GEOSCIENCE CANADA Volume 31 Number 4 December 2004 179

ISSUES IN CANADIAN GEOSCIENCE

WWWWWouter Bouter Bouter Bouter Bouter BleekerleekerleekerleekerleekerGeological Survey of Canada601 Booth StreetOttawa, Ontario K1A 0E8wbleeker@nrcan.gc.ca

SUMMARYHerein I propose a vision for a newmultidisciplinary “big science” projectfor Canada’s solid earth sciences. I callthis proposed project: “Taking the Pulseof Planet Earth”. At a modest cost, andover a 5 to 10 year life-span, it wouldaim at providing the mostcomprehensive and multidisciplinaryknowledge base of the complete recordof mafic magmatism in and aroundCanada, and through stimulatinginternational cooperation, around theworld.

A complete record of maficmagmatism (spatial distribution, ages,periodicities, rates, volume estimates,estimated geochemical fluxes toatmosphere and hydrosphere, tectonicsettings, structural trends, sequencestratigraphic framework, evolving majorand trace element compositions,evolving isotopic ratios, paleomagneticinformation, paleo-intensities,associated ore deposits, etc.) constitutescritical input data for numerous first-order questions about the past andpresent evolution of our planet. Many ofsuch questions relate to issues that arecurrently a focus of attention: global

Taking the Pulse ofPlanet Earth:A Proposal for a NewMulti-disciplinary Flag-ship Project in CanadianSolid Earth Sciences

change, past climate extremes, complexEarth systems, planetary evolution,extinction events, flood volcanism,potential relationships with large impactevents, and the discovery of new oreresources.

The proposed project is afocused, “smart”, and highly efficientapproach to solve a large number ofthese seemingly unrelated but first-orderquestions in contemporary earthscience. At its core, it would have alarge dating program, aiming to provideapproximately 200 new, high-precisionages of mafic magmatic events acrossCanada and adjacent regions. ASupporting Geoscience grant systemwould ensure that other aspects of themagmatic record receive equalattention. Finally, I illustrate the impactthis project would have on paleo-continental reconstructions. As part ofthis illustration, I synthesize existingdata on two dyke swarms, the ca. 2.45Ga Matachewan and Kaminak swarms,respectively, and propose a novelSuperior-Hearne reconstruction withinsupercraton Superia.

SOMMAIREJe propose ci-contre l’idée d’un nouveauprojet multidisciplinaire de grandeenvergure dans le domaine des sciencesdes roches solides au Canada. J’ainommé ce projet « Prendre le pouls dela planète Terre ». D’un coût modeste ets’échelonnant sur cinq à dix ans, ceprojet constituerait la base deconnaissances multidisciplinaire la pluscomplète de tout le répertoire desévénements magmatiques mafiques auCanada et à son pourtour, cela, enprofitant des effets stimulant de lacoopération internationale.

Un répertoire complet desévénements magmatiques mafiques

(distribution spatiale, âges, périodicités,taux, estimations des volumes,estimations géochimiques des flux dansl’atmosphère et l’hydrosphère, cadrestectoniques, styles structuraux, cadresstratigraphiques des séquences,compositions des suites évolutives deséléments majeurs et en trace, évolutiondes ratios isotopiques, donnéespaléomagnétiques, paléo-intensités,gisements associés, etc.) constitue unregistre de données de base crucialespour nombres de grandes questions surl’évolution passée et actuelle de notreplanète. Plusieurs de ces grandesquestions sont liées à des problèmesactuels qui mobilisent l’attention, tels :les changements à l’échelle planétaire,les événements climatiques extrêmes dupassé, les systèmes planétairescomplexes, l’évolution de la planète, lesgrandes extinctions biotiques, les grandsépanchements volcaniques, les lienséventuels avec de grands impactsmétéoritiques, ainsi que la découvertede nouvelles sources de minerai.

Le projet proposé constitue uneapproche ciblée, « habile », et trèsefficace permettant de solutionner ungrand nombre de ces grands problèmes,sans liens apparents, des géosciencescontemporaines. Au cœur du projet onretrouve un grand programme dedatation visant à établir quelques 200nouvelles datations de grande précisiond’événements magmatiques au Canadaet dans les régions périphériques. Unsystème de subvention d’appointpermettrait d’assurer que d’autresaspects de la problématique magmatiquereçoivent autant d’attention.Finalement, à titre d’illustration, jedécris les répercussions escomptées d’untel projet sur les reconstitutions paléo-continentales. Dans le cadre de cetteillustration, je présente une synthèse des

180

données disponibles concernant deuxréseaux de dykes, soit les réseaux deMatachewan et de Kaminak de 2,45Ga, respectivement, et propose unenouvelle reconstitution Supérieur-Hearne dans le super-craton Supéria.

INTRODUCTIONLITHOPROBE, for more than twodecades the premier flagship project ofCanada’s solid earth sciences, iswinding down. Lithosphere-scale cross-sections have been gathered across theCanadian landmass and final synthesesare now being prepared (e.g., Percivalet al., 2004). The inevitable ending ofthis project, on the heels of its success,will leave a growing vacuum across theCanadian earth science community at atime when competition for significantscience funding is stronger than ever.Much of this competition revolvesaround highly visible “big science”projects in such fields as physics,astronomy, biochemistry and materialscience. Hence, we need to positionourselves to compete in this arena—weneed a new flagship project. Here I willoutline my vision for such a project,which I call “Taking the Pulse of PlanetEarth” and which has at its core a largemultidisciplinary study of all maficmagmatism through time. I use “maficmagmatism” as a shorthand here, butmean to include all mantle-derivedmagmatic activity other than steady-state arc magmatism; that is, all basalt(±komatiite)-dominated and (or)bimodal magmatism associated withdivergent margins, or more generallyextensional regimes, intraplatemagmatic provinces, mantle plumes,anorogenic provinces, kimberlite andalkaline provinces.

Before describing this vision insome detail, it is important tounderstand the key reasons whyLITHOPROBE was successful as a “bigscience” project, so that we mayemulate some of the same provenformulae in any future project. First ofall, proponents of LITHOPROBE had agrand vision (e.g., Keen, 1981; Cloweset al., 1984)—to understand thefundamental architecture of theCanadian crust and lithosphere fromcoast to coast, in the third dimension,and through time. Second, rapid

advances in seismic reflection and deepprobing electromagnetic techniques,largely made possible by rapid growthin modern electronics and computingpower, opened an entirely newobservational window into thecrystalline crust and, increasingly overthe last decade, into the lithosphericmantle (e.g., Calvert et al., 1995; Cooket al., 1999). Third, the transectapproach divided the overall project upinto manageable chunks, linkingregional expertise pertinent toparticular transects with thematicexpertise from across the country.Fourth, the Supporting Geosciencegrant system assured “buy-in” from amulti-disciplinary science community,bringing together the best minds fromdifferent disciplines and organizationsto work together on cross-sectionsthrough the Canadian lithosphere. Andfinally, competent leadership andmanagement made it all happen. Ofcourse, favourable timing of these andother factors (e.g., funding sources andcycles, willingness of the GeologicalSurvey of Canada to be a significantpartner, industry interest), all comingtogether at the right time, was equallycritical.

Although all of the reasonsabove are necessary for a successful“big science” project, the first two - agrand vision, and a powerful newobservational window - are undoubtedlythe most important at a conceptualstage of any new project, and hence Iwill focus on those. The “vision” Ipropose is to understand the first-orderpulse of the Earth, as reflected in itsrecord of mafic magmatic events, at aglobal scale and over 4.5 billion years ofEarth history. The “new observationalwindow” is provided by ever increasingprecision, accuracy, and resolvingpower in a broad range of disciplinesand techniques: geochronology, isotopeand trace element geochemistry,paleomagnetism, sequence stratigraphy,and geophysical approaches such astomography, receiver function analysis,and geodynamic modeling.

