Journal of the Geological Society, London, Vol. 167, 2010, pp. 49–60. doi: 10.1144/0016-76492009-028.

Pulsed emplacement of the Mount Kinabalu granite, northern

M. COTTAM1*, R. HALL1,C.SPERBER1 & R. ARMSTRONG2 1SE Asia Research Group, Department of Earth Sciences, Royal Holloway University of London, Egham TW20 0EX, UK 2PRISE, Research School of Earth Sciences, The Australian National University, Canberra, ACT 0200, Australia *Corresponding author (e-mail: [email protected])

Abstract: High-precision U–Pb ion microprobe analyses provide new constraints on the emplacement and origin of the Kinabalu granite in , northern Borneo. The granite is a sheeted laccolith-like body comprising dyke-fed granitic units that young downwards, each emplaced beneath the previous sheet. Analyses of concentric growth zones in zircons indicate crystallization between 7.85 Æ 0.08 and 7.22 Æ 0.07 Ma, and show that the entire pluton was emplaced and crystallized within less than 800 ka. Several pulses of magmatism are recognized, each lasting for a maximum of 250 ka, and possibly as briefly as 30 ka. The oldest ages coincide with the highest elevations whereas the youngest ages are found at lower elevations around the edge of the body. Based on these new age data and field observations we identify the biotite granodiorite, hornblende granite and porphyritic facies as the Upper, Middle and Lower Units respectively. Inherited zircon ages indicate different protoliths for the Upper and Middle Units. The Upper Unit is derived from attenuated continental crust of the South China margin subducted beneath Sabah. The Middle Unit is sourced from melting of the crystalline basement in Sabah with little or no contribution from South China crust.

Supplementary material: Full U–Pb ion microprobe analytical data, and modal and major element composition data are available at http://www.geolsoc.org.uk/SUP18385.

Granite magmatism plays a fundamental role in the growth topography restrict geological observations below 3000 m. The and differentiation of the continental crust but there is still northeastern area of the pluton remains extremely inaccessible significant controversy about the generation and intrusion of because of thick rainforest cover, lack of roads and trails, and granitic magmas in the crust. In particular, the physical difficulty of the terrain (Connaughton 1996). processes of magma ascent and emplacement have been the Controversy still surrounds the origin and emplacement of subjects of considerable recent debate (e.g. McCaffrey & the Kinabalu Pluton. Amphibole geobarometry suggests intru- Petford 1997; Petford & Clemens 2000; Petford et al. 2000). sion at depths of between 3 and 10 km (Vogt & Flower Early ideas of granite diapirism (e.g. Mrazec 1915; Nicolesco 1989). Previous workers have proposed various models related 1929) are being substantially revised by studies employing to southward subduction of the proto-South China Sea beneath present-day geophysical and geochronological techniques (e.g. Borneo (Vogt & Flower 1989), melting during Miocene Evans et al. 1993; Michel et al. 2008). A growing body of isostatic rebound after subduction (Swauger et al. 2000), or evidence suggests that many plutons are emplaced at shallow crustal anatexis during post-subduction relaxation (Hutchison levels as tabular laccoliths (e.g. Vigneresse 1990; Evans et al. et al. 2000; Chiang 2002). Evaluating such models depends 1993; McCaffrey & Petford 1997) that grow incrementally in in part on the exact age of the body. Jacobson (1970) first a number of discrete magmatic pulses over a range of time estimated the age of the Kinabalu granite to be c. 9 Ma, scales (e.g. Coleman et al. 2004; Matzel et al. 2006; Michel based on K–Ar dating of biotite and hornblende. Subsequent et al. 2008). Single pulses are interpreted to be injected either K–Ar dating produced a wider spread of ages, ranging from as dyke-fed horizontal sheets (e.g. Evans et al. 1993; Michel 13.7 to 1.3 Ma (e.g. Rangin et al. 1990; Bellon & Rangin et al. 2008) or as subvertical dykes (e.g. Glazner et al. 2004) 1991; Hutchison et al. 2000; Swauger et al. 2000). Swauger that subsequently inflate. et al. (2000) suggested that the granite was formed by a The Kinabalu pluton provides an opportunity to test models tectonic event at 15 Ma that also led to the formation of the of granite emplacement. It is exposed over an area of c. important offshore unconformity known as the Deep Regional 120 km2 and forms the peak of Mt Kinabalu (c. 4100 m), Unconformity (e.g. Hutchison 1996; Sandal 1996). Similar U– situated at the northern end of the Crocker Ranges in the Pb ages (13–15 Ma) from the Capoas granite along-strike in Malaysian state of Sabah (Fig. 1). This collisional mountain Palawan (Encarnacion & Mukasa 1997) may indicate that range results from the Eocene to Early Miocene subduction melting was more widespread and linked to a regional of the proto-South China Sea beneath northern Borneo (Hall tectonic event. & Wilson 2000; Hutchison et al. 2000). The Mt Kinabalu This study presents the results of secondary ionization mass granite is the only pluton to intrude the deformed rocks of spectrometry (SIMS) U–Pb analyses on zircons from nine the Crocker Ranges. Most of the granite body is extremely samples of the Kinabalu granite. Based on these new data we well exposed on the relatively flat western and eastern summit suggest that the entire Kinabalu pluton was emplaced in less than plateaux of Mt Kinabalu (both c. 4000 m), well above the 800 ka as a series of dyke-fed subhorizontal granite sheets regional tree line at c. 2500 m. Rainforest cover and steep between 8 and 7 Ma, intruded in several pulses.

