mid-Proterozoic Time Research Articles

On the enigmatic mid-Proterozoic: Single-lid versus plate tectonics

The mid-Proterozoic (ca. 1850–850 Ma) is a peculiar period of Earth history in many respects: ophiolites and passive margins of this age are rare, whereas anorthosite and A-type granite suites are abundant; metamorphic rocks typically record high thermobaric (temperature/pressure) ratios, whereas ultrahigh pressure (UHP) rocks are rare; and the abundance of economic mineral deposits features rare porphyry Cu-Au and abundant Ni-Cu and Fe-oxide Cu-Ag (IOCG) deposit types. These collective observations have been used to propose that a stagnant-lid, or single-lid, tectonic regime operated at this time, between periods of plate tectonics in the Paleoproterozoic and Neoproterozoic. In our reappraisal of the mid-Proterozoic geological record, we not only assess the viability of the single-lid hypothesis for each line of evidence, but also that of the plate tectonic alternative. We find that evidence for the single-lid hypothesis is equivocal in all cases, whereas for plate tectonics the evidence is equivocal or supporting. We therefore find no reason to abandon a plate tectonic model for the mid-Proterozoic time period. Instead, we propose that the peculiarities of this enigmatic interval can be reconciled through the combination of two processes working in tandem: secular mantle cooling and the exceptionally long tenure and incomplete breakup of Earth's first supercontinent, where both of these phenomena had a dramatic effect on lithospheric behaviour and its resulting imprint in the geological record.

Spectral characteristics of banded iron formations in Singhbhum craton, eastern India: Implications for hematite deposits on Mars

Banded iron formations (BIFs) are major rock units having hematite layers intermittent with silica rich layers and formed by sedimentary processes during late Archean to mid Proterozoic time. In terrestrial environment, hematite deposits are mainly found associated with banded iron formations. The BIFs in Lake Superior (Canada) and Carajas (Brazil) have been studied by planetary scientists to trace the evolution of hematite deposits on Mars. Hematite deposits are extensively identified in Meridiani region on Mars. Many hypotheses have been proposed to decipher the mechanism for the formation of these deposits. On the basis of geomorphological and mineralogical studies, aqueous environment of deposition is found to be the most supportive mechanism for its secondary iron rich deposits. In the present study, we examined the spectral characteristics of banded iron formations of Joda and Daitari located in Singhbhum craton in eastern India to check its potentiality as an analog to the aqueous/marine environment on Mars. The prominent banding feature of banded iron formations is in the range of few millimeters to few centimeters in thickness. Fe rich bands are darker (gray) in color compared to the light reddish jaspilitic chert bands. Thin quartz veins (<4 mm) are occasionally observed in the hand-specimens of banded iron formations. Spectral investigations have been conducted in VIS/NIR region of electromagnetic spectrum in the laboratory conditions. Optimum absorption bands identified include 0.65, 0.86, 1.4 and 1.9 μm, in which 0.56 and 0.86 μm absorption bands are due to ferric iron and 1.4 and 1.9 μm bands are due to OH/H2O. To validate the mineralogical results obtained from VIS/NIR spectral radiometry, laser Raman and Fourier transform infrared spectroscopic techniques were utilized and the results were found to be similar. Goethite-hematite association in banded iron formation in Singhbhum craton suggests dehydration activity, which has altered the primary iron oxide phases into the secondary iron oxide phases. The optimum bands identified for the minerals using various spectroscopic techniques can be used as reference for similar mineral deposits on any remote area on Earth or on other hydrated planetary surfaces like Mars.

Major geological cycles substantiated by U–Pb ages and εHfi of detrital zircon grains from the Lower Rhine Basin