THE PULSE OF PLANET EARTHSilicate planets regulate their internalthermal state through melting events intheir mantles; through transport of

these melts and entrained heat to theirsurfaces; through creation of newlithosphere; and through recycling ofcold crust and lithosphere back intotheir hot convective interiors (e.g.,Davies, 1999). Today, on Earth, platetectonics is the dominant process and isdriven to a large extent by the negativebuoyancy of aged, and therefore cool,rigid, and relatively dense slabs ofoceanic lithosphere (i.e., the upperthermal boundary layer of theconvective mantle). Mantle plumes(e.g., Morgan, 1971; Ritsema andAllen, 2003; Montelli et al., 2004),originating from instabilities at deeperthermal boundary layers (i.e., the core-mantle boundary in whole-mantleconvection, e.g., Campbell, 2001), arethought to contribute about 10% of theglobal heat loss (Davies, 1999). Platetectonics and mantle plumes are twoindependent manifestations of mantleconvection, one strongly controlled bythe mechanical behaviour of the coldupper boundary layer (e.g., Anderson,2002a), the other by the fluid dynamicsof a hot deep boundary layer. Eachprocess produces voluminous maficmagmas, but on different andindependent time scales and withdifferent characteristic plan forms.

At longer time-scales, platetectonic activity appears quasi-periodicand self-organizes into thesupercontinent cycle (e.g., Schuiling,1973; Anderson, 1982; Gurnis, 1988).There are subtle indications that theperiodicity between successivesupercontinent cycles is getting shorter(Fig. 1), perhaps primarily because theaverage size of continental plates hasincreased, allowing less time lagbetween break-up and subsequentaggregation events (Bleeker, 2003; seealso Hoffman, 1997).

Running time forward into thefuture, average plate size may growfurther while the overall vigour ofmantle convection will decrease as aresult of secular heat loss andexponential decay of heat-producingelements. At some point in the future,plates will seize up (e.g., Anderson,2002b) and ultimately planet Earth willdie a “thermal death”, a state longreached by our much smallerneighbouring planet, Mars. There will

GEOSCIENCE CANADA Volume 31 Number 4 December 2004 181

be a transient era, in which heatproduction will still exceed conductiveheat loss so that the mantle underneaththe seized-up lithosphere will heat up,and transient plume activity or mantleoverturn events may temporarily bringback mobility to the lithosphere,resulting in significant resurfacing. Thisis the probable state of our closest sisterplanet, Venus. Smaller neighbouringplanets, although each having uniqueattributes, provide to some extentfuture snapshots of planet Earth as therate of thermal evolution scalesinversely with planetary radius (but, seealso Stevenson, 2003). In this sense,Venus, with a radius 0.95 that of Earth,is marginally ahead of us, whereas ourmuch smaller Moon, after initialcooling of its magma ocean, may neverhave had a mobile lithosphere, althoughthe 3.9–3.1 Ga mare basalts suggestsignificant transient heating andpossibly one or more mantle overturnevents (e.g., Hess and Parmentier,1995; Stegman et al., 2003).

Going back in time, it ispossible that mantle plumes may haveplayed a more significant role, althoughthis remains uncertain. Perhaps plumeactivity was highly episodic, in responseto strong spikes in heat flow across thecore-mantle boundary due to tidalresonances in the fluid outer core(Greff-Lefftz and Legros, 1999; seeFig. 1). Increased core heat flow wouldhave temporarily destabilized theboundary layer (D”) at the base of themantle and sharply increased plumeactivity. Because past rates ofdeceleration of the Earth’s spin (due totidal drag) are variable and not very wellknown (e.g., Denis et al., 2002, andreferences therein), the absolute timingof these resonances are poorlyconstrained but are estimated at ca.3.0±0.2 Ga and 1.8±0.2 Ga (Greff-Lefftz and Legros, 1999). Could thisexplain the post-Archean spike inkomatiite magmatism at ca. 1.9 Ga, asobserved in circum-Superior belts?

In the past, plate tectonicrecycling must have operated faster(e.g., Burke et al., 1976; Hargraves,1986) to keep pace with significantlyhigher heat production and a stillsignificantly higher primordial heatbudget retained from the era of

accretion and internal differentiation(e.g., Richter, 1988; Pollack, 1997;Davies, 1999). Thermal argumentssuggest that at some point in the distantpast, plate tectonics may not have been

Figure 1Figure 1Figure 1Figure 1Figure 1 Crustal aggregation states (supercratons, supercontinents) through time (afterBleeker, 2003). The diagram shows several first-order events in Earth evolution that mayrelate, directly or indirectly, to the crustal aggregation cycle: superplume events (e.g., Condie,2001), a compilation of mafic magmatic events (Ernst and Buchan, 2001), and the twoProterozoic time intervals during which low-latitude (global?) glaciations may have prevailed(e.g., Evans et al., 1997; Hoffman et al., 1998; Evans, 2000). Also shown are approximateintervals during which the fluid outer core may have experienced tidal resonances (Greff-Lefftz and Legros, 1999). Mid-Proterozoic Nuna was probably the first true supercontinent,whereas the late Archean may have been characterized by several discrete, transient,aggregations referred to here as supercratons: Vaalbara, Superia, Sclavia and possibly others.The diachronous breakup of these supercratons, during the Paleoproterozoic, spawned thepresent ensemble of approximately 35 Archean cratons, which now are variably incorporatedinto younger crustal assemblies. Since the assembly of Nuna, the time gaps between successivecrustal aggregation maxima appear to have become shorter (Hoffman, 1997; Bleeker, 2003).Note the correlation of intervals of global glaciation with periods of continental breakup anddispersal, and with apparent minima in the frequency of mafic magmatic events in thecontinental record (legend for the latter: red line, well-established mantle plume event; blackline, other mafic magmatic event; dashed line, poorly dated event; see Ernst and Buchan,2001).

able to keep pace with the required rateof heat loss (e.g., Davies, 1992, 1999)for the simple reason that oceaniclithosphere, on reaching a nearbytrench, was on average too young

182

(hence too buoyant to subduct), and toothin (and hence less capable of coolingthe mantle). Therefore, other and lessfamiliar geodynamic processes musthave mediated the required heat loss,perhaps involving 1) catastrophicmantle overturn and melting events(Davies, 1995; Condie, 1998; e.g., the2.72–2.70 Ga basalt-komatiite-rhyolitemagmatism documented in virtuallyevery fragment of Archean crust aroundthe world?); and 2) a less organizedreturn flow of lithospheric material intothe deep mantle in the form ofdelamination or “drip” (e.g., Davies,1992) of subcreted mafic andultramafic material after its partialtransformation into dense, refractoryeclogite. As part of the transformationinto dense eclogite, aqueous fluids andfusible tonalite-trondjhemite-granodiorite (TTG) components wereextracted from the subcreted mafic-ultramafic material and ascended intothe overlying crust to contribute to theformation of granite-greenstoneterrains. Numerous questions remain,but these less familiar, non-uniformitarian processes explain in partwhy Archean cratons do not look likemodern continental plates, even thoughmany of the second-order processeswere similar (e.g., Vlaar, 1986; Zegersand van Keken, 2001; Bleeker, 2002;Hamilton, 2003). Certainly there wasgreat mobility, involving both horizontaland vertical tectonics, but notnecessarily in the form of platetectonics sensu stricto, driven bystrongly asymmetrical subduction ofrigid, negatively buoyant slabs.