49 50 M. COTTAM ET AL.

Fig. 1. (a) Location map showing the position of Borneo within SE Asia. (b) Simplified outline map of Borneo showing the position of Mt Kinabalu, at the northern end of the Crocker Ranges, which form part of the Central Borneo Mountains (dark grey, elevation .1000 m; white, elevation .1500 m). Black dotted lines represent international boundaries between Malaysian Borneo (Sabah and Sarawak), Indonesian Borneo (Kalimantan) and Brunei (B).

Geological background as crystalline basement (Reinhard & Wenk 1951). The ultrabasic Country rocks rocks form part of a Middle Jurassic to Early Cretaceous ophiolite (Hutchison 2005) emplaced in the Late Cretaceous or The Kinabalu granite is intruded into Mesozoic igneous and Early Palaeogene (Newton-Smith 1967; Omang & Barber 1996). metamorphic rocks and their Cenozoic sedimentary cover (Fig. Crystalline basement rocks are Triassic to Cretaceous in age 2). Basement rocks in the Kinabalu area include variably (Reinhard & Wenk 1951; Dhonau & Hutchison 1966; Koopmans serpentinized peridotites and small exposures of rocks described 1967; Kirk 1968; Leong 1974), and most resemble deformed

Fig. 2. Simplified geological map of the Mt Kinabalu region, modified from Jacobson (1970). Line X–Y shows line of cross- section reproduced in Figure 5. PULSED EMPLACEMENT OF THE MT KINABALU GRANITE 51 ophiolitic rocks intruded by arc plutonic rocks, suggesting an rocks of the Kinabalu pluton have been described at different origin in a Mesozoic intra-oceanic arc (Hall & Wilson 2000). times as quartz monzonite, adamellite, granodiorite and granite Imai & Ozawa (1991) described unusual peridotites that are reflecting changes in classification and nomenclature, and differ- exposed close to Mt Kinabalu and suggested that they represent ent observations (Reinhard & Wenk 1951; Kirk 1968; Jacobson subcontinental mantle. 1970; Kasama et al. 1970; Vogt & Flower 1989). We recognize Throughout Sabah basement rocks are observed in faulted three distinct granitic lithologies: a biotite granodiorite, a contact with steeply dipping sedimentary cover sequences, hornblende granite and a porphyritic hornblende granite. Based predominantly deep-water turbidites and related deposits as- on new age data and field observations we term these the Upper, signed to the Eocene to Lower Miocene Trusmadi and Crocker Middle and Lower Units respectively. Formations (Collenette 1965). All of these rocks were folded and Reinhard & Wenk (1951) gave the first detailed petrological faulted during the Early Miocene Sabah orogeny (Hutchison account with a small number of chemical and modal analyses 1996) when the extended passive continental margin of South and reported that the rocks belonged to ‘the family of quartz- China collided with north Borneo (Hall & Wilson 2000; monzonites’. Kirk (1968) preferred the term adamellite from Hutchison et al. 2000) and subduction terminated. Much of the Nockolds’ (1954) classification