August 1, 2008

Chemical and Nd Isotope Constraints on Granitoid Sources Involved in the Caledonian Orogeny in Scotland

By Steinhoefel, Grit Hegner, Ernst; Oliver, Grahame J H

Abstract: Major- and trace-element data and Nd isotope compositions for granitoid samples from the Grampian Highlands in Scotland show a systematic evolution in the composition of their sources in the course of the Caledonian Orogeny. Granitoids of 511- 451 Ma, related to the collision of the Midland Valley island arc with the Grampian terrane, show S-type affinity and fractionated REE patterns with minor Eu anomalies and low initial epsilon^sub Nd^ values of -14.1 to -11.2 suggesting melting of predominantly Dalradian metasediments. Subsequently formed granitoids of 425-406 Ma derived from an assumed Andean plate margin comprise a wide spectrum of rock types including I-type granite-granodiorite, and S- type granitoids, monzonites and alkali granites. The trace-element patterns of these rocks and the range of initial epsilon^sub Nd^ values of -2.1 to -6.9 are consistent with melting of variably rejuvenated crust as found in continental margin settings. We conclude that the Grampian Highlands were affected by two major crust-modifying events during the Caledonian Orogeny: predominantly recycling of older crust during docking of the Midland Valley arc and addition of juvenile, mantle-derived material to the crust during the convergence of Avalonia with Laurentia.

The granitoids of the Grampian Highlands form an important member of the Caledonian mountain belt of Scotland and Ireland. They have been interpreted as originating from the Caledonian orogenic cycle from c. 600 to 395 Ma, encompassing continental rifting and opening of the Iapetus Ocean, arccontinent collision and subsequent subduction, and final plate collision (e.g. Stephenson & Gould 1995; Oliver 2002; Strachan et al. 2002). Thus the Grampian Highlands present a field laboratory to develop and test models for the geodynamic evolution of the margin of Laurentia as well as the Caledonian Orogeny. In the Grampian Highlands, widespread magmatism was caused by the collision of an island arc with the passive Laurentian margin (the Grampian Orogeny), and subsequent plate subduction beneath the accreted arc. Granitic rocks produced during these stages provide petrographical, geochemical and isotope evidence for melting of compositionally different sources during the Caledonian Orogeny (e.g. Brown 1979; Halliday 1984; Harmon et al. 1984; Stephens & Halliday 1984; Thirlwall 1988).

A drawback of earlier studies of the granitoids of the Grampian Highlands is the fact that in many cases geochronological, isotope or geochemical data were interpreted independently because of the lack of comprehensive geochemical-isotope datasets. To date, many plutons are known on a reconnaissance basis only, and their apparently complex petrogenesis together with potential magma sources need to be investigated in more detail. In this study, we interpret major- and trace-element data as well as Sm-Nd isotope systematics in granitoid samples related to the collision of the Midland Valley arc with Laurentia and the subsequent subduction of oceanic lithosphere when Avalonia converged with Laurentia. Age constraints on the igneous events are provided by new ion- microprobe zircon ages presented in a companion paper (Oliver et al. 2008).

Geological background

The Caledonian mountain belt of Scotland is a continuation of the Appalachian mountain belt in North America. It consists of terranes that were assembled at the Laurentian margin during closure of the Iapetus Ocean (e.g. Bluck 2001; Fig. 1a and c). During the early phase of the Caledonian Orogeny from c. 470 to 450 Ma (e.g. Dewey & Mange 1999; Oliver 2001; Flowerdew et al. 2005), the continent- facing Midland Valley arc, in Scotland now presumably buried in the Midland Valley terrane, collided with the Grampian terrane. This led to the Grampian Orogeny in Scotland and Ireland. The orogeny corresponds to the Taconian phase in North America. The accretion of the Midland Valley arc caused deformation, high-grade regional metamorphism, and magmatism in the Grampian Highlands. After the arc- continent collision, continental convergence was facilitated by plate subduction under Laurentia, which ceased with the closure of the Iapetus Ocean and soft docking of Avalonia.

The origin of the Caledonian granitoids of the Grampian Highlands has been a recurring issue in geological studies for more than a century. Structural analyses going back to Barrow (1893) yielded a general division into metamorphosed and deformed as well as undeformed granitoids. The two groups were termed the 'Older Granites' and 'Newer Granites', respectively, by Read (1961). Based on structural and geochemical evidence the Caledonian granitoids have been assigned to pre-tectonic, syntectonic, late-tectonic and post-tectonic stages of the Grampian Orogeny (Stephenson & Gould 1995). The Grampian terrane comprises a pre-Dalradian basement, overlain by Dalradian metasedimentary rocks, as well as cross- cutting Caledonian granitoids (Fig. 1b and c). The basement of the Grampian terrane is not exposed in the Grampian Highlands, but one example is thought to occur as the c. 1.8 Ga Rhinns Complex on Islay and Colonsay in the Inner Hebrides (Muir et al. 1994). Here it comprises syenitic gneisses derived from juvenile mantle-derived protoliths (Marcantonio et al. 1988; Daly & McLelland 1991; Muir et al. 1994). Muir et al. (1994) reported geochemical data for the complex that revealed a calc-alkaline composition and trace-element patterns with negative Nb and Ti anomalies suggesting an island-arc or subduction-related environment. The Rhinns basement probably forms a continuation of the Ketilidean belt of southern Greenland and is assumed to underlie the NW Grampian Highlands, as has been inferred from the Nd isotope compositions of Caledonian granites (Dickin & Bowes 1991). In addition to the Rhinns component, isotope data indicate that the Caledonian granites incorporate late Mesoproterozoic (Grenvillian) material of c. 1.2-1.0 Ga age (Clayburn 1988). Possible subordinate components of the basement originated from the Scourian event at c. 2.7 Ga and the Knoydartian event at c. 0.8 Ga. The evidence for the Scourian event has been inferred from the Nd model ages of the Corrieyairack granite (clayburn 1988). Some of the tectonic deformation within the Grampian Highlands has been attributed to the Knoydartian event (Noble et al. 1996; Highton et al. 1999). Dalradian metasedimentary rocks cover most of the Grampian Highlands and are made up of Archaean and Proterozoic detritus (Cawood et al. 2003) deposited at the passive margin of Laurentia in the Neoproterozoic and Early Palaeozoic. Contemporaneously with Dalradian sedimentation, the c. 600 Ma rift-related Tayvallich volcanic rocks and a series of small granitic intrusions were emplaced (e.g. Anderton 1985; Halliday et al. 1989; Tanner 1996; Dempster et al. 2002). These granitoids belong, among others, to the Older Granites' of Read (1961) and show S- and A-type characteristics with high initial Sr isotope ratios (see review by Stephenson & Gould 1995; Tanner et al. 2006). The syn- to late-tectonic stages of the Grampian Orogeny involved the collision of the Midland Valley island arc with the Grampian terrane from c. 470 to 450 Ma (Stephenson & Gould 1995; Hutchinson & Oliver 1998; Oliver 2001). Arc collision was accompanied by emplacement of granitic rocks and subordinate mafic to ultramafic intrusions (e.g. the c. 470 Ma 'Newer Gabbros'; Dempster et al. 2002). Among the late- tectonic granitoids, two types may be distinguished: (1) biotite- muscovite-bearing granites and granodiorites with S-type characteristics and high initial Sr isotope ratios (Pankhurst 1974; Pidgeon & Aftalion 1978); (2) granodiorites with I-type characteristics and subordinate diorites (e.g. Harrison 1987; Gould 1997). The origin of these granitoids has been explained by decompression melting of lower crust following post-orogenic uplift (Oliver 2002). During the post-tectonic stage of the Grampian Orogeny, plate subduction took place under Laurentia as a result of the continuing continental convergence. Associated magmatism lasted from c. 430 to 395 Ma and it is characterized by the emplacement of voluminous intermediate to felsic igneous rocks; that is, the 'Newer Granites' of Read (1961). They are undeformed, show I-type affinity, and may be subdivided into three suites (e.g. Halliday 1984; Stephens & Halliday 1984; Halliday et al. 1985; Plant 1986): (1) the southern Grampian Highland Suite, with small intrusions of dioritic to granitic composition and mafic dykes; (2) the Argyll Suite, including tonalites and granites with unusually high Ba und Sr concentrations; (3) the Cairngorm Suite, with highly differentiated biotitebearing granites of transitional I- to A-type composition. The predominantly calc-alkaline composition of the post-tectonic granitoids has been interpreted as evidence for an active ?Andeantype plate margin setting of the Grampian terrane after the Midland Valley island arc was accreted (Dewey 1971; van Breemen & Bluck 1981; Soper 1986; Thirlwall 1988; Oliver 2001). The final voluminous granite magmatism following docking of Avalonia has raised much speculation regarding the underlying geodynamic causes (e.g. Bluck 2001; Oliver 2001). It has been discussed in the context of an oblique collision of Avalonia and Baltica with Laurentia at c. 435 Ma and accretion of Armorica at c. 410 Ma (Soper & Hutton 1984; Oliver 2002). Atherton & Ghani (2002) were the first to explain the final magmatism by slab break-off during the late stage of subduction. Recent studies by Oliver (2002) and Oliver et al. (2008) have refined the slab break-off model for the British Caledonides. The Caledonian granites show mainly S- and I-type characteristics. This classic division was originally established for the granitoid suite of the Lachlan Fold Belt in SE Australia by Chappell & White (1974) and has been applied to granite suites all over the world. S- types are peraluminous and derived from sedimentary protoliths whereas I-types are peraluminous to metaluminous and extracted from igneous protoliths. Typically, S-types exhibit higher initial Sr isotope ratios and higher K, Rb and Pb but lower Na, Ca and Sr contents when compared with I-types (Chappell & White 1992). Further recognized varieties within the Caledonian suite are A-type granites. A-type granites are interpreted as of anorogenic origin and characterized by high SiO^sub 2^, Fe/Mg, Ga/Al, Zr, Nb, Ga, Y and Ce, and low Ca and Sr (e.g. Whalen et al. 1987). However, highly fractionated I- and S-type granites may exhibit similar chemical features (Whalen et al. 1987). Many of the Caledonian granitoids show transitional compositions of S- and I-type or I- and A-type, underlining the difficulty of unravelling their petrogenesis with geochemical discrimination diagrams.