Most importantly for the presentpurpose, all these processes of first-order geodynamic interaction betweencore, mantle, and lithosphere, and ofplanetary evolution in general, leave arecord in the form of mafic magmaticevents. The products of these maficmagmatic events are preserved in thecontinental geological record throughtime, and for the last 180 million yearsalso in the oceanic record. Withincreasing precision and accuracy thesemafic magmatic events can be dated(e.g., Krogh et al., 1987; Harlan et al.,2003), while at the same time the maficmagmas provide chemical probes intothe evolution of the lithosphere andconvective mantle (e.g., Sylvester et al.,

1997; Campbell, 2003). Some of thementrain deep crustal and mantlexenoliths to the surface, providing theonly direct rock samples of thelithosphere available to us. Remanentmagnetism of mafic magmatic rocks, ifprimary, can provide paleo-intensitiesof the geomagnetic field at the time ofcrystallization (e.g., Macouin et al.,2003), thus providing constraints onprocesses in the liquid outer core.When did the inner core start to form(e.g., Labrosse et al., 2001)? And whendid the geodynamo ignite?

The very presence of thesemafic magmatic events signifiesextension, rifting and break-up, andarrival of mantle plumes or astheno-spheric upwellings. Some of the largestmafic magmatic events may be relatedto catastrophic mantle overturn eventsor slab avalanches. Basalt flows, dykes,and sills signify critical events on riftedmargins and in the evolution of manysedimentary basins and, where they canbe dated, allow time calibration ofbasin evolution. Some of the largestevents are capable of venting enoughCO2, other volcanic gases, and (or) ashinto the atmosphere to result in majorenvironmental deterioration and upsetsof the greenhouse balance. Althoughultimate causes are still being debated(e.g., Courtillot and Renne, 2003), thetwo largest Phanerozoic extinctionevents, the Permo-Triassic and K-Tboundaries, overlap in time orimmediately follow the extrusion ofsome of the largest continental floodbasalt sequences, the Siberian (e.g.,Kamo et al., 2003) and Deccan(Hofmann et al., 2000) traps,respectively. Collectively, maficmagmatic events provide perhaps themost important record of “the pulse ofthe Earth” (e.g., Larson, 1991) in itsoverall thermal evolution from a hot,young planet to a cold, dead planet.

DATING ALL MAFIC MAGMATICEVENTS THROUGH TIMEMany mafic magmatic events have beenprecisely dated (e.g., see Ernst andBuchan, 2001 for a recent compila-tion; see also Ernst and Buchan, 2004);however, many others remain undatedor have yielded imprecise, less robust,or questionable ages (e.g., many old K-Ar ages). A few mafic magmatic events

may not be datable directly, althoughwith continuing advances in U-Pb and40Ar-39Ar dating, their number isshrinking. Hence, given thesignificance of mafic magmatic eventsto nearly all first-order geodynamicprocesses, either as direct manifest-ations of these processes, or as timecalibrators, it is surprising that noconcerted national or internationaleffort has yet been made to date allsuch events and obtain a complete timeseries of mafic magmatism and itscritical attributes. Perhaps thisdeficiency is more conspicuous nowthen ever, at a time when highly capablerobotic rovers are investigating thesurface of a neighbouring planet, yetkey elements of the much more readilyavailable terrestrial record remainpoorly known.

Significant interest in acomplete record of all terrestrial maficmagmatism exists in Canada and,indeed, Canadian geochronologistshave been leading efforts to date suchrocks. There is equal interest in thisaround the world, and a Canadianinitiative may be the critical impetusfor a global initiative (e.g., see plans foran international dating campaign onpre-Mesozoic large igneous provincesbeing developed by the Large IgneousProvinces Commission, atwww.largeigneousprovinces.org). Acomprehensive study of all maficmagmatism on Earth would have majorimpacts across numerous earth sciencedisciplines, ranging from the study ofancient tectonic regimes, and thechemical and dynamic evolution of thecore and mantle, to testing putativelinks between major flood volcanism,impacts, and extinctions, and the studyof what triggers major climaticexcursions. It is beyond the scope ofthis paper to do equal justice to allthese disciplines and questions; below, Iwill merely illustrate potential impactson two related fields: Precambrianevolution and paleo-continentalreconstructions.

PALEO-CONTINENTALRECONSTRUCTIONSLet us examine plate tectonicreconstructions of various times in thepast and the impact a global datingprogram would have on such

GEOSCIENCE CANADA Volume 31 Number 4 December 2004 183

reconstructions.The extant record of Archean

crust is highly fragmented and scatteredaround the globe in about 35 significantpieces, the “Archean cratons” (Bleeker,2003). Each of these cratons is crosscutby numerous mafic dykes swarms, theprecise ages of which provide a “barcode” (Fig. 2) that identifies that pieceof crust and which can be expected tomatch, at least in part, the “bar code”from a distant craton that has sincerifted off and been dispersed aroundthe globe. For example, from the timeof its formation in an ancestrallandmass (Kenorland or Superia? seeBleeker, 2003), up to its incorporationinto the Laurentian collage (Hoffman,1988, 1989), the Superior craton wascut by as many as 20 mafic dykeswarms and intrusive events (Buchanand Ernst, 2004). To date, only abouthalf of these events have been datedprecisely. In contrast, the Slave cratonin northwestern Canada was intruded,over a similar time interval, by about10 dyke swarms, few if any of whichmatch those in the Superior. Hence,these two distinct pieces of Archeancrust were never close neighbours andlikely originated from differentancestral supercratons (Bleeker, 2003).Only by 1.9–1.8 Ga were these crustalfragments brought into relativeproximity during their amalgamationinto Laurentia (Hoffman, 1988).

Subsequently, both cratons werecut by radiating dykes of the giantMackenzie swarm at 1267 Ma(LeCheminant and Heaman, 1989),which fan out from a focal point alongthe northern edge of the Slave craton tothe southeast, as far as the northwesternSuperior craton (e.g., Ernst andBaragar, 1992; Baragar et al., 1996).Were Laurentia to break up, it wouldbe the shared record of the Mackenzieswarm dykes, and their inherentpaleomagnetic record, that would allowreconstruction of Laurentia and reveal,most conclusively, a shared, transient,residence of the Slave and Superiorcratons in this continent from sometime prior to 1267 Ma to the time ofbreakup.

Although correlation ofplatformal sequences on dispersedcratons is one of the more definitive

tools to track initially adjacent pieces ofcrust (e.g., the sequence stratigraphiccorrelation of the Transvaal Supergroup,overlying the Kaapvaal craton, with thatof the Mount Bruce Supergroup,overlying the Pilbara craton; e.g.,Cheney, 1996), it is not the mostgenerally applicable tool, because itrelies on good preservation. Duringcollision and orogenesis, platformalsequences may be entirely removed byuplift and erosion. Even whereplatformal sequences are preserved andpotential sequence stratigraphiccorrelations can be made, they may notprovide the detailed spatial anddirectional information to tightlyconstrain past configurations.

Mafic dyke swarms, on theother hand, have a significant aerial anddepth extent and commonly survivesignificant uplift. Equally important,they provide a trend and thusdirectional information, as well aspotentially precise piercing points (Parket al., 1995; Buchan and Ernst, 1997).In theory, an analysis of flow direction(e.g., from magnetic fabrics orimbrication of tabular plagioclasephenocrysts in the laminar flow regimenear dyke margins) can provide uniqueazimuthal directions. Integratedmapping, high-precision age dating,fabric studies, and paleomagnetism ofmafic magmatic events and their dykeswarms thus allow continentalfragments to be placed: 1) at a specific latitude; 2) at a specific time; 3) with a known orientation; 4) in an position that optimizesgeological continuity prior to break upand dispersal; and 5) that satisfies the specificrequirements of precise piercing points.

In an ideal case, informationfrom multiple dyke swarms (e.g., alldyke swarms of the Superior, preciselydated) would allow definition ofimproved apparent polar wander paths,providing additional information onhow a certain continental fragmentarrived at a specific location and whichother blocks were fellow travelers.When a supercraton breaks up,apparent polar wander paths bifurcate;when two cratons are joined bycollision, two previously independent

paths merge into a common path.Better-defined and more denselypopulated polar wander paths cangreatly reduce or even eliminate theambiguities due to polarity. They mayalso contribute to the understanding oftrue polar wander events (e.g., Evans,2003).