scheme. Jacobson (1970) com- area then became emergent but most of present-day Sabah pleted the most extensive study of Mt Kinabalu and considered subsided again below sea level in the Late Early Miocene (Hall the dominant lithology to be hornblende adamellite using the et al. 2008). Neogene sedimentary rocks rest unconformably on classification of Bateman et al. (1963) based on visual estimates folded Eocene to Early Miocene rocks across Sabah. Fluviatile of modes. Vogt & Flower (1989) carried out a mineralogical and and marginal marine sediments were deposited across large areas chemical study using Jacobson’s samples. They described Kina- of south Sabah (Balaguru et al. 2003; Balaguru & Nichols 2004) balu as a batholith that ‘comprises variably phyric hornblende during the Middle and Late Miocene. Offshore of northern Sabah quartz monzonite and biotite quartz monzodiorite’ without shelf sequences pass northwards into deep marine sediments specifying their classification scheme, and presented modes for (Ingram et al. 2004; Franke et al. 2008), and the shelf edge 26 samples. Kasama et al. (1970) presented modes for 15 moved progressively north from the Middle Miocene (e.g. samples described as granites, granodiorites and porphyritic Hazebroek & Tan 1993; Sandal 1996). Onland in northern Sabah adamellites. marine clastic sediments are interpreted to have been deposited, In this study point counting was carried out on 36 granitoid but have now been removed by erosion after the region became samples and six aplites, and 22 of the samples were stained for emergent once more in the Late Miocene or Early Pliocene (Hall potassium feldspar (Sperber 2009). Five porphyritic rocks con- 2002; Balaguru et al. 2003; Tongkul & Chang 2003; Morley & tain phenocrysts of K-feldspar that are very large compared with Back 2008). No younger sedimentary rocks are known from the the thin-section size. For two of these samples the modal Kinabalu area except Pleistocene glacial tills of the Pinosuk proportion of K-feldspar was corrected based on an estimate Gravels (Collenette 1958). from the hand specimen. According to our study, the Upper Unit samples are biotite granodiorites or biotite tonalites, whereas the Middle and Lower Unit samples are almost all granites with a The Kinabalu granite few samples plotting as quartz monzonites or quartz syenite (Fig. The Kinabalu pluton has a roughly elliptical shape with a long 3c) using the classification scheme of Le Maitre (1989). Our axis oriented approximately NE–SW. The major axis is about modal plots are similar to those of Reinhard & Wenk (1951), 16 km and the minor axis about 10 km (Fig. 2). The igneous Kirk (1968) and Kasama et al. (1970) (Fig. 3b) whereas the Vogt

Fig. 3. QAP diagram (Le Maitre 1989) of the Kinabalu granite. (a) According to the data of Vogt & Flower (1989); (b) as proposed by Reinhard & Wenk (1951), Kirk (1968) and Kasama et al. (1970); (c) according to this study. 52 M. COTTAM ET AL.

