Many auriferous veins dated at ∼370 Ma in the Meguma terrane of Nova Scotia are concentrated in tight domes, and conical fold hinges that are transected by 378 Ma low-pressure and high-temperature metamorphic isograds and 380–370 Ma granitoid plutons. Published experiments suggest that such folds are diachronous initiating at inhomogeneities and propagating both in amplitude and along their length. Thus fold terminations are typically conical in geometry and record only the youngest increment of folding. The northeastern termination of the Oldham anticline is characterized by (1) a conical fold, (2) saddle reef auriferous veins, (3) divergent and convergent cleavage fans, (4) a downdip lineation on the slaty cleavage defined by biotitic mineral aggregates inferred to have developed during fold growth. The age of the biotitic mineral aggregates is bracketed between peak metamorphism at > 600°C, defined by a nearly concordant 378 Ma U–Pb monazite age from a xenolith in lamprophyre dyke, and ∼366 Ma 40Ar/39Ar muscovite–biotite ages, recording cooling through ∼400–300°C. This suggests that the conical termination of the Oldham anticline grew between 378 and 366 Ma, an observation that reconciles the empirical structural control of the saddle reef auriferous veins with the ∼370 Ma age dating of vein minerals. Application of this conclusion to saddle reef auriferous veins in domes suggests that mineralization was related to the youngest increment of fold amplification.
Auriferous veins of diverse ages are common in turbidite sequences throughout the world, and although a wide variety of origins (igneous, hydrothermal, syngenetic, and metamorphic) have been proposed, Boyle (1986) suggested that metamorphic secretion is the most probable. The gold-bearing veins in the Meguma Terrane of Nova Scotia (Atlantic Canada) represent an example of such turbidite-hosted deposits, for which the full range of genetic models have been proposed, and so provide an ideal location for evaluating these models. Hypotheses for the genesis of the gold mineralization include syngenetic-exhalative, emplacement during deformation, epigenetic related to granitoid plutonism, and hydrothermal veins emplaced along late shear zones. Thus, Haynes (1986) proposed that the bedding-parallel auriferous veins formed from syngenetic geyserites as the Cambro-Ordovician turbiditic sediments were being deposited. On the other hand, others (Faribault 1899; Malcolm 1929; Keppie 1976; Graves and Zentilli 1982; Mawer 1986, 1987; Henderson and Henderson 1986) have suggested that the auriferous veins formed by a crack-seal mechanism during Devonian folding with veins generally concentrated in anticlinal saddle reefs. Some other authors (e.g., Newhouse 1936; Douglas 1948; MacDonald and O’Reilly 1989) have suggested that the gold mineralization might be connected with the ∼380–370 Ma granitoid plutonism. Recent 40Ar/39Ar dating of vein muscovite, biotite, and amphibole have yielded ∼370 Ma ages leading to the hypothesis that the gold mineralization was synchronous with the magmatism and with subvertical shear zones (Kontak et al. 1990, 1998, and references therein). Isotopic data suggest that the fluids were derived from a mixed source including the host turbidites, middle–lower crust and mantle, and possibly granitoid magmas (Kontak et al. 1990). Keppie and Krogh (1999) and Murphy et al. (1999) have proposed that the passage of a mantle plume beneath the Meguma terrane was responsible for Late Devonian release of hydrothermal fluids associated with low-pressure metamorphism; voluminous plutonism, including intrusion of lamprophyres; and the gold mineralization.
Although the syngenetic model appears untenable because bedding-parallel veins commonly crosscut bedding or bifurcate (Mawer 1986), the empirical observation that auriferous veins are preferentially developed in anticlinal fold hinges appears to conflict with the ∼370 Ma age data because the ∼378 Ma metamorphic isograds do not appear to be deformed by the fold hinges, and the latter are truncated by the ∼380–370 Ma granitoid plutons (Fig. 1). This apparent contradiction may possibly be explained in terms of the experiments performed by Dubey and Cobbold (1977), which have shown that fold generation is diachronous with folds propagating in amplitude and along their length. Thus, the youngest increment of folding may be most easily distinguished from earlier stages of fold growth at the fold termination. To test this solution to the apparent contradiction, we selected for study a conical fold termination with saddle reef auriferous veins, the Oldham Gold District. We conclude that although fold generation probably started in the Early Devonian, the conical termination of the Oldham Anticline developed in the Late Devonian, and that the auriferous mineralizing fluids probably migrated with fold propagation.