Samples and analytical methods

For the purpose of constraining source evolution and possible tectonic settings of the granitoid and related rocks, 25 samples thought to be representative of the Caledonian Orogeny were analysed. Samples assigned in this study to the 'collision stage' are those referred to by Stephenson & Gould (1995) as belonging to the syn- and late-tectonic stages. They characterize the main tectonic event, namely, the collision of the Midland Valley arc with the Grampian Highlands. The subsequent stage with Andean-type magmatism is referred to the 'subduction stage' and includes rocks attributed to the post-tectonic stage by Stephenson & Gould (1995). A brief petrographie sample description, available ages including new zircon ages determined by sensitive high-resolution ion microprobe (SHRIMP) (Oliver et al. 2008), and sample localities are listed in the Supplementary Publication, which is available online at SUP 18313. The sampled intrusions are indicated in Figure 1b. The zircon data of Oliver et al. (2008) were used to establish the chronology of emplacement of the samples and to calculate initial ENd values. Some intrusions did not yield precise zircon ages as a result of open-system behaviour of the U- Pb system during Barrovian metamorphism at 470 Ma (Sm-Nd garnet age; Oliver et al. 2000). For example, the discordant U-Pb zircon data for the Rough Craig Intrusion (RC-15) yielded only an imprecise intercept age of 511 +-37 Ma. For the Dunfallandy Hill intrusion (samples DH-1 and DH-2), we assume an age similar to that of the Rough Craig intrusion (sample RC-15), as they are isotopically and geochemically similar and the Rb-Sr whole-rock isochron age of 481 +- 15 Ma for the Dunfullandy Hill intrusion (Pankhurst & Pidgeon 1976) overlaps within error that of RC-15.

Fresh rock chips were ground in a tungsten carbide mill. The major- and some trace-element concentrations were analysed by XRF at the University of St. Andrews, Scotland. Analytical details have been described by Stephens & Calder (2004). Repeated analyses of sample GK-3 (10 measurements) yielded an external precision of +-5% (2 SD) for most major elements and the reported trace elements. La, Ce, Nd, and Th at very low concentration levels (


Major- and trace-element data

The SiO^sub 2^ concentrations in the granitoid samples of the various tectonic stages range from 64 to 78 wt% and we classify them as granodiorite to high-silica granite. (Analytical data are available in the Supplementary Publication, see p. 819.) In the Streckeisen diagram using CIPW normative mineral compositions (Fig. 2), collision-related rocks are classified as granites, whereas subduction-related rocks reveal a wide compositional spectrum including monzodiorites, granodiorites, granites and alkali granites. All granitoid samples have a high-K calc-alkaline affinity as suggested by the K^sub 2^O v. SiO^sub 2^ diagram of Rickwood (1989; not shown). When considering the Al, Na, K and Ca concentrations (Shand's Index), the samples of the various tectonic stages reveal overall peraluminous compositions although at different levels (Fig. 3). The collision-related granitoids characterized by high SiO^sub 2^ concentrations of 70-74 wt% have a distinct peraluminous composition as indicated by an A/CNK ratio (molar Al^sub 2^O^sub 3^/(CaO + Na^sub 2^O + K^sub 2^O) of 1.1-1.3 (Clarke 1981)). This finding indicates a strong affinity to S-type granitoids. The subduction-related granitoids exhibit variable SiO^sub 2^ concentrations of 68-78 wt% and they have peraluminous to slightly metaluminous composition (A/CNK = 0.8-1.2). This group of samples comprises transitional S- to I-type granitoids.