None of these variousapproaches is strictly new and variouspioneers have applied them successfully,particularly over the last two decades.However, their scale of application andintegration, under the proposedproject, would be new. As recentexamples, Wingate et al. (2002) haveshown the power of every new preciselydated mafic event and associatedpaleopole in testing the putativeLaurentia-Australia connection in themid-Proteozoic, whereas Buchan et al.(2000) have applied similar tests toLaurentia-Baltica connections.However, no comprehensive effort hasbeen made to address these questionsfor all mafic magmatic events acrossCanada. For instance, about half of themafic magmatic events in the Slave andSuperior, two of the best-studiedcratons worldwide, remain to be datedprecisely. Worse, few if any mafic dykeswarms have been dated from the Raecraton and, hence, the ancestry of thissecond-largest building block ofLaurentia remains entirely unknown.

Given the improved precision inmany of the underlying techniques, andthe growing awareness that maficmagmatic events and their inherentinformation (precise age, composition,causative tectonic event, orientation,paleolatitude and polar wander path)provide the most efficient, robust, andgenerally applicable tools to unravel pastplate configurations, the time seemsright for a national and internationaleffort.

AN EXAMPLE: A SUGGESTEDSUPERIOR-HEARNE CORRELATIONAT 2.45 GAAs one further illustration of the powerof this approach, I will present apossible reconstruction of the Hearneand Superior cratons prior to theirbreakup and dispersal sometime before2.0 Ga. The Superior craton is cut bythe large Matachewan dyke swarm

184

GEOSCIENCE CANADA Volume 31 Number 4 December 2004 185

FFFFFigurigurigurigurigure 2 e 2 e 2 e 2 e 2 Comparison of the partial “bar codes” of the two best-known Archean cratons in Laurentia, the Slave and Superior cratons. Becauseof lateral heterogeneity of the Superior craton, the older record is not easily generalized and may be less relevant to craton correlations. YoungerArchean stratigraphy of the Superior craton is based on the Abitibi greenstone belt and surrounding granitoid terrains (e.g., Card, 1990;Corfu, 1993; Bleeker, 1999; Ayer et al., 2002). Ages of dyke swarms are from compilation by Ernst and Buchan (2001). Detrital zircon recordfrom ca. 2.8 Ga quartzites (Sircombe et al., 2001) used as a proxy for basement ages in the Slave craton (see also Bleeker and Davis, 1999).Except for widespread basalt-komatiite volcanism between 2730 Ma and 2700 Ma in both cratons, there are no significant age matches.Cratonization of the Superior craton preceded that of the Slave by at least 50 to 70 million years. Following cratonization, the Slave andSuperior cratons were part of different supercratons, Sclavia and Superia. Breakup of Superia was initiated at ca. 2.45 Ga (e.g., Heaman, 1997),whereas Sclavia did not break up before 2.23 Ga. Breakup of each supercraton must have spawned several independent Archean cratons someof which must have been “nearest neighbours” to either the Slave or the Superior in these ancestral landmasses. Finally, from about 2.0 Ga to1.8 Ga, the Slave and Superior, and several other cratonic fragments, were incorporated in the crustal collage of Laurentia.

(e.g., Halls and Zhang 1998, andreferences therein), which radiates froman apparent focal point along itsHuronian margin—a probable break-upmargin that was later obscured byyounger Grenvillian overthrusts.Although complicated by the presenceof two age populations, the youngerMatachewan dykes have been datedprecisely at 2446 Ma (e.g., Heaman,1997). Many of these dykes containabundant, large plagioclase megacrysts(Fig. 3a, b, c), which set these dykesapart petrographically from many otherdyke swarms. I estimate that such anabundance of large plagioclasemegacrysts occurs in fewer than 10% ofmafic dyke swarms worldwide.Following up on earlier suggestions ofsimilar dykes in the Kaminak area ofthe Hearne craton (Christie et al.,1975), Heaman (1994) showed thatthese dykes, characterized by a similarabundance of large plagioclasemegacrysts (Fig. 3d, e), are essentiallyidentical in age (ca. 2.45 Ga, Heaman,1994). Remarkably, both dyke swarmstrend in a northerly direction, at highangles to dominant late Archeantectonic trends in each craton.

Existing paleomagnetic data(Christie et al., 1975) allow the Hearnecraton to be placed, at ca. 2.45 Ga, justto the south of the Huronian margin ofthe Superior craton (Fig. 4a, b, c) - aposition that could equate the Kaminakdykes with a southern continuation ofthe younger age population of theMatachewan dykes, perhaps as anotherbranch of a giant radiating dyke swarmassociated with a triple junction alongthe Huronian margin (Fig. 4d). Thisconfiguration (Fig. 4d) suggests thatdyke intrusion, rifting, and eventualbreak-up of supercraton Superia mayhave occurred in response to the arrival

Figure 3Figure 3Figure 3Figure 3Figure 3 Field photographs of the Matachewan (a,b,c) and Kaminak (d,e) mafic dykeswarms. Both dyke swarms have been dated at ca. 2.45 Ga (Heaman, 1994, 1997). a.a.a.a.a. North-trending Matachewan swarm cutting southern Superior Province granitoids, along highway144 from Sudbury to Timmins. bbbbb..... Margin of typical Matachewan dyke, with abundant 1–5 cm plagioclase megacrysts. c. c. c. c. c. Close-up of photo b; scale bar: 5 cm. d. d. d. d. d. Typical Kaminakdyke with moderately sized plagioclase phenocrysts. e. As photo d, but with larger and moreabundant plagioclase megacrysts; scale bar: 5 cm.

of a mantle plume head—theMatachewan plume of Heaman (1994).The proposed reconstruction wouldsuggest a more or less direct correlationbetween the partly glaciogenicHuronian Supergroup (e.g., Young etal., 2001) and Hurwitz Group (e.g.,Young, 1973, 1975), perhaps on eitherside of a large continental rift (Fig. 5).By 2218 Ma, the age of the Nipissingdiabase sills in the Huronian

Supergroup (Corfu and Andrews, 1986;Krogh et al., 1987), this rift wasevidently wide enough to impedesouthward propagation of Nipissingmagmas across the rift into the Hurwitzbasin. Nipissing sill magmas arethought to be derived, via a radiatingswarm of feeder dykes, from a focalpoint northeast of the Superior craton(Buchan et al., 1998).

This Superior-Hearne

186

reconstruction (Fig. 4d), prior toPaleoproterozoic break-up and dispersalof these cratonic fragments, is arguablyone of the most specific, and testable,pre-2.0 Ga plate reconstructionsproposed to date, although theKaminak swarm paleomagnetic pole(Christie et al., 1975) and age need tobe improved. An immediate andstraightforward test would be toinvestigate the flow direction inKaminak dykes, as the reconstructionof Figure 4 would predict a north-to-south magma flow in these dykes, awayfrom a northerly plume source.

If correct, the newreconstruction would further suggestthat Hearne crust and its large, juvenileKaminak greenstone belt represent acontinuum of the belt-like crustal

growth observed in the southernSuperior craton. Alternatively, if refined2.45 Ga paleopoles for both theKaminak and Matachewan swarmswould prove to be statistically identical(i.e., the original interpretation ofChristie et al., 1975), it would placeboth cratons in approximately theircurrent relative position at ca. 2.45 Ga.This alternate result may require aradical re-interpretation of Trans-Hudson Orogen in terms of being theproduct of opening and closure of anarrow “intra-cratonic” ocean ratherthan the prevailing view of a wideManikewan ocean (Stauffer, 1984; seealso Symons, 1998; Halls and Hanes,1999). These first-order questionsabout the fundamental architecture andevolution of Laurentia have presented

themselves loud and clear since the1975 paper of Christie et al., yet theyremain unresolved. A related questionis the identity and significance of thetwo or more age populations among thedykes of the composite Matachewanswarm.