Detrital zircon grains of the Cenozoic Lower Rhine Basin were dated by the U–Pb method and simultaneously analyzed for initial Hf signatures to (1) identify distinct geodynamic events recorded in the W-European basement and cover rocks and (2) estimate the addition of juvenile crust in relation to these. Approximately 350 zircon grains extracted from 4 Miocene to Quaternary sand samples were investigated by back-scattered electron (BSE) and cathodoluminescence (CL) imaging. One third of these zircons were found to be sufficiently homogeneous for laser ablation U–Pb dating, which was performed by two individual and spatially well separated laser-spot analyses on the same grain, substantiating zircons have identical ages for the two intra-grain analyses. Approximately 40 zircons showed different ages. Our U–Pb dating results document distinct periods of primary magmatic zircon growth between 2.7 and 0.03Ga. These ages correspond to worldwide known subduction–collision cycles. The oldest terrains exposed today along the Rhine River and the Alps belong to the Variscan cycle. Therefore, all zircon grains older than ca. 0.4Ga must have been extracted from Paleozoic or Mesozoic–Cenozoic cover rocks containing detrital grains inherited from pre-Variscan basement; only the Cenozoic zircons require direct transport to the basin.From the dated homogeneous zircons 34 grains were selected for Hf isotope analysis, showing large variations in initial Hf isotope ratios (εHfi) between −10.9 and +15.2. Twelve of the 13 Precambrian grains yield positive or zero εHfi values, with a single 1025-Ma old grain having a negative value of −10.9. For the 21 zircon grains younger than 0.6Ga, a series of 9 grains yield positive or zero εHfi, whereas the remaining 12 zircons have negative values as low as −5.8. Approximately 60% of the zircon crystals reflect addition of juvenile crust having positive or zero εHfi being extracted from the mantle around the time indicated by their U–Pb ages. For some cases, juvenile crust addition may be somewhat older than indicated by the U–Pb ages because zircon growth may occur during later differentiation of rocks originally devoid of zircon. For such cases, εHfi values are necessarily less positive than that of the juvenile magma, depending on the Lu/Hf and crustal residence times of the zircon-free rock. A few εHfi values between +11.4 and +15.2 corroborate the existence of highly fractionated, Lu-enriched and Hf-depleted asthenospheric mantle in Mid-Proterozoic times. In agreement with earlier detrital zircon studies, our new data demonstrate that crustal recycling during continent collision events significantly increased in Phanerozoic times, being relatively rare during Precambrian times where intra-plate plume activity and oceanic arc attachment were dominant.

Detrital zircon provenance of Cambrian–Ordovician and Carboniferous strata of the Oaxaca terrane, southern Mexico

U–Pb geochronologic analyses have been conducted on 135 detrital zircon grains from Paleozoic strata of the Oaxaca terrane of southern Mexico. The grains are mainly mid-Proterozoic in age (age probability peak of 993 Ma), with subordinate clusters of ∼358 and ∼472 Ma. The mid-Proterozoic ages can be used to evaluate three possible paleopositions for the Oaxaca terrane during mid-Proterozoic time. Formation in proximity to northwest South America appears most likely, as southwestern portions of the Amazon craton (and perhaps basement massifs in the northern Andes) contain igneous rocks of the appropriate age. A more specific test of ties to these regions is not yet possible due to the limited number of U–Pb determinations from the region. Formation in proximity to Grenville-age rocks in northeast North America is less likely, as detrital zircons shed from these rocks are mostly older than 993 Ma. Previously proposed links with Grenville-age rocks in southwestern North America are not supported by our data, as there is little overlap of ages from the Oaxaca terrane and from Grenville-derived detrital zircons in southwestern US and northwestern Mexico. The presence of Paleozoic grains in our samples does not help constrain the Late Proterozoic–early Paleozoic displacement of the Oaxaca terrane, as magmatism of this age was apparently widespread within and adjacent to the Iapetus and Rheic ocean basins.

Evidences of the presence of old continental basement in Cheju volcanic Island, South Korea, revealed by radiometric ages and Nd-Sr isotopes of granitic rocks.

Major and trace element and Nd-Sr isotopic data are presented on the granitic rocks known as a basement in the Cheju volcanic island, together with CHIME age data on porphyritic hornblende-biotite granite xenolith from the Byeoldobong area. The CHIME ages from zircon and allanite grains of porphyritic hornblende-biotite granite were determined to be 169 ± 29 Ma and 183 ± 40 Ma, respectively, which are corresponding to a Jurassic age. It is concordant with a K-Ar age of 172.4 ± 5.2 Ma of a biotite granite in the same area, which is comparable to the Daebo granite in the Korean peninsula. Granitic xenoliths from the Byeoldobong area have Nd-Sr isotopic compositions (initial ratio (143Nd/144Nd)i = 0.51156∼0.51158, eNd = −19∼−14 and (87Sr/86Sr)i = 0.7156∼0.7164, eSr = 101∼179) which are close to those of Jurassic Daebo granite (eNd = −21∼−14, eSr = 104∼171). On the other hand, the micrographic granite from the drilling core in the Hadeokcheon area gives initial ratio (143Nd/144Nd)i = 0.51218 and (87Sr/86Sr)i = 0.7077), which are comparable to the Cretaceous Bulguksa granite ranging from 0.51184 to 0.51268 in 143Nd/144Nd and 0.7055 to 0.7167 in 87Sr/86Sr. The TNdDM model ages of 1.0 to 1.7 Ga from the Mesozoic granitic plutons in the Cheju island suggest that the source materials of the plutons in the island have evolved since mid-Proterozoic time as those of the peninsula. In combination with the basement geology and petrochemistry, Nd-Sr isotopes, and radiometric and model ages of granitic rocks, our results imply that the Okchon zone is continuously extended to the Cheju volcanic Island of the southeastern end of the continental margin of the Eurasian plate.