The trace-element patterns in Figure 4 show the typical characteristics of granitoid rocks, such as a strong enrichment in incompatible elements coupled with negative anomalies of Nb and Ti and positive Pb anomalies. In many samples there are paired negative anomalies of Eu and Sr. Below, we will describe in brief the trace- element characteristics of the samples. Of the four samples of the collision stage (DH-2, RC-15, GK-3, M-4, Fig. 4a-d) the older samples DH-2 and RC-15 (Fig. 4a and b) display steeper REE patterns (La/Yb^sub N^ = 25 and 13, respectively) than the younger ones, coupled with very small negative Eu and Nb anomalies. The younger samples GK-3 and M-4 (Fig. 4c and d) tend to have higher REE concentrations and less fractionated REE patterns (La/Yb^sub N^ = 6 and 13, respectively) than the older samples and they show slightly fractionated to unfractionated heavy REE (HREE) patterns. They have small positive Eu and Sr anomalies, and, in contrast to the older samples, distinctly negative Nb anomalies. The granitoids of the subduction stage depict a wide spectrum of trace-element patterns (Fig. 4e-1) and may be grouped into three types. Samples MC-5, GG- 12 and GD-16 (Fig. 4g and h) show subparallel REE patterns with moderately fractionated HREE patterns (La/Yb^sub N^ = 10-24). The samples show minor negative Eu anomalies and small anomalies for Pb and Ti. Samples P-22, P-27, LL-9, B-13, MB-19 (Fig. 4e, f, i and j) are characterized by slightly fractionated REE patterns (La/Yb^sub N^ = 9-14) and distinct negative Eu and Sr anomalies. The enrichment in Pb and Th in these samples is remarkable. A third type of pattern is confined to the dykes GG-11 and P-25 (Fig. 4k and 1). The samples show an enrichment of HREE over light REE (LREE) (La/Yb^sub N^ =0.3 and 1.0, respectively) with tetrad-type features (Bau 1996) such as convex patterns for four element groups (La-Nd, Nd-Gd, Gd-Er, Er- Lu). In addition, there are very large negative anomalies of Ba, Eu, Sr and Ti and significant enrichment of Th and Pb.

Sm-Nd isotope systematics

The Nd isotope data depicted in Figure 5 show a temporal variation with systematically increasing initial epsilon^sub Nd^ values with time and continuing orogeny. Full data are available in the Supplementary Publication, see p. 819. Samples of the collision stage exhibit low initial epsilon^sub Nd^ values of -14.1 to -11.2, corresponding to mean crustal residence ages (Arndt & Goldstein 1987) of 2.3-2.0 Ga. The granitoids produced during the subduction stage show much higher initial eNd values of -6.9 to -2.1 (Nd model ages 1.6-1.2Ga) than those formed earlier. Published Nd isotope data of Hamilton et al. (1980), Halliday (1984), Frost & O'Nions (1985) and clayburn (1988) recalculated for the new SHRIMP zircon ages of Oliver et al. (2008) plot on the data trend of samples of this study (Fig. 5) and provide additional support for the source evolution of the granitoids. Discussion

Temporal variation in source compositions and petrogenetic processes

A changing source composition for the granitoids of the arc- continent collision to the subduction stage is reflected in increasing epsilon^sub Nd^ values and a change from distinctly peraluminous to transitional peraluminous-metaluminous compositions. Noteworthy is the high abundance of felsic intrusive rocks and the lack of intermediate compositions, suggesting that crystal fractionation of mafic parental magmas was not important relative to partial melting of crustal sources during basalt underplating or during crustal delamination. We now explore the nature of the sources and petrogenetic processes for changing granitoid compositions with time.

Collision-related granitoids. The granitoids of 511 -451 Ma that intruded the Grampian basement during collision of the Midland Valley arc with the Grampian terrane have a distinct peraluminous S- type composition and exhibit very low initial epsilon^sub Nd^ values of -14.1 to -11.2 corresponding to old Nd model ages of 2.3-2.0 Ga (Fig. 5). All these characteristics are consistent with melting of large amounts of old sedimentary protoliths. The epsilon^sub Nd^ values of possible Dalradian metasedimentary sources reported in the literature are highly variable from c. -7 to -22 at 500 Ma (average about -14, corresponding to a Nd model age of 2.3 Ga; O'Nions et al. 1983; Frost & O'Nions 1985; recalculated for 500 Ma). In contrast, the uniform epsilon^sub Nd^ values in the granitoids of the collision stage require isotopically similar protholiths among the Dalradian source rocks. In Figure 5 it can be seen that the samples of the collision stage overlap the isotope evolution of the Rhinns, the Dalradian and the Grenvillian rocks. This coincidence shows that the Nd isotope data alone cannot constrain possible source rocks. However, involvement of Dalradian metasediments is supported by field evidence from the NE Grampian Highlands showing migmatites with sill-like granitic bodies and granites with rafts of metasedimentary rocks and/or gradational contacts with Dalradian country rock (Stephenson & Gould 1995; Johnson et al. 2001, 2003).

The REE patterns of samples RC-15 and DH-2 reveal subordinate negative Eu anomalies, indicating little plagioclase control during magma evolution, and fractionated HREE patterns are consistent with partial melting of garnet-bearing sedimentary rocks, probably greywackes (McMillan et al. 2003). Magma evolution by fractional crystallization from a mafic parental magma would have resulted in prominent negative Eu anomalies (Gromet & Silver 1987) in these rocks and can be ruled out. The unusually low total REE abundances suggest a source that underwent previous melting and possibly consumption of plagioclase. RC-15 and DH-2 have within-error limits similar ages and geochemical-isotopic characteristics, suggesting a comagmatic relationship. It can be seen that sample RC-15 with c. 73% SiO^sub 2^ has REE abundances a factor of two higher than the slightly more fractionated sample DH-2 with c. 74% SiO^sub 2^. If both samples were produced as separate magma batches by melting of a similar source, as required by Nd isotopes, the much higher REE abundances in sample RC-15 would require a much a smaller degree of source melting than for DH-2. However, the similar K^sub 2^O concentrations in both samples preclude large differences in the melting degrees for these samples and we need to explain the relationship of these samples and their different REE abundances by fractionation of accessory phases, rich in REE (e.g. Henderson 1984). The very low P^sub 2^O^sub 5^ concentration in DH-2 suggests fractionation of P-bearing phases such as apatite and monazite both having very high REE concentrations, and its low Ti concentrations may be due to involvement of titanite (e.g. Henderson 1984). Fractionation of apatite and titanite from a magma with the composition of RC-15 would drastically lower the REE abundances as required for sample DH-2. This would also explain the smaller negative Eu anomaly in DH-2 than in RC-15, as apatite and titanite discriminate against Eu. A higher Gd/Yb ratio in DH-2 is consistent with an origin by apatite-titanite fractionation from a parental magma such as RC-15, but a lower La/Sm ratio is the opposite of what would be expected. We suggest additional fractionation of highly LREE-enriched monazite, which would decrease the La/Sm in the evolved magma. The unusual small Nb anomaly of RC-15 and DH-2 appears to be inherited from the source, as trace-element patterns of Dalradian metapelitic rocks also exhibit very small negative Nb anomalies (Dalrymple 1995; Johnson et al. 2003). Applying the same line of reasoning to younger granitoids GK-3 and M-4, differing mainly in their contents of LREE at similar HREE, we suggest a comagmatic origin and fractionation of LREE-enriched monazite to account for the lower LREE abundances in the chemically more evolved sample GK-3. Both samples show slightly positive Eu anomalies indicating the presence of cumulate plagioclase, in particular in sample GK-3. Noteworthy are the slightly fractionated HREE patterns consistent with melting of garnet-free sources.