PALEOGEOGRAPHY AT CA. 1883MA: WHERE ARE THE CONJUGATEMARGINS OF THE SUPERIOR?Although possibly distracting thediscussion from the specific lateArchean Superior-Hearnereconstruction proposed above, it isworth pointing out that very similarpaleogeographic questions presentthemselves again when we consider thedistribution of ca. 1883 Ma mafic-ultramafic magmatism around themargins of the Superior craton (e.g.,Molson dykes, Fox River Sill). If thiswidespread magmatism is related torifting and break-up of the Superiorcraton, perhaps related to arrival of amantle plume underneath the westernpart of this craton, why do we findmafic sills of identical age (1883±5 Ma;R. Parrish, unpublished date;mentioned in Van Kranendonk et al.,1993) in the Piling Group on thesouthern flank of the Rae craton? Arethese identical ages mere coincidence,or a product of insufficient ageaccuracy? Or do they suggest proximityof the Piling Group basin, and byinference the Rae margin, to theSuperior craton and the Molson plumecentered on this latter craton?

In general, which margins wereconjugate to the circum-Superiormargin at ca. 1.88 Ga and where arethey now? Given the presence of world-class mineral deposits along the westernand northern Superior margin (e.g,Thompson, at 80 Mtonnes the largestkomatiite-associated Ni-Cu-PGEdeposit in the world), this question is ofmore than scientific interest. Again, allthese questions are most clearly andcritically addressed by a comprehensivesurvey of the record of ca. 1.88 Gamafic magmatism around the Superiorand other cratons and the informationinherent therein: spatial distribution,sequence stratigraphic framework,high-resolution age data, andpaleomagnetic information.

Figure 4Figure 4Figure 4Figure 4Figure 4 Inferences from paleomagnetic data for the ca. 2.45 Ga Kaminak (Hearne craton(H), Fig. a) and Matachewan dykes (Superior craton (S), Fig. b). Kaminak paleopole based onthree sampling sites and data of Christie et al. (1975); Matachewan paleopole based on 36sampling sites data of Bates and Halls (1990). Outline of Laurentia shown for reference only.Hemispherical ambiguity allows two orientations of the Hearne craton relative to the Superior.Because of a near-equatorial position, both orientations would permit the Hearne craton tohave originated south of the Huronian margin of the Superior craton (e.g., see Fig. c for a“normal” orientation of Hearne, which allows for a better geological fit). Inset (d) showsenlargement of the proposed Superior-Hearne configuration in supercraton Superia, at 2.45Ga, with locations and approximate orientations of Matachewan (M) and Kaminak (K) dykes.Dashed, 2000 km-diameter circle, is the putative “Matachewan plume” head (Heaman,1994) centered on the Huronian margin. The Karelian craton is also shown, in a positionsuggested by Mertanen et al. (1999), as is a hypothetical rift topology.

GEOSCIENCE CANADA Volume 31 Number 4 December 2004 187

DISCUSSIONThe time is right for a comprehensive,multidisciplinary, Canada-wide study ofall mafic and intraplate magmaticevents through time, their precise ages,their tectonic settings, and otherinherent information, which not onlyaddresses critical paleogeographicquestions but has the potential toilluminate secular evolution of themantle and the core.

Results from such a projectwould produce a quantum leap in theunderstanding of Precambrian terrains,younger break-up margins, and pastplate configurations such as theSuperior-Hearne reconstructionproposed herein, particularly if aCanadian project would help stimulatesimilar initiatives on other continents(e.g., Australia, southern Africa,Europe, Russia and China). Morecomplete “bar codes” for a significantnumber of Precambrian crustalfragments would immediately lead tomore robust reconstructions of pastcrustal aggregations, including that ofRodinia (e.g., Hoffman, 1991) andNuna, as well as the first outlines of

FFFFFigurigurigurigurigure 5e 5e 5e 5e 5 Schematic cross-section through Superior-Hearne prior to their Paleoproterozoicbreak-up and dispersal, based on matching age, petrographic character, and paleopoleinformation of Matachewan and Kaminak dykes. This configuration suggests directcorrelation of the Huronian Supergroup and Hurwitz Group, both of which containglaciogenic intervals (e.g., Young, 1973, 1975; Ojakangas, 1988). Nipissing diabase dykesand sills, which are thought to have invaded the Huronian basin from a northerly source(Buchan et al., 1998), apparently failed to cross the widening rift at 2218 Ma. Alternatively,break-up had already occurred by 2218 Ma.

Archean supercratons. Equallyimportant, with a complete time seriesof mafic magmatic events, we caninvestigate apparent periodicities (e.g.,Prokoph et al., 2004), pulses and gaps,and their ultimate causes—i.e. thepulse of planet Earth. With volumeestimates, we can address productionrates over time. Is there a connectionwith climate and extinctions (e.g.,Courtillot and Renne, 2003)? Is there aconnection with large impacts (Abbottand Isley, 2002; Glikson, 2003; Beckeret al., 2004)?

The overall price tag of theproposed project would be substantialbut modest compared to that ofLITHOPROBE. Funding on the orderof 10 M$ over 10 years would allowmany of the goals to be achieved,including approximately 200 new high-precision ages. Of course, increasedfunding would allow more supportinggeoscience and hence assure broaderbuy-in from a wider spectrum ofdisciplines.

Naturally, high-precisiongeochronology would be a leaddiscipline, as precise ages are critical to

the ordering of events in their correctsequence, the unraveling of cause andeffect, and the determination of rates.However, the following disciplineswould all be natural and equal partnersin this potential “big science” project:major and trace element geochemistry,isotope geochemistry, paleomagnetism,paleo-intensity studies and insights intocore dynamics, structural geology andtectonics, regional mapping, sequencestratigraphy and basin analysis,petrology and xenolith studies,geophysics, geodynamic modeling,economic geology (Ni-Cu-PGEs, Cr,diamonds), Earth systems science andthe study of extinction events.

A strategic link with theIntegrated Oceanic Drilling Program(IODP; http://www.iodp.org) wouldextend the scope into the marine andoceanic realm. Finally, this projectwould aim to understand the evolutionof planet Earth in its true naturalcontext - the dynamic evolution of theSolar System. Hence it would seekclose links with the planetary sciencecommunity. Perhaps more so than forLITHOPROBE, internationalparticipation should be encouraged, tobenefit from expertise not currentlyresident within Canada.

Major industry players aroundthe world are spending considerabletime and money on the understandingand exploration of mafic magmaticprovinces, with the potential rewards ofworld-class mineral deposits such asthose of Norilsk, Thompson, or aMerensky Reef. Even other deposittypes, not traditionally related to maficmagmatic events (VMS, SEDEX), mayowe their ultimate origin to mafic orultramafic intrusions, i.e. as the heatengines that drove hydrothermalcirculation (e.g., Pirajno, 2000). Keyexamples would be the 2716–2710 Ma,giant Kidd Creek Cu-Zn-Ag deposit,closely associated with ultramafic flowsand intrusions (Bleeker, 1999;Hannington and Barrie, 1999), or theSullivan Pb-Zn deposit in the Belt-Purcell basin, closely associated withthe 1468 Ma Moyie sills (e.g.,Anderson and Davis, 1995). Hence,there is no doubt that industry wouldbe a supportive partner, allowingsignificant opportunities for matchingfunding.