Seismic evidence for a mantle source for mid-Proterozoic anorthosites and implications for models of crustal growth

Abstract Voluminous anorthosite intrusions are common in mid-Proterozoic crust. Historically, two end-member models have been proposed for the origin of these anorthosites. In the first model anorthosites derive from fractionation of a mantle source leaving a residue of metagabbro in the lower crust; in the second model anorthosites are the product of partial melting of the lower crust with residual pyroxene and high-grade minerals (i.e. a pyroxenitic and/or metapelitic lower crust). Although a general consensus has developed that the first model provides the best fit to petrological and geochemical constraints, the sparse evidence for mafic and ultramafic counterparts to the anorthosites leaves the issue still unresolved. We use the absolute P-wave velocity and the ratio between P- and S-wave velocities ( V P / V S ) to infer the composition of the lower crust beneath the Marcy Anorthosite (New York State, USA). Seismic refraction data reveal a lower crust 20 km thick, where V P and V P /V S range from top to bottom between 7.0 km s −1 and 7.2 ± 0.1 and 1.84 km s −1 and 1.81 ± 0.02, respectively. Laboratory measurements on rock samples indicate that these seismic properties are typical of plagioclase-rich rocks. Magmatic underplating of basaltic melts is a mechanism to form plagioclase-rich bulk composition for the Grenville crust. At the bottom of the lower crust, increase of P-wave velocity, slight decrease of V P / V S ratios and the presence of a low-reflective seismic Moho are additional observations supporting crustmantle interactions related to magmatic underplating. High P-wave velocity (8.6 km s −1 ) in the upper mantle may indicate that the ultramafic portion (e.g. pyroxenites) of the underplated magma has become eclogite. High average P-wave velocity (6.7 km s −1 ) and V P / V S (1.81), and the exceptional abundance of anorthosites-norites-troctolites among the rocks exposed at the surface, indicate that the Grenville Proterozoic crust may have a unique plagioclase-rich bulk composition. We suggest magmatic underplating, occurring either over a wide time span or with separate syn- and post-collisional magmatic pulses, as being a major crust-forming mechanism operating in mid-Proterozoic time.

Igneous and metamorphic petrology of the eclogitic Seljeneset Meta-anorthosite and related jotunites, Western Gneiss Region, Norway