Subduction-related granitoids. Granitoids of 425-406 Ma are related to plate subduction along an Andean-type plate margin and represent the final igneous event of the Caledonian Orogeny in Scotland (e.g. Dewey 1971; Soper 1986; Oliver et al. 2008). They yielded much higher initial epsilon^sub Nd^ values of -6.9 to -2.1 and relatively young Nd model ages of 1.6-1.2 Ga when compared with those for the collision stage samples (Fig. 5). The Nd isotopes show clearly involvement of significant amounts of juvenile mantle- derived material. A very wide range of rock types ranging from granodiorites to alkali granites with I-type and transitional S- type characteristics were produced during this event. Alkali granites are associated with crustal rifting (e.g. Lee et al. 2003), so their presence suggests that the magmatic arc underwent extension (Soper & Hutton 1984; Dewey & Strachan 2003; Oliver 2002; Oliver et al. 2008). It is noteworthy that highly evolved granitoids dominate the rock spectrum in the subduction stage. Intermediate andesitic compositions are evidently rare and this finding suggests melting of mostly crustal protoliths, as would be expected in regions of thick crust heated from below by underplating of basalt, or input of heat from rising asthenosphere after delamination of lower crust. Development of a lower and middle crust with mixing, crustal assimilation, storage, and hybridization of melts (MASH), as originally proposed by Hildreth & Moorbath (1988) and refined by Annen et al. (2006) for Andean-type margins, may serve as a model. During plate subduction below Laurentia, mantle-derived melts probably underplated and assimilated older crust of possibly Rhinnian and Grenvillian age. The mixing of residual melts from basalt crystallization and crustal partial melts of metasedimentary and metaigneous rocks can explain the diverse isotope and trace- element compositions (Annen et al. 2006). The Nd isotopes of the subduction-related granititoids indicate variable mixing of different sources, whereas the trace elements are mainly controlled by melting and melt-fractionation processes. The range in initial epsilon^sub Nd^ values of -2 to -8 suggests a heterogeneous lower basement of the Grampian terrane, which is probably of Rhinnian and Grenvillian origin and has been rejuvenated during the Caledonian orogenic cycle. Associated Lome lavas yielded initial eNd values of c. -4 to +1, reflecting large amounts of mantle-derived material (Thirlwall 1982).

The 420-415 Ma granitoids are metaluminous in composition, indicating a high proportion of igneous material in their sources. The protoliths also contain in some cases large amounts of mantle- derived material as indicated by initial epsilon^sub Nd^ values of up to -2.1 (Fig. 5). The arc-like REE patterns in samples GD-16, MC- 5 and GG-12 strongly suggest an origin linked to plate subduction including melting of mantle and lower crustal sources. The samples are similar to the 'high Ba-Sr' granites of the Argyll Suite described by Stephens & Halliday (1984). High concentrations of the fluid-mobile elements Ba and Sr support an origin from subduction- modifled protoliths for these samples. Our interpretation of the origin of the samples of this study is at odds with the geodynamic model of Oliver et al. (2008), who argued for cessation of plate subduction already at 420 Ma, followed by calc-alkaline magmatism in extensional basins during slab roll-back. Accepting the latter scenario, we would need to conclude that the subduction characteristics were inherited from a previous tectonic setting.

When looking at the details of the trace elements in these samples it can be seen that their LREE-enriched patterns show only minor negative Eu anomalies, as is typical of many subduction- related magmas from mature island arcs or active continental margin (e.g. Gromet & Silver 1987). The magmas must have erupted rapidly from the lower crust without ponding in shallow-level magma chambers with major plagioclase fractionation (Gromet & Silver 1987). The high Sr and Ba concentrations in these rocks further reflect the subordinate role of feldspar and large contributions of slab- derived fluids, respectively (summary of references of Morris & Ryan 2003). These features, together with unfractionated HREE patterns, suggest limited melt fractionation of mantle-derived magmas outside the garnet stability field (Fowler & Henney 1996; Fowler et al. 2001). GD-16 is of basaltic-andesite composition (MgO 4.2 wt%) and represents a typical mantle-derived melt from subduction-modified mantle. Its low initial epsilon^sub Nd^ value of -5.9 indicates a large proportion of older crustal material inherited from assimilation of lower crust and/or inherited from subducted old sediment. The granodiorites MC-5 (420Ma) and GG-12 (415 Ma, Fig. 4d) have similar REE patterns, but much higher initial epsilon^sub Nd^ values of -3 to -2 indicate different sources from that of the more mafic granitoid GD-16. The origin of granodiorite in subduction zones has been discussed in the context of magma fractionation from basaltic parental magmas and partial melting of mafic protoliths (see discussion by Gromet & Silver 1987). The lack of intermediate rock compositions and distinct negative Eu anomalies in the samples do not support magma fractionation as the primary process for their composition. We suggest that, in addition to an origin of the mafic rocks by mantle melting, partial melting of mafic lower crust during magma storage in the lower crust was important for the granitoids (e.g. Rapp & Watson 1995). Another group of subduction-related granitoids comprises high-silica alkali granites of the Peterhead intrusion emplaced at c. 425 Ma (e.g. P-22, P-27, Fig. 4e), shortly before the end of plate subduction according to the model of Oliver et al. (2008). For these granitoids, magma evolution in shallow- level magma chambers in the presence of feldspar and Ti-oxides is indicated by their large negative Eu, Sr and Ti anomalies, respectively. For these samples we suggest a two-stage magma evolution by partial melting of mafic lower crust followed by magma fractionation at higher crustal levels. Initial epsilon^sub Nd^ values of -4 to -7 indicate again involvement of large proportions of older crust.

Besides the more typical subduction-related rocks, highly fractionated alkali granite dykes found in the Glen Gairn and Peterhead intrusions (samples GG-11 and P-25) were also emplaced during final plate subduction. These samples are strikingly distinct and show similar isotopic and geochemical characteristics so we assume them to be of similar age and comagmatic origin. Both ages that we cite for these samples have not been published, so that the definite age may be 415 Ma, as suggested by U-Pb zircon dating (Parry, pers. comm.), or 425 Ma (Torsvik, pers. comm.). The age difference is not important here for deciphering their petrogenesis. The evolved major-element composition and remarkable depletion in P^sub 2^O^sub 5^ and LREE may be due to fractionation of monazite and apatite, and low Ba, Sr, Eu and Ti contents can be explained by extensive magma fractionation of feldspar and titanite. ENO values of -6.2 and -6.4 suggest sources similar to those of the contemporaneously emplaced granitoids. The REE patterns of these samples show a slight tetrad effect that is characterized by four element groups (La-Nd, Nd-Gd, Gd-Er, Er-Lu) forming four segments of convex patterns. The tetrad effect is often accompanied by a non- charge-and-radius-controlled (non-CHARC) behaviour of other trace elements (Bau 1996), as reflected in Zr/Hf values of 12 and 14, respectively, which are distinct from a value of 37 +- 3 in mantle- and crust-derived igneous rocks (Bau 1996). These characteristics have been explained as the result of melt fractionation processes, a major influence of apatite (McLennan 1994), and interaction of residual granitic melts with hydrothermal fluids in a highly evolved magma system (Jahn et al. 2001). Adopting the hypothesis that the felsic rock types represent partial melts of lower crust, rather than the products of fractionation of mafic parental melts, these high-silica LREE-depleted granitoids may represent small-volume residual melts after extensive shallow-level melt fraction.