The Meguma terrane is the farthest outboard terrane in the northern Appalachian Orogen (Fig. 1). On its northern side, it is bordered by Minas Fault system that separates it from the Avalon terrane. The Meguma terrane is characterized by > 13 km of Cambrian–Ordovician turbidites of the Meguma Group (sandy Goldenville Formation, > 7 km thick, overlain by the slaty Halifax Formation, 6 km thick) overlain by 2.3–4.5 km thick, Silurian and Lower Devonian (Lockhovian – lower Emsian) shallow marine – continental deposits (Schenk 1970). These rocks have been interpreted as a passive margin sequence bordering either northwest Africa (Schenk 1981) or the Avalon terrane (Keppie and Krogh 2000). Although the basement beneath these Paleozoic rocks is not exposed, the chemistry of the ∼440 Ma bimodal volcanic rocks indicates the presence of a continental basement (Keppie et al. 1997).
These rocks were then deformed by generally upright, subhorizontal, northeast-trending folds associated with the development of fracture–slaty cleavage under low grade of metamorphism. This tectonothermal event is dated by 40Ar/39Ar whole-rock dating of phyllites and muscovite that yielded plateau ages ranging from ∼400 to 377 Ma, which are substantially younger than the detrital ages (Reynolds and Muecke 1978; Keppie and Dallmeyer 1987, 1995; Kontak et al. 1998; Hicks et al. 1999).
This was followed by several generally coeval events (no sequence implied): (1) widespread, generally static, greenschist– amphibolite facies, low-pressure metamorphism; (2) development of a local subhorizontal crenulation cleavage synchronous with the initial phases of the low-pressure metamorphism (Keppie and Dallmeyer 1987); (3) intrusion of lamphrophyre dykes closely dated by 370–367 ± 2 Ma 40Ar/39Ar hornblende plateau ages (Kempster et al. 1989); (4) intrusion of the Liscomb gneiss dome with 40Ar/39Ar cooling ages of ∼370 Ma (40Ar/39Ar mineral plateau ages: Kontak and Reynolds 1994); (5) development of hydrothermal veins (quartz-carbonate, plagioclase, biotite, muscovite, garnet, amphibole, apatite, chlorite, tourmaline, andalusite, epidote, K-feldspar, ilmenite, rutile, sphene, zircon, pyrite, pyrrhotite, galena, sphalerite, arsenopyrite, chalcopyrite, scheelite, molybdenite, sulphides, and native gold) closely dated by 40Ar/39Ar plateau ages on mica and amphibole recording cooling at ∼370 ± 8 Ma (Kontak et al. 1990, 1998): as some of these dated veins occur in chlorite-grade host rocks, their ages cannot have been reset by peak metamorphism and (or) granitoid emplacement; (6) intrusion of voluminous granitoid plutons at 380–370 Ma (concordant U–Pb zircon and monazite ages: Keppie and Krogh 1999 and references therein); (7) development of east-northeast-trending, subvertical, dextral shear zones with 40Ar/39Ar ages on syntectonic micas that record cooling through ∼300°C by ∼370–360 Ma in the northern and eastern Meguma terrane and between ∼350 and 345 Ma in the southwestern Meguma terrane (Reynolds and Muecke 1978; Keppie and Dallmeyer 1987, 1995); (8) northeast-trending, steeply dipping, ductile, dip-slip shear zones active during contact metamorphism with the downthrown side generally being away from the adjacent granite plutons (Williams and Hy 1990).
The relationship between the low-pressure and hightemperature metamorphism and the plutons is complex. For example, the metamorphic isograds are parallel to the contacts of the Barrington Passage pluton (dated at 372 ± 3 Ma: U–Pb concordant monazite analyses: Keppie and Krogh 1999), whereas they are truncated by the South Mountain batholith (dated at Ma: nearly concordant, U–Pb monazite analysis: Keppie et al. 1985). This suggests that the lowpressure metamorphism was nearly synchronous with intrusion of the granitoid plutons. This is consistent with the nearly concordant 378 ± 1 Ma U–Pb monazite ages recorded in a pelitic granulite xenolith in a lamprophyric dyke cutting the Meguma terrane, which were inferred to date this high-grade metamorphic event (Greenough et al. 1999).