188

Finally, the same tectonicquestions that control mafic magmaticevents are also relevant to the triggeringand emplacement of kimberliteprovinces. Indeed, as emphasized in theintroduction, this project shouldinclude all intra-plate magmatic eventsas they reflect equally on the tectonicpulse of the Earth. This project willclearly look into the mantle, and insome ways even into the core. In thissense it is a logical follow-up onLITHOPROBE, which had a definitecrustal and lithospheric perspective.Through its mantle perspective there ispotential for synergies with thePOLARIS project already underway inthe geophysical community.

CONCLUSIONS“Taking the Pulse of Planet Earth”, bymeans of a comprehensivemultidisciplinary study of the completerecord of mafic magmatism, includinga large dating program, is a “smart” andhighly efficient approach to address anumber of first-order questions incontemporary Earth science over thenext 5 to 10 years.

A complete record ofmafic magmatism (ages, periodicities,rates, volume estimates, geochemicalfluxes, spatial distribution, structuraltrends, evolving isotope and traceelement ratios, paleomagneticinformation, associated ore deposits,etc.) will provide critical constraints onissues as diverse as paleogeographicreconstructions and the supercontinentcycle; the triggers of past climateextremes; complex Earth systemsthrough time; crustal growth, core andmantle evolution; evolution of the coredynamo; causes of flood volcanism, andthe potential relationships withextinction events; the hypothesizedrelationship between flood volcanismand large impact events; and, last butnot least, the discovery of new oredeposits of strategic minerals.

The proposed project iscertain to stimulate a quantum leap inthe understanding of Precambrianterrains and their paleogeographicdistribution through time. Moreimportantly, a more completeunderstanding of mafic magmatism

through time will provide new insightsinto the evolution of planet Earth.Finally, this project would promote, ata modest cost, new links betweennumerous earth science disciplines andfacilitate further integration betweenearth and planetary sciences.

ACKNOWLEDGEMENTSIdeas discussed herein were firstpresented at the 2002 GAC conferencein Saskatoon and have since developedthrough discussions with many, inparticular Ken Buchan, Richard Ernst,Don Davis, Sandra Kamo, Bill Davis,Cees van Staal, Simon Hanmer, BobBarager, Tom Skulski, John Waldron,Paul Sylvester, Larry Heaman, HenryHalls, Keith Sircombe, John Ketchum,Kevin Cassidy, Dave Evans, IanDalziel, and many others. Many ofthese scientists share my vision,although individual approaches oremphases would naturally differsomewhat—this is to be expected. At aminimum, I hope this paper willstimulate discussion on the future ofthe Solid Earth Sciences in Canada.More so, I hope to convince fellowscientists that mafic magmatism is afirst-order process and diagnostic ofboth long-term and short-term planetaryevolution. As a tool, its enormouspotential remains grossly underutilized.I thank Ken Buchan for kindly assistingwith the plotting of Figure 4. The twofield photos of Kaminak dykes wereprovided by Simon Hanmer. BobBaragar suggested the flow directiontest of the Superior-Hearnereconstruction. Geological Survey ofCanada contribution number 2003305.

REFERENCESAbbott, D.H. and Isley, A.E., 2002,

Extraterrestrial influences on mantle plumeactivity: Earth and Planetary ScienceLetters, v. 205 (1-2), p. 53-62.

Anderson, D.L., 1982, Hotspots, polarwander, Mesozoic convection and thegeoid: Nature, 297 (5865), 391-393.

Anderson, D.L., 2002a, How many plates?:Geology, v. 30 (5), p. 411-414.

Anderson, D.L., 2002b, Plate tectonics as afar-from-equilibrium self-organized system,in Stein, S. and Freymueller, J.T., eds.,Plate Boundary Zones: AmericanGeophysical Union, Geodynamics Series,v. 30, p. 411-425.

Anderson, H.E. and Davis, D.W., 1995, U-Pbgeochronology of the Moyie Sills, PurcellSupergroup, southeastern BritishColumbia; implications for theMesoproterozoic geological history of thePurcell (Belt) Basin: Canadian Journal ofEarth Sciences, v. 32 (8), p. 1180-1193.

Ayers, J., Amelin, Y., Corfu, F., Kamo, S.,Ketchum, J., Kwok, and Trowell, N., 2002,Evolution of the southern Abitibigreenstone belt based on U-Pbgeochronology: autochthonous volcanicconstruction followed by plutonism,regional deformation, and sedimentation:Precambrian Research, v. 115, p. 63-95.

Baragar, W.R.A., Ernst, R.E., Hulbert, L. andPeterson, T., 1996, Longitudinalpetrochemical variation in the Mackenziedyke swarm, northwestern CanadianShield: Journal of Petrology, v. 37 (2),p. 317-359.

Bates, M.P. and Halls, H.C., 1990, Regionalvariation in paleomagnetic polarity of theMatachewan dykes swarm related to theKapuskasing Structural Zone, Ontario:Canadian Journal of Earth Sciences, v. 27,p. 200-211.

Becker, L., Poreda, R.J., Basu, A.R., Pope,K.O., Harrison, T.M., Nicholson, T.M.,and Iasky, R., 2004, Bedout: A possibleend-Permian impact crater offshore ofnorthwestern Australia: Science, 13 May2004 (10.1126/science.1093925).

Bleeker, W., 1999, Structure, stratigraphy, andprimary setting of the Kidd Creekvolcanogenic massive sulphide deposits: asemiquantitative reconstruction, inHannington, M.D. and Barrie, C.T., eds.,The giant Kidd Creek VolcanogenicMassive Sulfide Deposit, Western AbitibiSubprovince, Canada: Economic GeologyMonograph 10, p. 71-121.

Bleeker, W., 2002, Archaean tectonics: areview, with illustrations from the Slavecraton, in Fowler, C.M.R., Ebinger, C.J.and Hawkesworth, C.J., eds., The EarlyEarth: Physical, Chemical and BiologicalDevelopment: Geological Society SpecialPublications, v. 199, p. 151-181.

Bleeker, W., 2003, The late Archean record: apuzzle in ca. 35 pieces: Lithos, v. 71 (2-4),99-134.

Bleeker, W. and Davis, W.J., 1999, The 1991-1996 NATMAP Slave Province Project:Introduction: Canadian Journal of EarthSciences, v. 36 (7), p. 1033-1042.

Buchan, K.L. and Ernst, R.E., 1997,Restoration of the primary geometry ofgiant dyke swarms: constraints onpaleocontinental reconstructions [abstract]:Geological Association of Canada, AnnualMeeting, May 19-21, 1997, Abstractvolume, p. A-20.

Buchan, K.L. and Ernst, R.E., 2004, Diabasedyke swarms and related units in Canadaand adjacent regions: Geological Survey ofCanada, Map 2022A, scale 1:5,000,000.

GEOSCIENCE CANADA Volume 31 Number 4 December 2004 189

Buchan, K.L., Mortensen, J.K., Card, K.D.and Percival, J.A., 1998, Paleomagnetismand U-Pb geochronology of diabase dykeswarms of Minto block, Superior Province,Quebec, Canada: Canadian Journal ofEarth Sciences, v. 35, p. 1054-1069.

Buchan, K.L., Mertanen, S., Park, R.G.,Pesonen, L.J., Elming, S.A., Abrahamsen,N. and Bylund, G., 2000, Comparing thedrift of Laurentia and Baltica in theProterozoic; the importance of keypalaeomagnetic poles: Tectonophysics,v. 319 (3), p. 167-198.

Burke, K., Dewey, J.F. and Kidd, W.S.F., 1976,Dominance of horizontal movements, arcand microcontinental collisions during thelater permobile regime, in Windley, B.F.,ed., The Early History of the Earth: JohnWiley & Sons, New York, USA p. 113-129.

Calvert, A.J., Sawyer, E.W., Davis, W.J. andLudden, J.N., 1995, Archaean subductioninferred from seismic images of a mantlesuture in the Superior Province: Nature,v. 375 (6533), p. 670-674.