& Flower (1989) modes appear to be more plagioclase-rich (Fig. microgranites including porphyritic and pyroxene-bearing vari- 3a), and although several would now be classified (Le Maitre eties (Jacobson 1970). 1989) as granites or granodiorites many plot as quartz monzo- Based on observations of steeply dipping contacts (Jacobson nites, quartz monzodiorites or quartz diorites. 1970) and apparent geochemical zonation (Vogt & Flower 1989), To investigate the differences we compared rock compositions previous workers have interpreted the granite as a composition- reported by Reinhard & Wenk (1951), Kirk (1968) and Vogt & ally zoned, steep-sided pluton consistent with diapiric models Flower (1989) with bulk-rock compositions calculated from their (e.g. Mrazec 1915; Nicolesco 1929). The relationships of the modes using mineral composition data from Vogt & Flower various granites are unusual. The Lower Unit has previously been (1989). We repeated this exercise for our own samples using our interpreted as a ‘shell’ or marginal facies surrounding the own rock and mineral analyses. Vogt & Flower (1989) reported 18 pluton’s ‘main body’ of hornblende granite (e.g. Jacobson 1970; samples for which there are modes and major element composi- Vogt & Flower 1989) despite the fact that it contains the largest tions and we have 11 samples. The comparisons are shown for crystals. We regard the ‘main body’ as the Middle Unit. Intrusive SiO2 and K2O in Figure 4. The Vogt & Flower (1989) observed and/or chilled contacts are observable on steep cliff sections and calculated compositions show a great deal of scatter (e.g. Fig. within the Middle Unit, identified by contrasts between light and 4a and c), in contrast to our compositions, which show very good dark hornblende granites. The nature of the contact of the biotite agreement for all major elements (e.g. Fig. 4b and d). The granodiorite (our Upper Unit) with the hornblende granites observed and calculated compositions for the small number of (Middle Unit) is not clear. Jacobson (1970) reported a narrow (c. samples from Reinhard & Wenk (1951) and Kirk (1968) are also 1 m) gradational contact between the Upper and Middle Units on very similar; Kasama et al. (1970) provided no chemical analyses. the western summit plateau without interpreting any relative age Vogt & Flower (1989) did not report their methods but we suspect relationship between them. Vogt & Flower (1989) made no new that they did not stain the sections, may have misidentified K- field observations but interpreted the observations of Jacobson of feldspars and quartz, and may not have corrected their modes for a narrow gradational contact and flow structures to suggest that the very large size of the phenocrysts in the porphyritic granites. the magmas were only partially crystallized during intrusion. The major lithologies are granites not monzonites according to Based on a geochemical model they concluded that the biotite current classification schemes (Le Maitre 1989). granodiorite was the youngest part of the body. However, we The Upper Unit is referred to here as a biotite granodiorite, crossed the boundary in several places and saw no clear margin. although it varies from biotite–hornblende tonalite to granodior- The Middle and Lower Unit hornblende granites are magnesian, ite. It is exposed in a small area (about 2 km2) coinciding with predominantly calc-alkalic and metaluminous (e.g. Frost et al. the highest elevations of the western plateau. It contains 2001). Upper Unit biotite granodiorite samples are peraluminous. abundant biotite, which appears texturally to be in equilibrium All resemble most closely the Cordilleran granitoids of Frost et al. with less common hornblende, although in some samples (2001). The Upper Unit rocks have been termed volcanic arc hornblende is replaced by biotite. The Middle Unit forms most granites (Pearce et al. 1984), island arc and continental arc of the exposed body and is a hornblende granite. Hornblende granitoids (Maniar & Piccoli 1989), or amphibole-bearing calc- occurs as small phenocrysts texturally in equilibrium with minor alkalic granites (Barbarin 1999). Samples plot as volcanic arc biotite and potassium feldspar, plagioclase feldspar and quartz. granites on the Y, Nb and Rb plots of Pearce et al. (1984); the The colour varies from darker to lighter grey, which corresponds evolved samples (aplites and some porphyritic samples) plot in to the abundance of hornblende. The Lower Unit is a porphyritic the syncollisional granite field. Several workers have suggested hornblende granite that is megacrystic in places. It is chemically that trace element classifications reflect the melt protolith rather similar to the Middle Unit but contains large potassium feldspar than the tectonic setting in which melting occurred (Arculus phenocrysts (2–3 cm in length) and megacryst-rich zones. The 1987; Twist & Harmer 1987; Roberts & Clemens 1993; Frost et Lower Unit is exposed in a band running along the southern and al. 2001). Least-squares modelling confirms the suggestion (Vogt western edges of the pluton, varying in width from 0.5 to 1.7 km & Flower 1989) that the upper biotite granodiorite cannot be (Fig. 2). derived from the hornblende granite by closed-system fractional Each of the three units is intruded by dykes and the number of crystallization alone. The other granites can be related by dykes varies from place to place. Field observations indicate fractional crystallization involving K-feldspar, plagioclase, quartz, there are two broad types: abundant aplites and a range of hornblende, magnetite and biotite.

