investigations58, 207 pp.

Miller, J.D., Jr., Green, J.C., Severson, M.J., Chandler, V.W., Peter-son, D.M., 2001. Geologic map of the Duluth Complex and relatedrocks, northeastern Minnesota. Minnesota Geological Survey Mis-cellaneous map M-119, scale 1:200,000.

Miller, J.D., Jr., Nicholson, S.W., Cannon, W.F., 1995. The Midcon-tinent Rift in the Lake Superior region. In: Miller, J.D., Jr. (Ed.),Field Trip Guidebook for the Geology and Ore Deposits of the Mid-continent Rift in the Lake Superior Region. Minnesota GeologicalSurvey, Guidebook Series 20, pp. 1–22.

Miller, J.D., Jr., Green, J.C., Boerboom, T.J., Chandler, V.W., 1993.Geologic map of the Doyle Lake and Finland quadrangles, LakeCounty, Minnesota. Minnesota Geological Survey MiscellaneousMap M-72, scale 1:24,000.

Miller Jr., J.D., Ripley, E.M., 1996. Layered intrusions of the DuluthComplex, Minnesota, USA. In: Cawthorn, R.G. (Ed.), LayeredIntrusions. Elsevier Science, Amsterdam, pp. 257–301.

Miller Jr., J.D., Vervoort, J.D., 1996. The latent magmatic stage of theMidcontinent rift: a period of magmatic underplating and melting
of the lower crust. In: Inst. Lake Superior Geol., 42nd Ann. Mtg.,Proceedings, vol. 42, pp. 33–35.
Miller, J.D., Jr., Severson, M.J., 2002. Geology of the Duluth Com-plex. In: Miller, J.D., Jr., Green, J.C., Severson, M.J., Chandler,V.W., Hauck, S.A., Peterson, D.M., Wahl, T.E. (Eds.), Geology

rian R

M

M

N

N

N

N

P

P

P

P

P

P

P

P

P

R

S


and Mineral Potential of the Duluth Complex and Related Rocksof northeastern Minnesota. Minnesota Geological Survey Reportof Investigations 58, pp. 106–143.

orey, G.B., Mudrey, M.G., Jr., 1972. Keweenawan volcanic rocks ineast-central Minnesota. In: Sims, P.K., Morey, G.B. (Eds.), Geol-ogy of Minnesota: A Centennial Volume. Minnesota GeologicalSurvey, pp. 425–430.

udrey, M.G., Jr., 1973. Structure and petrology of the sill on PigeonPoint, Cook County, Minnesota. Ph.D. Thesis. University of Min-nesota, Minneapolis, MN, 310 pp.

elson, N., 1991. A petrologic and geochemical study of granophyricgranite in the roof of the Keweenawan Duluth Complex, EagleMountain, Cook County, Minnesota. M.S. Thesis. University ofMinnesota, Duluth, MN, 151 pp.

icholson, S.W., Shirey, S.B., 1990. Midcontinent rift volcanism inthe Lake Superior region: Sr, Nd, and Pb isotopic evidence for amantle plume origin. J. Geophys. Res. 95, 10851–10868.

icholson, S.W., Shirey, S.B., Schulz, K.J., Green, J.C., 1997. Rift-wide correlation of 1.1 Ga Midcontinent rift system basalts:implications for multiple mantle sources during rift development.Can. J. Earth Sci. 34, 504–529.

icholson, S.W., 1992. Geochemistry, petrography, and volcanologyof rhyolites of the Portage Lake Volcanics, Keweenawan Peninsula,Michigan. U. S. Geol. Surv. Bull. 1970, B1–B57.

aces, J.B., 1988. Magmatic processes, evolution, and mantle sourcecharacteristics contributing to the petrogenesis of MidcontinentRift basalts: Portage Lake Volcanics, Keweenaw Peninsula, Michi-gan. Ph.D. Thesis. Michigan Technological University Houghton,MI.

aces, J.B., Bell, K., 1989. Non-depleted sub-continental mantlebeneath the Superior Province of the Canadian Shield: Nd–Sr iso-topic and trace element evidence from Midcontinent Rift basalts.Geochim. Cosmochim. Acta 53, 2023–2035.