The Seljeneset Meta-anorthosite consists of deformed and metamorphosed anorthosite and mafic rocks, is situated in the high-grade Western Gneiss Region (WGR) of Norway, and is adjacent to the 1.52 Ga Flatraket quartz monzonite (mangerite). The WGR was metamorphosed to eclogite fades during Caledonian tectonism 414 ± 16 Ma ago. The Meta-anorthosite contains 46–57 wt.% SiO 2 and displays chondrite-normalized REE patterns typical of anorthosite, indicating that metamorphism did not significantly affect REE abundances. Norms indicate that igneous anorthosite consisted of ≥ 90% modal plagioclase, minor mafic silicates, and ~ 1% Fe-Ti oxide. Mafic rocks (43–48 wt.% SiO 2) are rich in total Fe 2O 3 (13–20%), TiO 2 (3–4%), and P 2O 5 (1–3%), poor in Cr (< 4 ppm) and Ni (~ 30 ppm), and are thus classified as jotunites, which are characteristic of Proterozoic anorthosite-mangerite complexes. Jotunite originally consisted of > 50% modal andesine, ∼ 25% mafic silicates, ∼ 10% Fe-Ti oxide, and up to 7% apatite. REE patterns of jotunite are highly variable, which reflects variable igneous abundances and modification by metasomatism. Filter pressing of residual liquid trapped between cumulus plagioclase may have been important in the evolution of anorthosite and jotunite. Anorthosite contains ultramafic xenoliths (high MgO, Cr and Ni and depleted in LREE), which suggests that some mantle-derived peridotite was present in the Baltic crust by mid-Proterozoic time; this may extend the crustal history of Norwegian garnet peridotites by over one billion years. Seljeneset rocks display evidence of retrogression from eclogite through amphibolite facies. Meta-anorthosite consists of ovoidal clots of margarite, quartz, plagioclase, and relict kyanite set in a matrix of zoisite, clinozoisite, plagioclase, quartz, hornblende, and biotite. Phase equilibria analysis of sequential assemblages shows that meta-anorthosite records a retrograde P-T path from ≥ 15–17 kbar to 4.3–8.2 kbar at ~ 450–600 °C. Mafic rocks vary from moderately well-preserved eclogite, through varieties containing symplectite, to dynamically recrystallized amphibolite. In one sample, the rare Na-Ca amphibole magnesio-alumino-taramite appears to be part of the eclogite assemblage. Calculations indicate eclogite formation at P ≥ 16–18 kbar at ~ 750 °C. Some mafic rocks contain veins and segregations of biotite, muscovite, pyrite, allanite, and zircon, that, considered together with geochemical data, indicate metasomatic introduction of K, Rb, Ba, Ga, S and rare-earth and other high-field-strength elements by channelized fluids during retrogression. The Seljeneset Meta-anorthosite, associated jotunites, and adjacent Flatraket quartz monzonite likely constitute a single complex, here named the Nordpollen Plutonic Complex. This complex was emplaced 1.52 Ga ago, metamorphosed to eclogite facies during Silurian continent-continent collision, largely retrograded to amphibolite facies during uplift of the orogen, and locally metasomatised during retrogression.

Constraints on temperature–pressure conditions and fluid composition during metamorphism of the Sullivan orebody, Kimberley, British Columbia, from silicate–carbonate equilibria

The Sullivan mine, in southeastern British Columbia, is one of the world's largest sediment-hosted, massive sulphide deposits. It has undergone at least one period of metamorphism since it was deposited in mid-Proterozoic times. Mineral textures within the deposit are predominantly of metamorphic origin. A well-constrained estimate of metamorphic conditions is required to understand how the original, depositional character of the orebody has been modified by metamorphism. Metamorphic conditions were estimated using multiequilibrium thermobarometric techniques involving silicate–carbonate–fluid equilibria. Peak metamorphic temperature constrained by calibration of the garnet–biotite Fe–Mg exchange equilibrium is 450 ± 50 °C. Lower temperature estimates from some samples are interpreted to record the temperature of cessation of garnet growth prior to the attainment of peak metamorphic temperature. Peak metamorphic pressure as determined from equilibria applicable to the assemblage garnet–biotite–muscovite–chlorite–calcite–quartz–fluid is 380 ± 100 MPa. The fluid composition accompanying this pressure estimate is [Formula: see text], [Formula: see text]. This estimate is particular to one sample and may not be representative for the deposit as a whole. Metamorphic fluids at the estimated P–T conditions would not have contained significant concentrations of C–O–H–S species other than H2O and CO2. Textural evidence and temperature–pressure results from a titanite-bearing metamorphosed mafic intrusion in the deposit suggest published titanite ages near 1330 Ma in the area of the mine represent the age of the peak metamorphic event. The results of this study carry tectonic implications for the Sullivan area, and may have application to other metamorphosed ore deposits and low-grade metamorphic settings.

Lead isotope analyses of leached feldspars: Constraints on the early crustal history of the Grenville Orogen

Lead isotopic data on leached K-feldspars, plagioclases, and muscovite provide constraints on the premetamorphic evolution of the Grenville Orogen in Ontario and New York. A boundary between reworked Archean crust to the northwest and new mantle-derived crust generated during mid-Proterozoic times to the southeast is identified within the northern portion of the Gneiss Belt of Ontario. The identification of juvenile mid-Proterozoic crust at the southeastern margin of the Orogen is consistent with previous investigations based on neodymium model ages. However, near the Grenville Front lead isotopes indicate the presence of reworked Archean crust rather than primary early Proterozoic crust postulated on the basis of neodymium model ages. This discrepancy between the lead and neodymium “isotopic terranes” underscores the necessity of using a multidisciplinary approach to identify possible terrane boundaries. Only through a combination of the lead with neodymium isotopes is it possible to distinguish between crust that formed during a single major orogenic cycle and crust that represents a mixture of old with new mantle-derived material. The combined neodymium and lead data imply that the Grenville Orogen in Ontario consists essentially of reworked Archean crust and juvenile mantle derived material generated around 1.5 Ga. The Grenville Orogen involved Archean crust that was depleted in thorium and uranium several hundred Ma before high grade metamorphism at ca. 1.1 Ga. Such a depletion may have a significant impact on the geotherm achieved during orogeny and may explain the lower metamorphic thermal gradients observed for the Gneiss Belt.