These events were followed by 5–12 km rapid denudation before deposition of continental – shallow marine Late Devonian – Carboniferous rocks, the oldest of which are late Fammenian (∼365 Ma) along the northern margin of the Meguma terrane and ∼345 Ma along the southern margin of the Meguma terrane (Martel et al. 1993; Keppie and Dallmeyer 1995; Tucker et al. 1998). These rocks were deformed by generally northeast-trending folds associated with slides formed under sub-greenschists facies conditions and preferentially located in Visean evaporite horizons (Giles and Lynch 1993). These late Paleozoic units are unconformably overlain by early Mesozoic rocks along the northwestern margin of the Meguma terrane.
The Oldham anticline occurs in the eastern Meguma terrane. It lies between the biotite and garnet isograds and is ∼25 km in length, extending from Shubenacadie Lake to 3 km northeast of the town of Oldham (Figs. 1, 2). It represents one of the northeast-trending folds that affect the Meguma Group: neighbouring folds are truncated by the basal Carboniferous unconformity. The Oldham anticline was mapped at a 1: 63 360 scale by Faribault (1908), who showed a syncline on its northern side that terminates at the same location as the anticline. Keppie (1976) analyzed bedding data from Faribault’s 1: 63 360 map sheets and determined that the Oldham anticline terminates in two conical folds: a 45° cone in the northeast and a 20° cone in the southwest. The Oldham Gold District at the northeastern end of the Oldham anticline was mapped in detail by Faribault (1898) at a 1: 6000 scale during active mining when outcrop was readily available. This latter map shows the location of the bedding-parallel veins that were mined for gold (Fig. 2). It also shows a major fault with several convergent splays in the hinge zone of the anticline, which brecciated and displaced the veins (Fig. 2).
Planar and linear structures: field description
Primary structures in the Goldenville Formation include bedding that is generally defined by thick sandy turbidite layers that grade upwards into slaty horizons. The bottoms of the sandy layers are generally sharply defined and may show a variety of bottom structures (Schenk 1970). The top of the sandy layers may show dewatering structures, such as sand volcanoes fed by a central pipe and dewatering sheets (Pickerill and Harris 1979).
At Oldham, these rocks are folded into a major anticline and syncline (Fig. 2): no minor folds were observed. A spaced cleavage occurs in the psammitic layers and this grades into a slaty cleavage in the pelitic layers with the development of cleavage refraction in the limbs of the folds. In the fold limbs, some psammites contain two or three spaced cleavages, with the younger cleavage(s) generally dipping more steeply than the older cleavage. Furthermore, sigmoidal spaced cleavage was locally recorded in psammites in the fold limbs. A prominent lineation is defined by the intersection of bedding and cleavage. In addition, a downdip mineral lineation defined by aggregates that include biotite is present on the slaty cleavage planes. Both sinistral and dextral kink bands deform the cleavage.
Bedding-parallel auriferous quartz veins, occurring mainly in the slate horizons, occasionally cut obliquely across the bedding. These veins are folded in the hinge region of the Oldham anticline and their geometry varies from cylindrical to discontinuous and non-cylindrical. A ridge–valley or groove lineation subparallel to the minor fold axes is present on the vein margins. The cleavage is refracted across these veins. A series of en echelon veins may cut across or merge into the concordant veins. Highly discordant quartz veins are also present in both psammites and pelites.
Bedding, cleavage, bedding-cleavage intersection lineation, and mineral lineation
Remapping of the northeastern end of the Oldham anticline shows that it is an upright, northeast-trending periclinal structure with an axial plane oriented at 056°/87°NW and a curvilinear hinge, which varies from 236°/20°SW in the southwest to 056°/36°NE in the northeast (Fig 2). Measurements along the #102 Highway show that the central part of the fold is cylindrical with bedding poles lying on a great circle with a pole (the fold axis) generally oriented parallel to bedding– cleavage intersection lineations (Fig. 3a). The spread of the cleavage poles along the same great circle is consistent with the field observation of cleavage refraction with spaced, convergent and slaty, divergent cleavage fans in the sandstone and slate layers, respectively.