aces, J.B., Miller Jr., J.D., 1993. Precise U–Pb ages of Duluth Com-plex and related mafic intrusions, northeastern Minnesota: newinsights for physical, petrogenetic, paleomagnetic, and tectono-magmatic processes associated with 1.1 Ga Midcontinent rifting.J. Geophys. Res. 98, 13997–14013.

ankhurst, R.J., Weaver, S.D., Bradshaw, J.D., Storey, B.C., Ire-land, T.R., 1998. Geochronology and geochemistry of pre-Jurassicsuperterranes in Marie Byrd Land, Antarctica. J. Geophys. Res.103, 2529–2547.

atchett, P.J., Ruiz, J., 1987. Nd isotopic ages of crust formation andmetamorphism in the Precambrian of eastern and southern Mexico.Contrib. Miner. Petrol. 96, 523–528.

earce, J.A., Harris, N.B., Tindle, A.G., 1984. Trace element discrim-ination diagrams for the tectonic interpretation of granitic rocks. J.Petrol. 25, 956–983.

hillips, E.H., 2000. Petrogenesis of the enigmatic Aurora Sill, MesabiRange, Minnesota. B.A. Honors Thesis. Macalester College, St.Paul, MN, 96 pp.

hillips, E.H., Wirth, K.R., 2000. Petrogenesis of the enigmatic Aurorasill, Mesabi Range, Minnesota. Geol. Soc. Am. Abstr. 32, 150.

hillips, E., Wirth, K.R., 2001. Nd and U–Pb isotope studies of theSyenitic Aurora Sill, Mesabi Range, Minnesota. In: Inst. LakeSuperior Geol., 47th Ann. Mtg, Proceedings, vol. 47, pp. 80–81.

ousseau, R.M., 2001. Detection limit and estimate of uncertainty of
analytical XRF results. Rigaku J. 18, 33–47.
andland, T.O., Wirth, K.R., Gehrels, G.E., Vervoort, J.D., Kennedy,B.K., Harpp, K.S., 2001. Roles of fractional crystallization andassimilation in the production of Midcontinent Rift granophyres.Inst. Lake Super. Geol. 47, 85–86.

esearch 157 (2007) 235–268 267

Sandland, T.O., 2001. A fusion based method of whole rock dissolutionfor ICP-Mass Spectrometry and the origin of Midcontinent Riftgranophyres. B.A. Honors Thesis. Macalester College, St. Paul,MN, 88 pp.

Schmitz, M.D., Bowring, S.A., Ireland, T.E., 2003. Evaluation of theDuluth Complex Anorthositic Series (AS3) zircons as a geochrono-logical standard: new high-precision IDTIMS analyses. Geochim.Cosmochim. Acta 67, 3665–3672.

Schmitz, M.D., Bowring, S.A., Southwick, D.L., Boerboom, T.J.,Wirth, K.R., 2006. High-precision U–Pb geochronology in theMinnesota River Valley subprovince and its bearing on theNeoarchean to Paleoproterozoic evolution of the southern SuperiorProvince. Geol. Soc. Am. Bull. 118, 82–93.

Schwartz, G.M., Sandberg, A.E., 1940. Rock series in diabase sills atDuluth, Minnesota. Geol. Soc. Am. Bull. 51, 1135–1171.

Shirey, S.B., Klewin, K.W., Berg, J.H., Carlson, R.W., 1994. Tem-poral changes in the sources of flood basalts: Isotopic and traceelement evidence from the 1100 Ma old Keweenawan MamainsePoint Formation, Ontario, Canada. Geochim. Cosmochim. Acta58, 4475–4490.

Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial leadisotope evolution by a two-stage model. Earth Planet. Sci. Lett. 26,207–221.

Streckeisen, A.L., 1976. Classification and nomenclature of igneousrocks. Earth Sci. Rev. 12, 1–35.

Stevenson, R.J., 1974. A mafic layered intrusion of Keweenawan agenear Finland, Lake County, Minnesota. M.S. Thesis. University ofMinnesota, Duluth, MN, 160 pp.

Sutcliffe, R.H., 1987. Petrology of Middle Proterozoic diabases andpicrites from Lake Nipigon, Canada. Contrib. Miner. Petrol. 96,201–211.