Fig. 4. Comparison of observed and calculated values for SiO2 and K2O for Kinabalu granites using data reported by Vogt & Flower (1989) and this study. The major element compositions for the rocks were calculated from the modes and mineral analyses. PULSED EMPLACEMENT OF THE MT KINABALU GRANITE 53 Analytical methods interpreted to represent the magmatic age of the zircons and thus Zircon separates from nine samples were analysed by sensitive the crystallization age of the Kinabalu pluton. The results of high-resolution ion microprobe (SHRIMP) on the SHRIMP II (at these analyses are plotted as isotopic compositions, uncorrected the Australian National University: one sample from the Upper for common Pb, in the form of Tera–Wasserburg concordia Unit, five from the Middle Unit, and three from the Lower Unit; diagrams (Tera & Wasserburg 1972) in Figs 5 and 6. Mean Table 1). Three of the five Middle Unit samples were collected sample ages (Table 1) were calculated using the SQUID 1.03 and from the western summit plateau (two hornblende granites and an Isoplot/Ex 2.29 programs of Ludwig (2001a,b). Data points aplite dyke), and two from the eastern plateau. The three Lower plotting off a regression line from common lead are not included Unit samples include two megacrystic samples and an associated in mean age calculations. pyroxene microgranite dyke. Samples were crushed and graded using disposable nylon cloth sieves in a brass holder. Zircon Upper Unit: biotite granodiorite separates were prepared from the resulting 100–250 ìm fraction using standard heavy liquid and Frantz isodynamic separation Thirty-one spot analyses were made on 24 grains from a biotite techniques. High-purity zircon separates were then hand picked granodiorite sample CS-021. Twenty-five analyses form a coher- and the grains were mounted in epoxy resin, alongside the Temora ent group interpreted as reflecting the magmatic age. Six zircon standard (Black et al. 2003) and the SL13 standard analyses yield ages that are clearly inherited, being more than (Claoue´-Long et al. 1995) of the Research School of Earth twice the mean magmatic age. Twenty-four of the magmatic ages Sciences, The Australian National University. The grains were plot close to concordia, elongated along the mixing line to then sectioned and polished until exposed through their midsec- common Pb. This group defines a mean magmatic age of 7.85 Æ tions and Au coated. The internal zonation and structure of single 0.08 Ma (Fig. 5a). A single analysis with an abnormal uranium grains were mapped using cathodoluminescence and reflected content plots off the common lead mixing line and is not light images, allowing spot analyses to be targeted on grain areas included in magmatic age calculations. free of cracks and inclusions. U–Pb analyses were performed following the analytical Middle Unit: hornblende granite, Western plateau procedures outlined by Williams (1998). Data reduction and all age calculations were achieved using the SQUID 1.03 and The two hornblende granite samples from the western plateau Isoplot/Ex 2.29 programs of Ludwig (2001a,b). Uranium and (CS-018 and CS-020) and an aplite dyke (CS-059) were thorium concentrations were calibrated against the SL-13 stan- collected close to the mapped contact with the Upper Unit. dard. Young ages are assumed to be concordant, and were Thirty-one spot analyses on 23 grains from CS-018 yield 20 determined solely using the 238U/206Pb ratio; common Pb was coherent magmatic ages. Excluding an outlier that plots off the estimated using the 206Pb/207Pb ratio. This method has been common Pb mixing line (Fig. 5b), and one analysis rejected on shown to be very reliable for analyses that have low inherent the basis of an anomalously high uranium content, 18 data points common Pb and lie close to the concordia (e.g. Muir et al. define a mean magmatic age of 7.64 Æ 0.11 Ma. There are 11 1994), as is the case here. Rare Precambrian crystals and cores inherited ages. Nineteen analyses from CS-020, sampling 17 were corrected for common Pb using the 204Pb/206Pb ratio, and crystals, yield 17 coherent magmatic ages. Two analyses yield the age was based on the 206Pb/207Pb ratio. Uncertainties in inherited ages. Discarding a single analysis with abnormally high isotopic ratios and ages (including data tables and error bars for uranium, the remaining 16 magmatic analyses are tightly clus- plotted data) are reported at the 1ó level. Final, weighted mean tered (Fig. 5c), defining a mean age of 7.69 Æ 0.07 Ma. Twenty- ages are reported as 95% confidence limits. six grains were analysed from the aplite dyke (CS-059). Two analyses were discarded because they have abnormal uranium Magmatic ages contents. The 24 remaining analyses yield a mean magmatic age Analyses of concentric growth zones from all nine samples of 7.58 Æ 0.07 Ma (Fig. 5d). No inherited ages were observed for reveal tightly clustered Late Miocene ages. These ages are sample CS-059.