Basement–cover relationships in the East Greenland Caledonides: evidence from the Eleonore Bay Supergroup at Ardencaple Fjord

AbstractThe Eleonore Bay Supergroup (EBG) is a 16 km-thick shallow-water sequence of Neoproterozoic age that is preserved within the East Greenland Caledonides in several tracts, surrounded by crystalline gneisses and schistose supracrustal rocks. The apparent downward transition from non-metamorphic EBG into gneiss gave rise to the classic ‘stockwerke’ hypothesis, in which all the metamorphism was regarded as Caledonian, and differences in grade were ascribed to the ascent of a migmatite front to different levels within the orogen. Field and isotopic studies in the 1970s however revealed that the underlying gneisses and schists had undergone orogenic reworking in mid-Proterozoic time; the EBG–basement contact was then interpreted as an approximately bedding-parallel décollement with apparent lag geometry, that is with EBG cover rocks in its hangingwall.Recent work in the northernmost EBG tract, at Ardencaple Fjord, has shed light on the problems posed by the basal relationships of the EBG, and together with regional structural and stratigraphic data leads to the following interpretation. There are two regionally important basement-cover interfaces within the East Greenland Caledonides. The earlier one is between Archaean/early Proterozoic gneisses and early Proterozoic supracrustal rocks, which were pervasively deformed in mid-Proterozoic time and form the basement to the Neoproterozoic Eleonore Bay cover sequence. This was deposited on a vast, continually subsiding shelf that is now preserved in East and NE Greenland and Svalbard, and contains Grenville detritus. EBG deposition was terminated by major extensional faulting of Vendian age; the succeeding Tillite Group is interpreted as a syn-rift sequence, presumably associated with the opening of Iapetus.The EBG–basement contacts that are not late faults are inferred to be extensional shear zones of Vendian age. These were reactivated in compression during the Caledonian orogeny in the Silurian, with metamorphic and fabric convergence, which accounts for the apparent downward transition from sedimentary rocks through schists into gneisses. Caledonian shortening was not large; inversion of the Vendian grabens was incomplete, so that the marginal shear zones retained their lag geometry and large tracts of low grade Eleonore Bay sediments are preserved at the present erosion level, surrounded by Proterozoic basement rocks, within the Caledonian belt of East Greenland.

Un exemple de métasomatisme alcalin albite–uranium dans le bassin des Monts Otish, Québec

Metallographic studies of uraniferous occurrences in the Otish basin (Quebec) have shown that these mineralizations are characterized by the association albite–uranium (successively, brannerite–uraninite). This paragenesis corresponds to a sodium metasomatic event related to Hudsonian (1730 ± 30 Ma) gabbroic sills and dykes that intrude both Archean granitic basement and the overlying Aphebian detritic cover. The mineralization (1723 ± 3 Ma) is interpreted as a late Hudsonian event and is in many ways comparable to other occurrences in the Precambrian shields of Brazil, Scandinavia, and the Ukraine. A comparison of these different occurrences with the Otish mineralization has resulted in the recognition of several characteristics common to late-orogenic metallogenic events from mid-Proterozoic time in Canada and elsewhere up to late Hercynien in Europe. To explain specific assemblages of tellurides, selenides, and sulphides within a dolomitic gangue, one suggests that deep basinal Na-rich brines were mobilized to form mineralizing fluids during a regional tectono-magmatic event. The pitchblende–calcite association in the Otish basin corresponds to the latest mineralized event (306 Ma) caused by a remobilization of the primary Na–U metallogeny.