On the other hand, measurements from the fold termination in the Oldham Gold District show bedding poles lying on a small circle 70° ± 10° (corresponding to an apical angle of 40° ± 20°) with a cone axis located at the center of the small circle oriented at 056°/56°NE, and with a hinge line orientation of 054°/34°NE (Fig. 3b). As the fold hinge is more gently plunging than the cone axis, the conical fold must die out northeastwards by widening into an apron (cf. Figs. 3c–3d). This is consistent with the map pattern and cross-sections, which show that the fold varies from tight and angular (Fig. 2, cross-section A–B) to more open and rounded (Fig. 2, cross-section C–D) before dying out first into a monocline and finally into planar bedding (Fig. 2). This geometry is also consistent with the fact that the northeastern terminations of the Oldham anticline and its neighbouring syncline do not terminate at the same location (Fig. 2). The cleavage poles appear to be distributed about a great circle centred on an axis oriented at 057°/10°NE. The bedding– cleavage intersections spread about 50° along the fold axial plane, consistent with the intersection of cleavage with a conical fold (Fig. 3b). A steeply plunging lineation defined by biotitic mineral aggregates in the slaty cleavage show a 40° spread along the great circle about the divergent cleavage fan axis (Fig. 3b).
Measurments of 30 sand volcanoes show that the long axes are generally parallel to the bedding–cleavage intersection lineation (Fig. 4). Although not enough sand volcanoes were measured to be statistically meaningful, the mean aspect ratio of the sand volcanoes is ∼2:1, however there is considerable variation: on bedding surfaces dipping at < 50°, aspect ratios vary from 1:1 to 6:1, whereas on bedding planes with dips > 50°, aspect ratios show a more restricted range between 0.7:1 and 2.5:1. These measurements are similar to those of Henderson et al. (1986), who show that aspect ratios of the deformed sand volcanoes vary from 1.5:1 to 3.3:1, at dips < 60°, to a more restricted range between 1.5:1 and 2.1:1 at dips > 60°. Furthermore, Henderson et al. (1986) showed that the sand volcano dewatering pipes are skewed nearer to bedding-normal than the cleavage.
The tightness of folding in the fold hinge may be expressed by the formula
The radius is measured in a circle that is tangential at the fold hinge (Figs. 5a–5c; Lisle and Robinson 1995). The maximum curvature for a cylindrical surface will be found in the profile plane: this also applies to individual surfaces within a conical fold (N.B. neighbouring surfaces in conical folds have different fold plunges). In a cross-section of parallel Class 1B folds, the radius decreases by the thickness between neighbouring surfaces (Keppie 1976). Thus, in an anticline, structurally lower surfaces will have a greater curvature, and this increases exponentially when the radius is < 1 (Fig. 5d). This is qualitatively illustrated by the vertical cross-sections (Fig. 2) because the plunge of the fold axis of the Oldham anticline varies from 20°SW to 34°NE. This is quantitatively expressed by the formula (Figs. 5a–5c) Where
a=distance between the axial trace and the closest bedding reading to the fold hinge measured in a direction parallel to the line of intersection of the profile plane with the map surface;
β= the pitch of the axial plane in the profile plane;
α = the angle between bedding plane pole and the axial plane in the profile plane.
The curvature of the Oldham anticline has been calculated in the two cross-sections on either side of the axial trace (Table 1) and shows that the curvature of the fold increases rapidly from ∼1 at C–D to between 7 and 29 at A–B over a distance of 3 km (≡ true thickness of 1.68 km). If the Oldham anticline were Class 1B, the curvature of the hinge at cross section A–B would be unrealistically large. However, Keppie (1976) measured minor folds in the Meguma Group and demonstrated that they approximate Class 1B in the psammites, whereas they are Class 1C-2-3 in the pelites. Using these observations, the increase in the curvature may be reduced to realistic values because (1) the interbedded slates are thickened in the hinge, (2) the strain in the hinge is associated with reverse faulting (Fig. 2), and (or) (3) the space is filled with introduced material (i.e., quartz veins). The fact that auriferous saddle reef quartz veins were preferentially developed in the northeast-plunging conical termination of the Oldham anticline may be related to the non-parallelism of neighbouring surfaces in a conical fold hinge. This leads to the creation of more space between neighbouring psammites in the conical part of the fold.
The various structural measurements may be used to determine the mechanism of fold generation. The greater range of aspect ratio of the sand volcanoes observed in the bedding planes in the hinge versus the limbs of the Oldham anticline is incompatible with a purely flexural mechanism of folding, but may be due to the operation of other fold mechanisms and (or) a component of flattening (Ramsay 1967). Thus, the presence of reverse faults in the hinge of the Oldham anticline (Fig. 2) is indicative of a tangential longitudinal strain mechanism, the downdip biotite mineral lineation may indicate a component of vertical stretching in the slaty cleavage, and the presence of several generations of cleavage is consistent with a component of superimposed flattening (Ramsay 1967).