Table 1. Sample locations, descriptions and mean magmatic ages

Sample Longitude Latitude Elevation Lithology Spot Magmatic Age Error Æ number (decimal (decimal (m) analyses* analyses† (Ma)‡ (Ma)§ degrees) degrees)

Upper Unit CS-021 E 116.551 N 6.078 3827 Biotite granodiorite 31 25 7.85 0.08 Middle Unit CS-020 E 116.562 N 6.068 3847 Hornblende granite 19 16 7.69 0.07 CS-018 E 116.567 N 6.065 4096 Hornblende granite 31 19 7.64 0.11 CS-059 E 116.560 N 6.071 3980 Aplite dyke 26 26 7.58 0.07 CS-070 E 116.583 N 6.097 3527 Hornblende granite 18 18 7.44 0.09 CS-072 E 116.579 N 6.074 4007 Hornblende granite 27 22 7.46 0.08 Lower Unit CS-027 E 116.565 N 6.056 3096 Porphyritic hb granite 21 21 7.32 0.09 CS-054 E 116.598 N 6.056 2125 Porphyritic hb granite 32 17 7.22 0.07 CS-007 E 116.563 N 6.048 2888 Microgranite dyke 18 13 7.38 0.08

* Total number of spot analyses made on sample. † Number of analyses giving magmatic ages. ‡ Mean age of all accepted magmatic analyses. § Errors are 1ó. Longitude and latitude data are based on global positioning system (GPS) measurements. Altitude data are based on position within a detailed digital elevation model (DEM). 54 M. COTTAM ET AL.

Fig. 5. Tera–Wasserburg concordia diagrams (Tera & Wasserburg 1972) showing zircon SHRIMP U–Pb analyses for samples from the Upper Unit (a) and Middle Unit (b–f). Data-point error ellipses are 68.3 confidence. Grey ellipses reflect data points plotting off the mixing line to common Pb, which are not included in mean magmatic age calculations. Analyses with anomalously high uranium contents (not included in mean magmatic age calculations) are not shown. Uncertainties are 1ó; weighted mean ages are reported as 95% confidence limits.

age data (Fig. 6a) giving a mean age of 7.32 0.09 Ma. No Middle Unit: hornblende granite, Eastern plateau Æ inherited ages were observed. Twenty-three spot analyses on a Two hornblende granite samples (CS-070 and CS-072) were total of 20 zircons from the second sample, CS-054, give 17 collected from the mountain’s eastern plateau. Eighteen spot magmatic ages and yield a mean age of 7.22 Æ 0.07 Ma (Fig. analyses from different grains yield exclusively magmatic ages 6b). There are six inherited ages. Eighteen spot analyses were for sample CS-070. The data form a coherent group (Fig. 5e) made on 11 grains from a microgranite dyke (CS-007). Two close to concordia, with a single analysis that plots further along analyses yield inherited ages. Sixteen analyses form a coherent the mixing line with common lead. These analyses have a mean group of magmatic ages. Two analyses plot off the mixing line to age of 7.44 Æ 0.09 Ma. Twenty-seven spot analyses on 22 zircons common Pb; a third has anomalously high uranium content. from sample CS-072 give 23 magmatic ages. Twenty-one These analyses are not included in calculations of mean mag- analyses define a continuous array towards common Pb (Fig. 5f) matic age. The remaining 13 data points give a mean age of 7.38 that yields a mean age of 7.46 Æ 0.08 Ma. Two data points with Æ 0.08 Ma (Fig. 6c). anomalously high uranium contents are not included in age calculations. Four analyses yield inherited ages. Crustal inheritance patterns: ages of zircon cores Lower Unit: porphyritic hornblende granite All three lithological units contain inherited zircon that forms a Three samples from the Lower Unit were dated. Two are small but significant age population (Table 2). Thirty-one of the porphyritic hornblende granites (CS-027 and CS-054) and the 214 spot analyses give ages that are unquestionably inherited, third is a microgranite dyke (CS-007) intruding country rock being greater than twice the magmatic age. All represent close to the south margin of the pluton. Twenty-one spot analyses analyses of crystal cores. The small number is a true reflection on 20 crystals from sample CS-027 all yield coherent magmatic of the proportion of older cores. Overall, there is a strong peak PULSED EMPLACEMENT OF THE MT KINABALU GRANITE 55