Campbell, I.H., 2001, Identification of ancientmantle plumes, in Ernst, R.E. and Buchan,K.L., eds., Mantle plumes: TheirIdentification Through Time: GeologicalSociety of America, Special Paper 352, p. 5-21.

Campbell, I.H., 2003, Constraints oncontinental growth models from Nb/ Uratios in the 3.5 Ga Barberton and otherArchaean basalt-komatiite suites: AmericanJournal of Science, v. 303, p. 319-351.

Card, K.D., 1990, A review of the SuperiorProvince of the Canadian Shield, a productof Archean accretion: PrecambrianResearch, v. 48 (1-2), p. 99-156.

Cheney, E.S., 1996, Sequence stratigraphy andplate tectonic significance of the Transvaalsuccession of Southern Africa and itsequivalent in Western Australia:Precambrian Research, v. 79 (1-2), p. 3-24.

Christie, K.W., Davidson, A. and Fahrig, W.F.,1975, The paleomagnetism of Kaminakdikes; no evidence of significant Hudsonianplate motion: Canadian Journal of EarthSciences, v. 12 (12), p. 2048-2064.

Clowes, R.M., Green, A.G., Yorath, C.J.,Kanasewich, E.R., West, G.F. and Garland,G.D., 1984, Lithoprobe; a nationalprogram for studying the third dimension ofgeology: Journal of the Canadian Society ofExploration Geophysicists, v. 20 (1), p. 23-39.

Condie, K.C., 1998, Episodic continentalgrowth and supercontinents: a mantleavalanche connection?: Earth and PlanetaryScience Letters, v. 163 (1-4), p. 97-108.

Condie, K.C., 2001, Mantle plumes and theirrecord in Earth history: CambridgeUniversity Press, New York, USA, 306 p.

Cook, F.A., van der Velden, A.J., Hall, K.W.and Roberts, B.J., 1999, Frozen subductionin Canada’s Northwest Territories;

Lithoprobe deep lithospheric reflectionprofiling of the western Canadian Shield:Tectonics, v. 18 (1), p. 1-24.

Corfu, F., 1993, The evolution of the southernAbitibi greenstone belt in light of preciseU-Pb geochronology: Economic Geology,v. 88 (6), p. 1323-1340.

Corfu, F. and Andrews, A.J., 1986, A U-Pb agefor mineralized Nipissing Diabase,Gowganda, Ontario; Canadian Journal ofEarth, v. 23 (1), p. 107-109.

Courtillot, V.E. and Renne, P.R., 2003, Onthe ages of flood basalt events: ComptesRendus, Geoscience, v. 335 (1), p. 113-140.

Davies, G.F., 1992, On the emergence of platetectonics: Geology, v. 20, p. 963-966.

Davies, G.F., 1995, Punctuated tectonicevolution of the Earth: Earth and PlanetaryScience Letters, v. 136 (3-4), p. 363-379.

Davies, G.F., 1999, Dynamic Earth: CambridgeUniversity Press, Cambridge, UnitedKingdom, 458 p.

Denis, C., Schreiber, A.A., Varga, P., andZavoti, J., 2002, Despinning of the earthrotation in the geological past andgeomagnetic paleointensities: Journal ofGeodynamics, v. 34, p. 667-685.

Ernst, R.E. and Baragar, W.R.A., 1992,Evidence from magnetic fabric for the flowpattern of magma in the Mackenzie giantradiating dyke swarm: Nature, v. 356,p. 511-513.

Ernst, R.E. and Buchan, K.L., 2004, Igneousrocks associations in Canada 3. Largeigneous provinces (LIPs) in Canada andadjacent regions: 3 Ga to present:Geoscience Canada, v. 31, p. 103-126.

Ernst, R.E. and Buchan, K.L., 2001, Largemafic magmatic events through time andlinks to mantle-plume heads, in Ernst, R.E.and Buchan, K.L., eds., Mantle Plumes:Their Identification Through Time:Geological Society of America, SpecialPaper 352, p. 483-575.

Evans, D.A., 2000, Stratigraphic,geochronological, and paleomagneticconstraints upon the Neoproterozoicclimatic paradox: American Journal ofScience, v. 300, p. 347-433.

Evans, D.A.D., 2003, True polar wander andsupercontinents: Tectonophysics, v. 362,p. 303-320.

Evans, D.A., Beukes, N.J., and Kirschvink,J.L., 1997, Low-latitude glaciation in thePaleoproterozoic era. Nature, v. 386(6622), p. 262-266.

Glikson, A., 2003, Comment on“Extraterrestrial influences on mantleplume activity”: Earth and PlanetaryScience Letters, v. 215, p. 425-427.

Greff-Lefftz, M. and Legros, H., 1999, Corerotational dynamics and geological events:Science, v. 286 (5445), p. 1707-1709.

Gurnis, M., 1988, Large-scale mantleconvection and the aggregation anddispersal of supercontinents: Nature, v. 332

(6166), p. 695-699.Halls, H.C. and Hanes, J.A., 1999,

Paleomagnetism, anisotropy of magneticsusceptibility, and argon-argongeochronology of the Clearwateranorthosite, Saskatchewan, Canada:Tectonophysics, v. 312, p. 235-248.

Halls, H.C. and Zhang, B., 1998, Upliftstructure of the southern Kapuskasing zonefrom 2.45 Ga dike swarm displacement:Geology, v. 26, p. 67-70.

Hamilton, W.B., 2003, An alternative Earth:GSA Today, 13 (11), 4-12.

Hannington, M.D. and Barrie, C.T., 1999,The giant Kidd Creek volcanogenic massivesulfide deposit, western AbitibiSubprovince, Canada. Economic Geology,Monograph 10, 672 p.

Hargraves, R.B., 1986, Faster spreading orgreater ridge length in the Archean:Geology, v. 14, p. 750-752.

Harlan, S.S., Heaman, L., LeCheminant, A.N.and Premo, W.R., 2003, Gunbarrel maficmagmatic event: a key 780 Ma time markerfor Rodinia plate reconstructions: Geology,v. 31, p. 1053-1056.

Heaman, L.M., 1994, 2.45 Ga global maficmagmatism: Earth’s oldest superplume?[abstract]: Eighth International Conferenceon Geochronology, Cosmochronology, andIsotope Geology; Program with Abstracts;U.S. Geological Survey Circular 1107,p. 132.

Heaman, L.M., 1997, Global maficmagmatism at 2.45 Ga; remnants of anancient large igneous province?: Geology,v. 25 (4), p. 299-302.

Hess, P.C. and Parmentier, E.M., 1995, Amodel for the thermal and chemicalevolution of the Moon’s interior;implications for the onset of marevolcanism: Earth and Planetary ScienceLetters, v. 134 (3-4), p. 501-514.

Hoffman, P.F., 1988, United plates of America,the birth of a craton: early Proterozoicassembly and growth of Laurentia: AnnualReview of Earth and Planetary Sciences,v. 16, p. 543-603.

Hoffman, P.F., 1989, Precambrian geology andtectonic history of North America, in Bally,A.W. and Palmer, A.R., eds., The Geologyof North America: An Overview:Geological Society of America, Boulder,Colorado, p. 447-512.

Hoffman, P.F., 1991, Did the breakout ofLaurentia turn Gondwanaland inside-out?:Science, v. 252 (5011), p. 1409-1412.

Hoffman, P.F., 1997, Tectonic genealogy ofNorth America, in van der Pluijm, B.A. andMarshak, S., eds., Earth Structure andIntroduction to Structural Geology andTectonics: McGraw Hill, New York, USA,p. 459-464.

Hoffman, P.F., Kaufman, A.J., Halverson, G.P.and Schrag, D.P., 1998, A Neoproterozoicsnowball Earth: Science, v. 281, p. 1342-1346.