The c. 406Ma granites (sensu stricto; e.g. MB-19, LL-9, B-13; Fig. 4) are transitional between I- and S-type and may have been generated after active subduction (Oliver et al. 2008). As mantle melting cannot have been important at that stage, we interpret these typical granites as resulting from melting of lower crust. Apparently, the protoliths were similar to those involved during final plate subduction, as can be inferred from similar initial exd values of -4 to -5. Large negative Eu, Sr and Ti anomalies are consistent with magma evolution in shallow-level magma chambers in the presence of feldspar and titanite. The homogeneous compositions and small range in initial eNlj values of these granites indicate melting of similar sources and magma evolution under similar conditions. The geochemical evidence in these granites fails to identify the tectonic environment of their genesis; instead, characteristics of older tectonic processes are recycled in their trace elements and isotopes.

Plate tectonic development

The Caledonian magmatism in Scotland is closely related to the tectonic evolution of the Laurentia margin and can be unravelled with the chemical and isotope data for igneous rocks that represent the different stages of orogeny. In the Neoproterozoic, Laurentia rifted from West Gondwana during the break-up of the supercontinent Rodinia and the initial opening of the Iapetus Ocean. In the Ordovician, collision of island arcs with Laurentia marked the beginning of the closure of the Iapetus Ocean, which resulted in the amalgamation of Laurentia, Avalonia, and Baltica. Below, we discuss a possible geodynamic evolution of the Grampian terrane (Fig. 1c), which follows recently published models (Oliver 2001, 2002; Oliver et al. 2008) and integrates the new geochemical and isotope information obtained in this study. It can be seen that the geochemical data of the highly dynamic orogenic development provide a rather simple image of crustal evolution with remobilization of pre-existing crust followed by production of increasingly more rejuvenated crust.

Rifting stage. Crustal thinning and lithospheric rifting at c. 600 Ma produced bimodal mafic-felsic magmatism consistent with melting of mantle and crustal sources (Fig. Ic, stage 1). The c. 590 Ma Ben Vuirich A-type monzogranite (U-Pb zircon TIMS age, Rogers et al. 1989) yielded initial eNd values of -4.3 to -6.3 consistent with melting of mixed recycled-juvenile sources (Hamilton et al. 1980; Tanner et al. 2006). The c. 600 Ma Tayvallich volcanic rocks (Dempster et al. 2002) with initial epsilon^sub Nd^ values of up to +4 (Halliday et al. 1989) reveal mostly depleted upper mantle sources. The Tayvallich volcanic rocks are probably related to extensive mantle melting and crustal underplating during continental break-up (Anderton 1985; Graham 1986; Kamo et al. 1989). From c. 585 to 470Ma the margin of Laurentia remained passive as the Iapetus Ocean was opening (Fig. 1c, stage 2; Oliver et al. 2008).

Collision stage. There is general agreement that in the early Ordovician a continent-facing island arc collided with the Laurentia margin, and this caused the Grampian Orogeny in Scotland (e.g. Dewey & Shackleton 1984; Fig. Ic, stage 3). Magmatic and cooling ages constrain magmatism, deformation, high-grade regional metamorphism, and unroofing of Scotland and Ireland to a short time interval from c. 470 to 460 Ma (Dewey & Mange 1999; Friedrich et al. 1999; Soper et al. 1999; Oliver et al. 2000). Geochemical and isotope data for the collision-related granitoids indicate rather uniform, mostly metasedimentary sources probably of Dalradian provenance. Melting of sedimentary protoliths may have been facilitated by heat supplied from mantle-derived melts forming intrusions such as the 'Newer Gabbros' at c. 470 Ma (Dempster et al. 2002). The origin of these mafic to ultramafic intrusions as well as the highgrade regional metamorphism has been explained by a number of processes such as slab break-off (Oliver 2001, 2002; Oliver et al. 2008), delamination of the lower mafic crust (Draut et al. 2002, 2004), and contact of hot asthenosphere with the Laurentia margin during collision with the arc (Dewey & Mange 1999). Decompression melting of Dalradian metasedimentary rocks is thought to have resulted from isostatic readjustment of the thickened Grampian terrane (Oliver 2001, 2002). Evidence for decompression, uplift and erosion is provided by K-Ar cooling ages of c. 455 Ma for mica (Dempster 1985) and the contemporaneous deposition of metamorphic detritus from the Grampian terrane in the Southern Uplands accretionary prism (Oliver et al. 2000).

The data presented here emphasize the importance of the melting of a Dalradian metasedimentary source during the collision of the Midland Valley arc with the Grampian terrane. The trace-element data suggest melting of material with characteristics of garnet-bearing and garnet-free greywackes. The necessary heat may have been generated during crustal thickening and intrusion of mantle-derived melts. The lack of evidence for juvenile material in these granitoids, if representative for the collision stage, indicates that subduction zone melting below Laurentia was not important before c. 430 Ma as proposed in recent models for Scotland (Oliver 2001, 2002).

Subduction stage. After the Grampian Orogeny, an active Laurentia margin probably developed with subduction beneath the accreted Midland valley arc and Grampian terrane (Fig. 1c, stage 4). Zircon ages constrain the final igneous activity to c. 430-390Ma (Oliver et al. 2008). The geochemical data presented above show a very wide compositional spectrum of mostly granitoids, suggesting melting of protoliths with a high proportion of mantle-derived material. The sources of the granitoids apparently were produced in a plate subduction setting. It has been argued that subduction ceased with the soft docking of Avalonia before magmatism ceased (e.g. Soper et al. 1992). This would imply that the late generation of magma resulted from post-subduction processes. This possibility cannot be precluded with the geochemical data presented here, which show that the youngest granitoids are highly evolved granites probably derived from melting of earlier rejuvenated crustal sources. Strike-slip faulting caused by the oblique collision of Avalonia and Baltica with Laurentia (e.g. Soper & Button 1984; Dewey & Strachan 2003), slab roll-back (Oliver 2002) and/or slab break-off (Atherton & Ghani 2002) would all be associated with heating of the lower crust and decompression melting. Processes such as slab roll-back and slab break-off have been proposed during the late stage of subduction but cannot be evaluated with geochemical data, as they would result in magmatic underplating and production of granitoids with similar compositions to those formed during plate subduction. The Nd isotope evidence for melting of mantle-derived material may be taken as evidence for plate subduction until c. 400 Ma, but extensive lower crustal melting in a collision zone cannot be precluded. Conclusions

The geochemical-geochronological data for granitoids from the Grampian Highlands are consistent with melting of crust and mantle during continental rifting and associated magmatic underplating at c. 600Ma, arc-continent collision from c. 470 to 450 Ma, and subsequent development of an Andean-type plate margin. The geochemical-isotope data reveal two major igneous events that affected the crust. Granitoids produced during docking of the Midland Valley arc with Laurentia were predominantly derived by melting and recyling of older crust. Contemporaneously emplaced mantle-derived material apparently did not contribute to granite magmatism, and probably simply provided a heat source for melting of Dalradian metasediments. During the final convergence of Avalonia and Laurentia, melting of mixed and distinctly more juvenile sources reflects the input of mantlederived material during plate subduction. Contemporaneous emplacement of alkali granites within the Andean-type plate margin could have resulted from either oblique plate subduction or post-orogenic melting of collision zone assemblages.

