Search

Precambrian high-grade metamorphic rocks (granulite facies, transitional granulite to amphibolite facies, and amphibolite facies) from the Amata area, Mus-grave Ranges, central Australia, display contrasting weighted average K/Rb ratios (628, 310, and 203 respectively). This pattern of higher K/Rb ratios in medium to high-pressure granulite facies rocks than in amphibolite facies rocks is in keeping with patterns found elsewhere in areas of metamorphosed Precambrian continental crust. The weighted average values are strongly influenced by the predominantly quartzo-feldspathic (acidic to sub-acidic) nature of the three metamorphic terrains. The ratios therefore reflect extensive control exerted by host-rock mineralogy on the distribution of K and Rb.The K/Rb ratios displayed by the quartzo-feldspathic and certain of the mafic rocks of the area are probably the result of crustal fractionation processes, whereby Rb is relatively depleted in the granulite facies rocks and relatively concentrated in the lower grade rocks. However, the K/Rb ratios in the mafic granulites may represent original igneous patterns, preserved as a result of isochemical metamorphism.

Anorogenic granites of middle to late Proterozoic age in the Davis Inlet — Flowers Bay area of Labrador are subdivided on the basis of petrology and geochemistry into three coeval suites. Two of these are high-temperature anhydrous hypersolvus granites: a peralkaline aegirine-sodic-calcic to sodic amphibole-bearing suite and a non-alkaline fayalite-pyroxene-bearing suite. The third is a group of non-alkaline subsolvus hornblende-biotite-bearing granites. Associated with the hypersolvus peralkaline suite is a group of genetically related syenites and quartz syenites. The granites cut ca. 3,000 Ma old Archaean gneisses as well as Elsonian layered basic intrusions of the Nain Complex. One of these, a crudely layered mass which ranges in composition from gabbro to diorite and monzonite, appears to be related to the syenites. The peralkaline granites and some of the syenites are extremely enriched in the high field-strength elements such as Y, Zr, Nd, as well as Rb, Ga and Zn, and have low abundances of Ba, Sr and most of the transition elements. In contrast, the non-alkaline hypersolvus and subsolvus granites do not show the same degree of enrichment. Concentration of the highly charged cations in the peralkaline suite is believed to be the result of halogen-rich fluid activity during fractionation of the magma. The sodic evolution trend in the peralkaline suite is reflected mineralogically by the development of aegirine and aegirine-hedenbergite solid solutions, and by a spectacular amphibole compositional range from katophorite through winchite, richterite, riebeckite to arfvedsonite and ferro eckermannite. Accessory phases which are ubiquitous in these rocks include aenigmatite, astrophyllite, fluorite, monazite and zircon. The non-alkaline hypersolvus granites typically contain iron-rich phases such as fayalite, eulite, ferrosilite-hedenbergite, and annite rich biotite. In the subsolvus granites, amphiboles range in composition from edenite through common hornblende to actinolite and also coexist with annite-rich biotite.

A petrogenetic model is developed to explain the evolution and geochemical character of granitic rocks in early Archean (pre 3.6 Gyr) continental crust taking into account the following important geological constraints, viz.:1.High geothermal gradients (probably in excess of 90 ° C/km) and resulting widespread granulite facies metamorphism even at relatively shallow depths2.The fractionation of certain major and trace elements under granulite facies conditions3.The composition and geochemical behaviour of fluids which emanate from or pass through terrains undergoing granulite facies metamorphism viz. carbonic fluids containing significant amounts of SO2 and halogens.

The 3,622±72 m.y. Uivak gneisses of northern Labrador consist of two major rock suites: regionally developed layered granodioritic gneisses (Uivak I suite) interpreted as derived from earlier tonalitic igneous parents by the massive introduction of potassium and rubidium, and a less extensive group of iron-rich porphyritic granodiorites and ferro-diorites (Uivak II suite) emplaced after at least one major period of deformation and migmatisation had affected the Uivak I gneisses. Samples from both suites fall on the 3,622 m.y. Rb/Sr whole-rock isochron.

Two groups of Archaean orthogneisses from the Western Gneiss Terrain, approximately 100 km east of Perth, Western Australia, give Nd model ages of 3150 Ma to 3240 Ma, and 2950 Ma to 3050 Ma respectively. A Sm-Nd whole-rock isochron of 3210 ± 190 Ma (2σ), obtained by including a mafic granulite with samples having the older group of model ages, is in close agreement with U-Pb zircon ages of Nieuwland & Compston (1981). This agreement, together with a chondritic initial 143Nd/144Nd, indicates a limited prior crustal residence time (<200 Ma) for the gneiss protoliths. Leucosome and melanosome components of nearby migmatite yield significantly younger Nd model ages of 2860 Ma and 2660 Ma, respectively. Post-tectonic granites also give younger model ages of 2770 Ma and 2680 Ma, consistent with the addition of new material to the continental crust at this time. When contrasted with Nd model ages of 3630 Ma and 3510 Ma from the northern extension (de Laeter et al., 1981) of the Western Gneiss Terrain, these results indicate that this is a composite belt with an igneous and metamorphic history spanning almost 1000 Ma. Comparison with isotopic results for gneisses from the eastern margin of the Yilgarn Block indicates that the Western Gneiss Terrain is fundamentally older, suggesting a dominant pattern of crustal growth from west to east.