190

Hofmann, C., Feraud, G. and Courtillot, V.,2000, 40Ar/39Ar dating of mineralseparates and whole rock from the WesternGhats lava pile: further constraints onduration and age of the Deccan traps: Earthand Planetary Science Letters, v. 180,p. 13-27.

Keen, C.E., 1981, Lithoprobe; geosciencestudies on the third dimension; a co-ordinated national geoscience project forthe 1980s: Geoscience Canada, v. 8 (3),p. 117-125.

Krogh, T.E., Corfu, F., Davis, D.W., Dunning,G.R., Heaman, L.M., Kamo, S.L.,Machado, N., Greenhough, J.D., andNakamura, E., 1987, Precise U-Pb isotopicages of diabase dykes and mafic toultramafic rocks using trace amounts ofbaddeleyite and zircon, in Halls, H.C. andFahrig, W.F., eds., Mafic Dyke Swarms:Geological Association of Canada, SpecialPaper 34, p. 147-152.

Kamo, S.L., Czamanske, G.K., Amelin, Y.,Fedorenko, V.A., Davis, D.W. andTrofimov, V.R., 2003, Rapid eruption ofSiberian flood-volcanic rocks and evidencefor coincidence with the Permian–Triassicboundary and mass extinction at 251 Ma:Earth and Planetary Science Letters, v. 214(1-2), p. 75-91.

Labrosse, S., Poirier, J.P. and Le, M.J.L., 2001,The age of the inner core: Earth andPlanetary Science Letters, v. 190 (3-4),p. 111-123.

Larson, R.L., 1991, The latest pulse of theEarth. Geology, v. 19, p. 547-550.

LeCheminant, A.N. and Heaman, L.M., 1989,Mackenzie igneous events, Canada: middleProterozoic hotspot magmatism associatedwith ocean opening: Earth and PlanetaryScience Letters, v. 96 (1-2), p. 38-48.

Macouin, M., Valet, J.P., Besse, J., Buchan, K.,Ernst, R., LeGoff, M. and Sharer, U., 2003,Low paleointensities recorded in 1 to 2.4Ga Proterozoic dykes, Superior Province,Canada: Earth and Planetary ScienceLetters, v. 213, p. 79-95.

Mertanen, S., Halls, H.C., Vuollo, J.I.,Personen, P.J., Stepanov, V.S., 1999,Paleomagnetism of 2.44 Ga mafic dykes inRussian Karelia, eastern FennoscandianShield—implications for continentalreconstructions: Precambrian Research,v. 98, p. 197-221.

Montelli, R., Nolet, G., Dahlen, F.A., Masters,G., Engdahl, E.R. and Hung, S-H., 2004,Finite frequence tomography reveals avariety of plumes in the mantle: Science,v. 303, p. 338-342.

Morgan, W.J., 1971, Convection plumes in thelower mantle: Nature, v. 230 (5288), p. 42-43.

Ojakangas, R.W., 1988, Glaciation: anuncommon “mega-event” as a key tointracontinental and intercontinentalcorrelation of early Proterozoic basin fill,

North American and Baltic cratons, inKleinspehn, K.L. and Paola, C., eds., NewPerspectives in Basin Analysis: Springer-Verlag, New York, USA, p. 431-444.

Park, J.K., Buchan, K.L. and Harlan, S.S.,1995, A proposed giant radiating dykeswarm fragmented by the separation ofLaurentia and Australia based onpaleomagnetism of ca. 780 Ma maficintrusions in western North America: Earthand Planetary Science Letters, v. 132,p. 129-139.

Percival, J.A, Bleeker, W., Cook, F.A., Rivers,T., Ross, G. and van Staal, C., 2004, Pan-LITHOPROBE workshop IV: Intra-orogen correlations and comparativeorogenic anatomy: Geoscience Canada, v.31, p. 23-39.

Pirajno, F., 2000, Ore Deposits and MantlePlumes: Kluwer Academic Publishers,Dordrecht, The Netherlands, 556 p.

Pollack, H.N., 1997, Thermal characteristicsof the Archaean, in de Wit, M.J. andAshwal, L.D. eds., Greenstone Belts:Oxford Monographs on Geology andGeophysics 35, p. 223-232.

Prokoph, A., Ernst, R.E. and Buchan, K.L.,2004, Time-series analysis of large igneousprovinces: 3500 Ma to present: The Journalof Geology, v. 112, p. 1-22.

Richter, F.M., 1988, A major change in thethermal state of the Earth at the Archean-Proterozoic boundary: consequences forthe nature and preservation of continentallithosphere, in Menzies, M.A. and Cox,K.G., eds., Oceanic and ContinentalLithospheres: Similarities and Differences:Journal of Petrology, Special LithosphereIssue, p. 39-52.

Ritsema, J. and Allen, R.M., 2003, The elusivemantle plume: Earth and Planetary ScienceLetters, v. 207 (1-4), p. 1-12.

Schuiling, R.D., 1973, Active role ofcontinents in tectonic evolution—geothermal models, in de Jong, K. andScholten, R., eds., Gravity and Tectonics:John Wiley and Sons, New York, USA,p. 35-47.

Sircombe, K.N., Bleeker, W., and Stern, R.A.,2001, Detrital zircon geochronology andgrain-size analysis of ~2800 MaMesoarchean proto-cratonic coversuccession, Slave Province, Canada: Earthand Planetary Science Letters, v. 189 (3-4),p. 207-220.

Stauffer, M.R., 1984, Manikewan: an earlyProterozoic ocean in central Canada, itsigneous history and orogenic closure:Precambrian Research, v. 25 (1-3), p. 257-281.

Stegman, D.R., Jellinek, A.M., Zatman, S.A.,Baumgardner, J.R. and Richards, M.A.,2003, An early lunar core dynamo driven bythermomechanical mantle convection:Nature, v. 421 (6919), p. 143-146.

Stevenson, D.J., 2003, Styles of mantleconvection and their influence on planetaryevolution: Comptes Rendus Geoscience,v. 335, p. 99-111.

Sylvester, P.J., Campbell, I.H., and Bowyer,D.A., 1997, Niobium/uranium evidence forearly formation of the continental crust:Science, v. 275, p. 521-523.

Symons, D.T.A., 1998, Precambrian platetectonic models: shifting the paleomagneticparadigm for orogens such as the Trans-Hudson in Canada: Physics and Chemistryof the Earth, v. 23 (7), p. 753-759.

Van Kranendonk, M.J., St-Onge, M.R. andHenderson, J.R., 1993, Paleoproterozoictectonic assembly of Northeast Laurentiathrough multiple indentations:Precambrian Research, v. 63, p. 325-347.

Vlaar, N.J., 1986, Archean global dynamics:Geologie and Mijnbouw, v. 65, p. 91-101.

Wingate, M.T.D., Pisarevsky, S.A., and Evans,D.A.D., 2002, Rodinia connection betweenAustralia and Laurentia: no SWEAT, noAUSWUS?: Terra Nova, v. 14 (2), p. 121-128.

Young, G.M., 1973, Tillites and aluminousquartzites as possible time marker formiddle Precambrian (Aphebian) rocks ofNorth America, in Young, G.M., ed.,Huronian Stratigraphy and Sedimentation:Geological Association of Canada, SpecialPaper 12, p. 97-127.

Young, G.M., 1975, Geochronology ofArchean and Proterozoic rocks in thesouthern District of Keewatin: Discussion:Canadian Journal of Earth Sciences, v. 12,p. 1250-1254.

Young, G.M., Long, D.G.F., Fedo, C.M., andNesbitt, H.W., 2001, PaleoproterozoicHuronian basin: product of a Wilson cyclepunctuated by glaciations and a meteoriteimpact: Sedimentary Geology, v. 141-142,p. 233-254.

Zegers, T.E. and van Keken, P.E., 2001,Middle Archean continent formation bycrustal delamination: Geology, v. 29,p. 1083-1086.

Accepted as revised 7 August 2004