A. Calder carried out the XRF analyses at St. Andrews and A. Rocholl supported the analysis of Nd isotopes in Munich. We thank M. Fowler and I. Millar for their helpful and constructive reviews of the manuscript. G.S. would like to thank C. HeIo for discussions.


ANDERTON, R. 1985. Sedimentation and tectonics in the Scottish Dalradian. Scottish Journal of Geology, 21, 407-436.

ANNEN, C., BLUNDY, J.D. & SPARKS, R.S.J. 2006. The genesis of intermediate and silicic magmas in deep crustal hot zones. Journal of Petrology, 47, 505-539.

ARNDT, N.T. & GOLDSTEIN, S.L. 1987. Use and abuse of crust- formation ages. Geology, 15, 893-895.

ATHERTON, M.P. & GHANI, A.A. 2002. Slab breakoff: a model for Caledonian, Late Granite syn-collisional magmatism in the orthotectonic (metamorphic) zone of Scotland and Donegal, Ireland. Lithos, 62, 65-85.

BARROW, G. 1893. On an intrusion of muscovite biotite gneiss in the southeast Highlands of Scotland and its accompanying metamorphism. Quarterly Journal of the Geological Society of London, 49, 330-358.

BAU, M. 1996. Controls on the fractionation of isovalent trace elements in magmatic and aqueous systems: evidence from Y/Ho, Zr/ Hf, and lanthanide tetrad effect. Contributions to Mineralogy and Petrology, 123, 323-333.

BLUCK, B.J. 2001. Caledonian and related events in Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, 91, 375-404.

BROWN, G.C. 1979. Geochemical and geophysical constraints on the origin and the evolution of Caledonian granites. In: HARRIS, A.L., HOLLAND, C.H. & LEAKE, B.E. (eds) The Caledonides of the British Isles-Reviewed. Geological Society, London, Special Publications, 8, 645-652.

CAWOOD, P.A., NEMCHIN, A.A., SMITH, M. & LOEWY, S. 2003. Source of the Dalradian Supergroup constrained by U-Pb dating of detrital zircon and implications of the East Laurentian margin. Journal of the Geological Society. London, 160, 231-246.

CHAPPELL, B.W. & WHITE, A.J.R. 1974. Two contrasting granite types. Pacific Geology, 8, 173-174.

CHAPPELL, B.W. & WHITE, A.J.R. 1992. I- and S-type granites in the Lachlan Fold Belt. Transactions of the Royal Society of Edinburgh: Earth Sciences, 83, 1-26.

CLARKE, D.B. 1981. The mineralogy of peraluminous granites. Canadian Mineralogist, 19, 3-17.

CLAYBURN, J.A.P. 1988. The crustal evolution of Central Scotland and the nature of the lower crust: Pb, Nd and Sr isotope evidence from Caledonian granites. Earth and Planetary Science Letters, 90, 41-51.

DALRYMPLE, D.J. 1995. Contact anatexis of Dalradian metapelites from the Huntly-Knock area, Aberdeenshire, N.E. Scotland. PhD thesis, University of Manchester.

DALY, J.S. & MCLELLAND, J.M. 1991. Juvenile middle Proterozoic crust in the Adirondack Highlands, Grenville province, northeastern North America. Geology, 19, 119-122.

DEMPSTER, T.J. 1985. Uplift patterns and orogenic evolution in the Scottish Dalradian. Journal of the Geological Society, London, 142, 111-128.

DEMPSTER, T.J., ROGERS, G. & TANNER, P.W.G. ET AL. 2002. Timing and deposition, orogenesis and glaciation within the Dalradian rocks of Scotland: constraints from U-Pb ages. Journal of the Geological Society, London, 159, 83-94.

DEPAOLO, D.J. 1981. Neodymium isotopes in the Colorado Front range and crustmantle evolution in the Proterozoic. Nature, 291, 193- 196.

DEWEY, J.F. 1971. A model for the Lower Palaeozoic evolution of the southern margin of the early Caledonides of Scotland and Ireland. Scottish Journal of Geology, 7, 219-240.

DEWEY, J.F. & MANGE, M. 1999. Petrography of Ordovician and Silurian sediments in the western Irish Caledonides: tracers of a short-lived Ordovician continent-arc collision orogeny and the evolution of the Laurentian Appalachian-Caledonian margin. In: MACNIOCAILL, C. & RYAN, P.D. (eds) Continental Tectonics. Geological Society, London, Special Publications, 164, 55-107.

DEWEY, J.F. & SHACKLETON, R.J. 1984. A model for the evolution of the Grampian tract in the early Caledonides and Appalachians. Nature, 312, 115-120.

DEWEY, J.F. & STRACHAN, R.A. 2003. Changing Silurian-Devonian relative plate motion in the Caledonides; sinistral transpression to sinistral transtension. Journal of the Geological Society, London, 160, 219-229.

DICKIN, A.P. 1998. Nd isotopic mapping of a cryptic continental suture, Grenville Province of Ontario. Precambrian Research, 91, 433- 444.

DICKIN, A.P. 2000. Crustal formation in the Grenville Province: Nd-isotope evidence. Canadian Journal of Earth Sciences, 37, 165- 181.

DICKIN, A.P. & BOWES, D.R. 1991. Isotopic evidence for the extent of early Proterozoic basement in Scotland and northwest Ireland. Geological Magazine, 128, 385-388.

DICKIN, A.P. & HIGGINS, M.D. 1992. Sm/Nd evidence for a major 1.5 Ga crust-formation event in the central Grenville province. Geology, 20, 137-140.

DICKIN, A.P. & MCNUTT, R.H. 1989. Nd model age mapping of the southeast margin of the Archean Foreland in the Grenville Province of Ontario. Geology, 17, 299-301.

DRAUT, A., CLIFT, P.D., CHEW, D.M., HANNIGAN, R.E., LAYNE, G. & SHIMIZU, N. 2002. A model for continental crust genesis by arc accretion; rare earth element evidence from the Irish Caledonides. Earth and Planetary Science Letters, 203, 861-877.

DRAUT, A., CLIFT, P.D., CHEW, D.M., COOPER, M.J., TAYLOR, R.N. & HANNIGAN, R.E. 2004. Laurentian crustal recycling in the Ordovician Grampian Orogeny; Nd isotopic evidence from western Ireland. Geological Magazine, 141, 195-207.

DULSKI, P. 2001. Reference material for geochemical studies: new analytical data by ICP-MS and critical discussion of reference values. Geostandards Newsletter, 25, 87-125.

FLOWERDEW, M.J., DALY, J.S. & WHITEHOUSE, M.J. 2005. 470 Ma granitoid magnetism associated with the Grampian Orogeny in the Slishwood Division, NW Ireland. Journal of the Geological Society, London, 162, 563-575.