Except for the addition of non-planar strain in the slaty cleavage, these conclusions are generally consistent with those of Keppie (1976), who used dip isogons and thickness measurments around minor fold hinges to deduce that the mechanism of folding involved a combination of flexural and tangential longitudinal strain with a small component of homogeneous strain (flattening) superimposed normal to the axial plane. Henderson et al. (1986), using the aspect ratios of deformed sand volcanoes and the relative orientations of dewatering pipes and the cleavage, deduced that folding mainly took place by a flexural mechanism with variations in bedding-parallel shear related to lithology and a component of superimposed flattening. Furthermore, Henderson et al. (1986) deduced that ∼50% shortening took place before folding: this interpretation was based upon the relatively constant aspect ratios of the deformed sand volcanoes regardless of dip. Williams and Hy (1990) observed a downdip mineral lineation in steeply dipping ductile shear zones associated with the development of a new foliation. This mineral lineation is parallel to that observed in the Oldham anticline. These observations confirm that the mechanism of folding in the Meguma Group is composite and probably involved flexural, tangential longitudinal and shear folding with components of flattening before, during and after folding.
Age of the NE Oldham conical fold termination and its associated auriferous veins
The presence of aligned, downdip biotitic mineral aggregates in the slaty cleavage that form at > 350°C provides a means for determining the age of the northeastern termination of the Oldham anticline. Although neither this biotite mineral lineation nor the correlative downdip shear zones have been dated, the time of biotite growth may be bracketed between the age of peak low-pressure and high-temperature metamorphism > 600°C (378 ± 1 Ma U–Pb nearly concordant monazite in granulite xenolith in lamprophyre dyke: Greenough et al. 1999) and ∼400–300°C cooling ages recorded by muscovite and biotite (367 ± 1 Ma and 366 ± 1.5 Ma 40Ar/39Ar plateau ages: Keppie and Dallmeyer 1987). This Late Devonian deformation is consistent with the observation that a component of upright folding deforms a ∼380 Ma, subhorizontal crenulation cleavage (Keppie 1983). It is also supported by the 376 ± 2 Ma 40Ar/39Ar muscovite plateau age recorded from the Ovens Anticline that has been related to the development of a second cleavage during late stage flexural-slip deformation (Hicks et al. 1999). The vertical stretching in the slaty cleavage appears to be synchronous with dextral shear fabrics observed in the Minas Fault system and may indicate strain partitioning between shear zones and inter-shear domains.
Dubey and Cobbold (1977) performed a buckling experiment, which shows that folds are diachronous, initiating at points of local inhomogeneity and growing by amplification and propagation along their length. Our data suggest any one point on a fold would progress through a conical fold to a cylindrical fold. Thus conical fold terminations represent locations where only the youngest increment of folding occurs. Elsewhere in a fold, it is difficult to distinguish younger from older increments. The preferential, spatial association of the auriferous saddle reef veins in the conical termination of the Oldham anticline is probably a consequence of the greater volume of space between neighbouring layers in a conical fold in comparison with a cylindrical fold. Thus, it is possible that the mineralizing fluids migrated with the conical fold termination as it propagated along its length. Therefore, by association, the concentration of auriferous quartz veins in the conical northeastern termination of the Oldham anticline implies that this mineralization is also bracketed between ∼378 and 367 Ma. On the other hand, gold mineralization is also preferentially developed in domes and in conical segments of anticlines that are not located at fold terminations (Keppie 1976). This raises the possibility that some of the gold mineralization is older, however, younger increments of folding in such locations would produce fold amplification resulting in increased curvature and vein volume, both of which were empirically recorded by Keppie (1976). However, such geometric factors cannot distinguish older and younger increments of folding. Thus, gold mineralization in domes may also have developed during the younger increments of folding. This is consistent with the conclusions of Kontak et al. (1990) that gold and associated sulphide mineralization is ∼370 ± 8 Ma, including auriferous veins within the chlorite zone, which suggests that the introduction of mineralizing fluids was short-lived (Kontak et al. 1990).