Taylor, R.B., 1964. Geology of the Duluth Gabbro Complex nearDuluth, Minnesota. Minnesota Geol. Surv. Bull. 44, 63.

Tommasini, S., Poli, G., Manetti, P., Conticelli, S., 1994. Oligo-Miocene A-type granites and granophyres from Yemen: isotopicand trace-element constraints bearing on their genesis. Eur. J. Min-eral. 6, 571–590.

Van Schmus, R., 1980. Chronology of igneous rocks associated withthe Penokean orogeny in WI. In: Morey, G.B., Hanson, G.N. (Eds.),Selected studies of Archean gneisses and lower Proterozoic rocks,southern Canadian Shield. Geol. Soc. Am. Spec. Paper 182. Geo-logical Society of America, Boulder, CO, pp. 159–168.

Van Schmus, W.R., Green, J.C., Halls, H.C., 1982. Geochronology ofKeweenawan rocks of the Lake Superior region: a summary. In:Wold, R.J., Hinze, W.J. (Eds.), Geology and Tectonics of the LakeSuperior Basin. Geol. Soc. Am. Memoir 156. Geological Societyof America, Boulder, CO, pp. 165–171.

Van Schmus, W.R., Hinze, W.J., 1985. The Midcontinent Rift system.Ann. Rev. Earth Planet. Sci. 13, 345–383.

Van Wyck, N., Johnson, C.M., 1997. Common lead, Sm–Nd, and U–Pbconstraints on petrogenesis, crustal architecture and tectonic settingof the Penokean orogeny (Paleoproterozoic) in Wisconsin. U. S.A. Geol. Soc. Am. Bull. 109, 799–808.

Venzke, E.A., 1994. The geology and petrogenesis of the GreenwoodLake Area, Lake County, Minnesota. M.S. Thesis. University ofMinnesota, Duluth, MN, 166 pp.

Vervoort, J.D., Green, J.C., 1997. Origin of evolved magmas in the
Midcontinent rift system, northeast Minnesota: Nd-isotope evi-dence for melting of Archean crust. Can. J. Earth Sci. 34, 521–535.
Vervoort, J.D., Kennedy, B.C., Sandland, T.O., Wirth, K.R., 2000. Iso-topic constraints on the origin of granitic rocks in the Midcontinentrift of northern Minnesota. Geol. Soc. Am. Abstr. 32, 399.

brian R

continent rift. Can. J. Earth Sci. 34, 536–548.Zartman, R.E., Nicholson, S.W., Cannon, W.F., Morey, G.B., 1997.


Vervoort, J.D., Wirth, K.R., Gehrels, G.E., Kennedy, B.C., Sandland,T.O., 2001. Granites and rhyolites of the Midcontinent Rift. Geol.Assoc. Canada–Miner. Assoc. Can. Ann. Mtg. Abstr. 26, 157.

Wager, L.R., Brown, G.M., 1968. Layered Igneous Rocks. Oliver andBoyd, London, 588 pp.

Wager, L.R., Deer, W.A., 1939. Geological investigations in EastGreenland, Part III. The petrology of the Skaegaard intrusion.
Kangerdlugssuaq, East Greenland. Medd. om Grønland 105,1–352.
Whalen, J.B., Currie, K.L., Chappell, B.W., 1987. A-type granites: geo-chemical characteristics, discrimination and petrogenesis. Contrib.Miner. Petrol. 95, 407–419.

esearch 157 (2007) 235–268

Winchell, N.H., 1899. The Geology of Minnesota. Geologicaland Natural History Survey of Minnesota, Final Report 4,629 pp.

Wirth, K.R., Vervoort, J.D., Naiman, Z.J., 1997. Petrography andgeochemistry of southernmost volcanic rocks exposed in the Mid-

U–Th–Pb zircon ages of some Keweenawan Supergroup rocksfrom the south shore of Lake Superior. Can. J. Earth Sci. 34,549–561.

Documents

The magmatic evolution of the Midcontinent rift: New ...mille066/Teaching/5100_07/Articles/Vervoort et al... · and Machado, 1992; Davis and Paces, 1990; Paces and Miller, 1993; Davis