FOWLER, M.B. & HENNEY, P.J. 1996. Mixed Caledonian appinite magmas: Implications for lamprophyre fractionation and high Ba-Sr granite genesis. Contributions to Mineralogy and Petrology, 126, 199- 215.

FOWLER, M.B., HENNEY, P.J., DARBYSHIRE, D.P.F. & GREENWOOD, P.B. 2001. Petrogenesis of high Ba-Sr granites: the Rogart pluton, Sutherland. Journal of the Geological Society, London, 156, 1217- 1230.

FRIEDRICH, M.A., HODGES, K.V., BOWRING, S.A. & MARTIN, M.W. 1999. Geochronological constraints on the magmatic, metamorphic and thermal evolution of the Connemara Caledonides, western Ireland. Journal of the Geological Society. London, 156, 1217-1230.

FROST, C.D. & O'NioNS, R.K. 1985. Caledonian magma genesis and crystal recycling. Journal of Petrology, 26, 515-544.

GOULD, D. 1997. Geology of the Country around Inverurie and Alford. Memoir of the British Geological Survey, Sheets 76W and 76E (Scotland).

GRAHAM, C. 1986. Petrochemistry and tectonic setting of the Dalradian metabasic rocks of the SW Scottish Highlands. Journal of the Geological Society, London, 132, 61-84.

GROMET, P. & SILVER, L.T. 1987. REE variations across the Peninsular Ranges Batholith: Implications for batholithic petrogenesis and crustal growth in magmatic arcs. Journal of Petrology, 28, 75-125.

HALLIDAY, A.N. 1984. Coupled Sm-Nd and U-Pb systematics in late Caledonian granites and basement under northern Britain. Nature, 307, 229-233.

HALLIDAY, A.N., STEPHENS, W.E., HUNTER, R.H., MENZIES, M.A., DICKIN, A.P. & HAMILTON, P.J. 1985. Isotopic and chemical constraints on the building of the deep Scottish lithosphere. Scottish Journal of Geology, 21, 456-491.

HALLIDAY, A.N., GRAHAM, C.M., AFTALION, M. & DYMOKE, P. 1989. The depositional age of the Dalradian Supergroup: U-Pb and Sm-Nd isotopic studies of the Tayvallich Volcanics, Scotland. Journal of the Geological Society, London, 146, 3-6.

HAMILTON, P.J., O'NIONS, R.K. & PANKHURST, R.J. 1980. Isotopic evidence for the provenance of some Caledonian granites. Nature, 287, 279-284. HARMON, R.S., HALLIDAY, A.N., CLAYBURN, J.A.P. & STEPHENS, W.E. 1984. Chemical and isotope systematics of the Caledonian intrutions of Scotland in northern England: a guide to magma source regions and magma-crust interaction. Philosophical Transactions of the Royal Society of London, Series A, 310, 709- 742.

HARRISON, T.N. 1987. The granitoids of eastern Aberdeenshire. In: TREWIN, N.H., KNELLER, B.C. & GILLEN, C. (eds) Excursion Guide to the Geology of the Aberdeen Area. Scottish Academic Press, Edinburgh, 243-250.

HEGNER, E., RODDICK, J.C., FORTIER, S.M. & HULBERT, L. 1995. Nd, Sr, Pb, Ar, and O isotopic systematics of Sturgeon Lake kimberlite, Saskatchewan, Canada: constraints on emplacement, alteration, and source contamination. Contributions to Mineralogy and Petrology, 122, 212-222.

HENDERSON, P. 1984. Rare earth geochemistry. In: FYFE, W.S. (ed.) Developments in Geochemistry. Elsevier, Amsterdam, 2, 1-29.

HIGHTON, A.J., HYSLOP, E.K. & NOBLE, S.J. 1999. U-b geochronology of migmatisation in the Northern Central Highlands: evidence for pre- Caledonian (Neoproterozoic) tectonometamorphism in the Grampian block, Scotland. Journal of the Geological Society, London, 156, 1195-1204.

HILDRETH, W. & MOORBATH, S. 1988. Crustal contributions to arc magmatism in the Andes of central Chile. Contributions to Mineralogy and Petrology, 98, 455-489.

HUTCHINSON, A.R. & OLIVER, G.J.H. 1998. Garnet provenance studies, juxtaposition of Laurentian marginal terranes and timing of the Grampian Orogeny in Scotland. Journal of the Geological Society, London, 155, 541.

JAHN, B.M., WU, F., CAPDEVILA, R., MARTINEAU, F., ZHAO, Z.H. & WANG, Y.X. 2001. Highly evolved juvenile granites with tetrad REE patterns: the Woduhe and Baerzhe granites from the Great Xing'an Mountains in NE China. Lithos, 59, 171-198.

JOHNSON, T.E., HUDSON, N.F. & DROOP, G.T.R. 2001. Partial melting in the Inzie Head gneisses; the role of water and a petrogenetic grid in KFMASH applicable to anatectic pelitic migmatites. Journal of Metamorphic Geology, 19, 99-118.

JOHNSON, T.E., HUDSON, N.F.C. & DROOP, G.T.R. 2003. Evidence for the genetic granite-migmatite link in the Dalradian of NE Scotland. Journal of the Geological Society, London, 160, 447-457.

KAMO, S.L., GOWER, C.F. & KROGH, T.E. 1989. Birthday of the Iapetus Ocean? A precise U-Pb zircon and baddeleyite age for the Long Range dikes, southeast Labrador. Geology, 17, 602-605.

LEE, S.R., CHO, M., CHEONG, C.S., KIM, H. & WINGATE, M.T.D. 2003. Age, geochemistry, and tectonic significance of Neoproterozoic alkaline granitoids in the northwestern margin of the Gyeonggi massif, South Korea. Precambrian Research, 122, 297-310.

LEMAITRE, R.W. 2002. Igneous Rocks-a Classification and Glossary of Terms. Cambridge University Press, Cambridge.

MARCANTONIO, F., DICKIN, A.P., MCNUTT, R.H. & HEAMAN, L.M. 1988. A 1800-million-year-old Proterozoic gneiss terrane in Islay with implications for the crustal structure and evolution of Britain. Nature, 335, 62-64.

MCDONOUGH, W.F. & SUN, S.S. 1995. The composition of the Earth. Chemical Geology, 120, 223-253.

MCLENNAN, S.M. 1994. Rare earth element geochemistry and the 'tetrad' effect. Geochimica et Cosmochimica Acta, 58, 2025-2033.

MCMILLAN, A., HARRIS, N.B.W., HOLNESS, M., ASHWAL, L., KELLEY, S. & RAMBELOSON, R. 2003. A granite-gabbro complex from Madagascar: constraints on melting of the lower crust. Contributions to Mineralogy and Petrology, 145, 585-599.

MORRIS, J.D. & RYAN, J.G. 2003. Subduction zone processes and implications for changing composition of the upper and lower mantle. In: HOLLAND, H.D. & TUREKIAN, K.K. (eds) The Mantle and Core. Pergamon, Oxford, 2, 451-470.

MUIR, R.J., FITCHES, W.R. & MALTMAN, A.J. 1994. The Rhinns Complex: Proterozoic basement on Islay and Colonsay, Inner Hebrides, Scotland, and on Inishtrahull, NW Ireland. Transactions of the Royal Society of Edinburgh: Earth Sciences, 85, 77-90.