U-Pb zircon ages for granitoid gneisses in northern Namibia and their significance for proterozoic crustal evolution of southwestern Africa

The pre-Damara (> 900 Ma) basement gneisses and granitoids of northern Namibia and southern Angola constitute the southern margin of the Congo craton and are poorly known geochronologically. We report U-Pb zircon concordia intercept ages for an augengneiss of the Epupa Complex (∼ 1800 Ma), a banded quartzofeldspathic gneiss of the Huab Complex (∼1810 Ma), a post-Huab granitic gneiss (∼1750 Ma) and for inherited zircons (∼2120 Ma) from a Damaran syenite that has a true emplacement age of 800–840 Ma. These ages suggest that the Epupa and Huab Complexes are broadly coeval, and their formation, together with the emplacement of the widespread Fransfontein granitoid suite, constitutes a significant and regionally extensive crust-forming event in northern Namibia and southern Angola (the Huabian Episode) that corresponds to the Eburnian event in western, central and eastern Africa. Older crust in excess of 2120 Ma is likely to be present in the region but has not been documented geochronologically. Similarities in age, rock association and broad geologic evolution for pre-Damara basement complexes in northern, central and southern Namibia suggest that the entire territory may have been part of the Congo craton in mid-Proterozoic times, whose southwestern portion was largely reworked during the late Precambrian Namaqua and Pan-African events.

Crustal growth and late Precambrian-early Palaeozoic plate tectonic evolution of England and Wales

Comparison of the basement rocks of Britain reveals major contrasts to the N and S of the Iapetus suture (Solway-Shannon line). North of the suture, polymetamorphic high-grade basement schists and gneisses and igneous rocks date from about 2900 Ma to c. 400 Ma, the latter age corresponding to closure of the Iapetus ocean. By contrast, S of the suture, in England and Wales, the basement is dominantly low grade metamorphic rock, together with intrusive and extrusive igneous, and sedimentary rocks. There is little geological or geochemical evidence for the presence of basement older than c. 900 Ma below this area and the predominant expression of crustal growth in England and Wales is indicated by radiometric ages between c. 700 Ma and 450 Ma, reflecting late Precambrian-Palaeozoic magmatism and metamorphism. Palaeomagnetic data from Scotland demonstrate that the crust in northern Britain developed as an integral part of the Laurentian Shield during mid-Proterozoic times (1600–1000 Ma). The palaeomagnetic record from the pre-Ordovician rocks of England and Wales shows that this area was unrelated to the Laurentian Shield prior to closure of the Iapetus Ocean. The apparent polar wander (a.p.w.) path is that of a distinct microplate involved in rapid growth of Iapetus during early Cambrian times, and emplaced in its present configuration adjacent to northern Britain as a result of subduction of the ocean basin between Cambrian and Silurian times. We conclude that whereas the basement to the north of the Iapetus suture formed an essentially continental unit from 2900 to 400 Ma, the basement of England and Wales is a result of crustal growth from about 900 Ma to 400 Ma and formed by accretion of island arcs, associated accretionary prisms and fore-arc basin sediments within the Iapetus Ocean.

Space and time in granite plutonism

Granitic suites formed in the last 2500 Ma are either calc-alkaline or, less commonly, alkalic. The space-time trends of granitoid rocks from modern compressional arcs include variations towards more silicic and potassic products. But even the most ‘mature’ magmas of these arcs have Fe/Mg ratios near to 1, in contrast with the strongly iron-enriched alkalic suites that characterize zones of crustal extension. Both calc-alkaline and alkalic suites are mainly mantle-derived; they evolved, respectively, from tholeiitic and alkali basalt parental magmas which have been subject, in most examples, to a two-stage fractionation history. Other factors that may influence the evolution of these parental magmas include, first, the addition of variable, but minor, amounts of crustal melt and, secondly, the persistence of refractory mineral phases in the zone of melting. Contributions from subducted continental detritus are thought to be of minor significance. A long-term trend, over the entire history of the crust from more sodic to more potassic calc-alkaline magmatism may reflect the declining importance of subducted ocean lithosphere in magma generation and the increasing role of the overlying volatile-fluxed mantle wedge. This trend and, in particular, the widespread development of alkalic granite suites in mid-Proterozoic times, may be linked to the declining vigour of tectonic and associated magmatic processes during the Earth’s history. These alkalic suites may indicate a period when the continental lithosphere had become stable enough to resist fragmentation, leading to intracontinental ‘rift and swell’ zones of magmatism. In some cases, new plate cycles, with more normal calcalkaline igneous rocks, seem to have occurred in the same zones. Although much of the evidence favours the operation of plate tectonic and subduction processes in magma generation during Archaean and early Proterozoic times, this link cannot be taken as proved except for the last ca . 1000 Ma.