Age of the cleavage
As noted earlier, Henderson et al. (1986) believe that some cleavage development occurred prior to folding. The 40Ar/39Ar dating of whole-rock phyllites appears to date this earlier cleavage development and ranges from 400 to 377 Ma (Reynolds and Muecke 1978; Keppie and Dallmeyer 1987, 1995; Kontak et al. 1998; Hicks et al. 1999). The youngest deformed rocks are the lower Emsian Torbrook Formation (Bouyx et al. 1992), and the Emsian spans the period 409.5–394 Ma (Tucker et al. 1998). Thus the transition from shallow-marine to subaerial environment recorded by the Torbrook Formation may reflect the onset of the deformation. The 400–377 Ma age range of the early cleavage development is also consistent with the observation that some of the folds in the Meguma Group are truncated by the ∼380–370 Ma granitoid plutons (Keppie and Dallmeyer 1987, 1995; Benn et al. 1997). On the other hand, later cleavage development is recorded in shear zones within the 380–370 Ma granitoid plutons and in the ∼360 Ma high-grade dextral and normal shear zones (Keppie and Dallmeyer 1987; Williams and Hy 1990). Thus development of cleavage appears to have lasted for at least ∼40 Ma. Although not rigorously documented, these observations are consistent with the development of several successive cleavages at Oldham. However, it is puzzling to find that in other places, only one cleavage is present. This may be explained by experiments that show later deformation producing structures that mimic an original anisotropy (Price and Thompson 1991) and suggests that some of the cleavage in the Meguma terrane may be composite.
Fold geometry and location of a decollement plane
With a wavelength of ∼2 km, the Oldham anticline is a third- or fourth-order fold compared with the ∼14.4–16 km wavelengths of the first-order folds (Keppie 1976). Applying the exponential increase in curvature of folds with a radius of < 1 to the Meguma Group (Fig. 5d), the anticlines should terminate at depth in a decollement surface. Using the fold geometry, it is possible to calculate the depth to this decollment plane. Assuming that the ratio of psammite:pelite is 3:1, that the folds approximate a sine curve, that their fold class is 1B and 2 for the psammites and pelites, respectively, (Keppie 1976), and that the decollement is a planar surface, then the decollement should lie at a depth of ∼9–10 km below the Halifax–Goldenville boundary in synclinal keels. This depth is just ∼2–3 km greater than the minimum thickness of the Goldenville Formation (> 7 km), and it is possible that the Goldenville–basement contact could coincide with such a decollement. This decollement may be recorded in the reflection seismic data by reflectors at 3–4 s (two-way travel time: Keen et al. 1991) and coincides with the velocity boundary between 6.2 and 6.4 km/s on line 99-2 (Jackson et al. 2000).
The apparent contradiction between the observation that auriferous veins in fold hinges yielded 370 ± 8 Ma ages and yet the folds are transected by 378 Ma metamorphic isograds and truncated by 380–370 Ma granitoid plutons appears to have been resolved by the study of the conical northeastern termination (Oldham Gold District) of the Oldham anticline. Experimental studies have shown that buckle folds initiate at a point of inhomogeneity and propagate both in amplitude and along their length (Dubey and Cobbold 1977). Thus, fold terminations clearly represent the youngest increment of folding. The age of aligned biotitic mineral aggregates in the slaty cleavage at the Oldham Gold District, which record a vertical stretching component of deformation, is bracketed between 378 and 366 Ma in adjacent areas. Thus, the final stages of folding were synchronous with the 370 ± 8 Ma 40Ar/39Ar ages on minerals in the veins. This implies that the auriferous vein fluids moved into the conical fold termination as it formed. The present location of auriferous saddle reef veins in the northeastern conical termination of the Oldham anticline is consistent with the creation of space for the veins where it changes from cylindrical to conical and where the curvature is > 2. Application of these constraints to auriferous saddle reefs veins in domes suggests that mineralizing fluids were only present during the youngest increments of fold amplification. This reconciles the apparent contradiction between the presence of auriferous veins as saddle reefs preferentially located in domes and cones with high curvatures (Keppie 1976) and their young age (Kontak et al. 1990). The deep source for the gold and associated sulphide mineralization is compatible with the high heat flow associated with the overriding of a mantle plume, which would also produce the low-pressure and high-temperature metamorphism, releasing fluids from the lower crust that would rise into the upper crust, induce melting of the crust to produce the granitoid magmas, and cause rapid denudation (Murphy et al. 1999; Keppie and Krogh 1999).
This paper includes results from D.F. Keppie’s Honours thesis undertaken at St. Francis Xavier University. Funds for this project were provided by the Natural Sciences and Engineering Research Council to JBM. We would like to thank Stephen F. Cox and Paul K. Smith for constructive criticism of the manuscript. We are grateful to Jose Luis Arce for preparing the figures for publication.
- Received January 23, 2001.
- Accepted July 26, 2001.
- Published on the NRC Research Press Web site at http://cjes.nrc.ca on January 7, 2002.
- © 2002 NRC Canada