NOBLE, S.R., HYSLOP, E.K. & HIGHTON, A.J. 1996. High-precision U- Pb monazite geochronology of the c. 806 Ma Grampian Shear Zone of the Central Highlands of Scotland. Journal of the Geological Society, London, 153, 511-514.

OLIVER, G.J.H. 2001. Reconstruction of the Grampian episode in Scotland: its place in the Caledonian Orogeny. Tectonophysics, 332, 23-49.

OLIVER, G.J.H. 2002. Chronology and terrane assemblage, new and old controversies. In: TREWIN, N.H. (ed.) The Geology of Scotland, 4th. Geological Society, London, 201-211.

OLIVER, G.J.H., CHEN, F., BUCHWALDT, R. & HEGNER, E. 2000. Fast tectonome-tamorphism and exhumation in the type area of the Harrovian and Buchan zones. Geology, 28, 459-462.

OLIVER, G.J.H., WILDE, S.A. & WAN, Y. 2008. Geochronology and geodynamics of Scottish granitoids from the late Neoproterozoic break-up of Rodinia to Palaeozoic collision. Journal of the Geological Society, London, 165, 661-674.

O'NIONS, R.K., HAMILTON, P.J. & HOOKER, P.J. 1983. A Nd isotope investigation of sediments related to crustal development in the British Isles. Earth and Planetary Science Letters, 63, 229-240.

PANKHURST, R.J. 1974. Rb-Sr whole-rock chronology of Caledonian events in Northeast Scotland. Geological Society of America Bulletin, 85, 345-350.

PANKHURST, R.J. & PIDGEON, R.T. 1976. Inherited isotope systems in the source region pre-history of early Caledonian granites in the Dalradian Series of Scotland. Earth and Planetary Science Letters, 31, 55-68.

PIDGEON, R.T. & AFTALION, M. 1978. Cogenetic and inherited zircon U-Pb systems in granites: Palaeozoic granites of Scotland and England. Geological Journal, Special Issue, 10, 183-220.

PLANT, J.A. 1986. Models for granites and their mineralising systems in the British and Irish Caledonides. In: ANDREW, C.J. (ed.) Geology and Genesis of Mineral Deposits in Ireland. Irish Association for Economic Geology, Dublin, 121-156.

RAPP, R.P. & WATSON, E.B. 1995. Dehydration melting of metabasalt at 8-32 kbar: Implications for continental growth and crystal recycling. Journal of Petrology, 36, 891-931.

READ, H.H. 1961. Aspects of Caledonian magmatism in Scotland. Proceedings of the Liverpool and Manchester Geological Society, 2, 653-683.

RICKWOOD, P.C. 1989. Boundary lines within petrologic diagrams, which use oxides of major and minor elements. Lithos, 22, 247-263.

ROGERS, G., DEMPSTER, T.J., BLUCK, BJ. & TANNER, P.W.G. 1989. A high-precision age for the Ben Vuirich granite: implications for the evolution of the Dalradian Supergroup. Journal of the Geological Society, London, 146, 789-798.

SOPER, N.J. 1986. The Newer Granile problem: a geotectonic view. Geological Magazine, 123, 227-236.

SOPER, N.J. & BUTTON, D.H.W. 1984. Late Caledonian sinistral displacements in Britain: Implications for a three-plate model. Tectonics, 3, 781-794.

SOPER, N.J., STRACHAN, R.A., HOLDSWORTH, R.E., GAYER, R.A. & GREILING, R.O. 1992. Sinistral transpression and the Silurian closure of the Iapetus. Journal of the Geological Society, London, 149, 871-880.

SOPER, N.J., RYAN, P.D. & DEWEY, J.F. 1999. Age of the Grampian Orogeny in Scotland and Ireland. Journal of the Geological Society, London, 156, 1231-1236.

STEPHENS, W.E. & CALDER, A. 2004. Analysis of non-organic elements in plant foliage using polarised X-ray fluorescence spectrometry. Analytica Chimica Acta, 527, 89-96.

STEPHENS, W.E. & HALLIDAY, A.N. 1984. Geochemical contrasts between late Caledonian granitoid plutons of northern, central and southern Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, 75, 259-273.

STEPHENSON, D. & GOULD, D. 1995. The Grampian Highlands. British Regional Geology, HMSO, London, 115-148.

STRACHAN, R.A., SMITH, M., HARRIS, A.L. & FETTES, D.J. 2002. The Northern Highland and Grampian terranes. In: TREWIN, N.H. (ed.) The Geology of Scotland, 4th. Geological Society, London, 81-147.

STRECKEISEN, A.L. 1976. To each plutonic rock its proper name. Earth-Science Reviews, 12, 1-33.

TANNER, P.W.G. 1996. Significance of the early fabric in the contact metamorphic aureole of the 590 Ma Ben Vuirich Granite, Perthshire, Scotland. Geological Magazine, 133, 683-695.

TANNER, P.W.G., LESLIE, A.C. & GILLESPIE, M.R. 2006. Structural setting and petrogenesis of the Ben Vuirich Granite Pluton of the Grampian Highlands: a pre-orogenic, rift-related intrusion. Scottish Jounal of Geology, 42, 113-136.

THIRLWALL, M.F. 1982. Systematic variation in chemistry and Nd- Sr isotopes across a Caledonian calc-alkaline volcanic arc: implications for source materials. Earth and Planetary Science Letters, 58, 27-50.

THIRLWALL, M.F. 1988. Geochronology of the Late Caledonian magmatism in northern Britain. Journal of the Geological Society, London, 145, 951-967.

VAN BREEMEN, O. & BLUCK, B.J. 1981. Episodic granite plutonism in the Scottish Caledonides. Nature, 291, 113-117.

WHALEN, B.J., CURRIE, K.L. & CHAPPELL, B.W. 1987. A-type granites: geochemical characteristics, discrimination and petrogenesis. Contributions to Mineralogy and Petrology, 95, 407- 419.

WASSERBURG, G.J., JACOBSEN, S.B., DEPAOLO, D.J. & MCCULLOCH, M.T. 1981. Precise determination of Sm/Nd ratios, Sm and Nd isotopic abundances in standard solutions. Geochimica et Cosmochimica Acta, 45, 2311-2323.

ZEN, E. 1986. Aluminium enrichment in silicate melts by fractional crystallization: some mineralogic and petrographic constraints. Journal of Petrology, 27, 1095-1117.

Received 17 July 2007; revised typescript accepted 22 February 2008.

Scientific editing by Rob Strachan


1 Department fur Geo- und Umweltwissenschaften, Universitat Munchen, Theresienstr. 41, D-80333 Munchen, Germany

2 Present address: Institut fur Mineralogie, Universitat Hannover, Callinstr. 3, D-30167 Hannover, Germany

(e-mail: [email protected])

3 Crustal Geodynamics Group, School of Geography and Geosciences, University of St. Andrews, Irving Building, St. Andrews KY16 9AL, UK

Copyright Geological Society Publishing House Jul 2008

(c) 2008 Journal of the Geological Society. Provided by ProQuest Information and Learning. All rights Reserved.