of Mesozoic ages, accounting for 67% of all inherited ages, dominated by 16 Early Cretaceous ages.

Upper Unit: biotite granodiorite The Upper Unit biotite granodiorite (CS-021) contains the oldest inherited ages found in any of the Kinabalu samples (Table 2) and they are very different from those for the Middle and Lower Units. This sample contains two zircons with Palaeoproterozoic ages (1658 Æ 18 and 1896 Æ 20 Ma). There is one Permian (289 Æ 4 Ma), one Early Triassic (242 Æ 3 Ma) and one Early Jurassic age (179 Æ 3 Ma). There are no Early Cretaceous ages. A single sector-zoned grain has an Eocene (49 Æ 1 Ma) core with a younger magmatic rim.

Middle Unit: hornblende granite, Western plateau The two hornblende granites from the western plateau (CS-018 and CS-020) and the aplite dyke (CS-059) contain 15 inherited ages (Table 2). There are two Jurassic (186 Æ 2 Ma, 173 Æ 2 Ma) ages, eight Early Cretaceous (138 Æ 2 to 102 Æ 1 Ma) one Late Cretaceous (68 Æ 1 Ma), and two Cenozoic ages (41 Æ 1 and 25 Æ 1 Ma). There are no ages older than Jurassic.

Middle Unit: hornblende granite, Eastern plateau Only one of the two hornblende granite samples from the mountain’s eastern plateau yielded inherited cores (CS-072); CS- 070 was not examined for inherited grains because of limitations on the machine time available. There are four ages from CS-072: three are Early Cretaceous (142 Æ 1 to 114 Æ 1 Ma) and one is Eocene (43 Æ 1 Ma) (Table 2). Again, there are no ages older than Jurassic.

Lower Unit: porphyritic hornblende granite Because of time constraints, only one sample from the Lower Unit was searched for inherited cores. The porphyritic hornble- nde granite (CS-054) contains six inherited ages (Table 2). There is a single Neoproterozoic (970 Æ 11 Ma) age, one Ordovician (469 Æ 7 Ma), one Jurassic (182 Æ 2 Ma), and three Early Cre- taceous ages (130 Æ 2 to 105 Æ 1 Ma).

Discussion The U–Pb age data from zircons reported here record the emplacement of the Kinabalu pluton and provide information on the crust beneath Kinabalu that was the source of the melt.

Age of the Kinabalu granite It has always been difficult to interpret the previous K–Ar ages obtained from the Kinabalu granite, which range between 13.7 and 1.3 Ma (e.g. Jacobson 1970; Rangin et al. 1990; Bellon & Rangin 1991; Hutchison et al. 2000; Swauger et al. 2000). The new ages presented here provide a precise age for the Kinabalu Fig. 6. Tera–Wasserburg concordia diagrams (Tera & Wasserburg 1972) pluton. They show that the entire body was intruded in a narrow showing zircon SHRIMP U–Pb analyses for samples from the Lower time interval between c. 7.85 and c. 7.22 Ma, and that there were Unit. Data-point error ellipses are 68.3 confidence. Grey ellipses reflect several separate pulses of magmatism. data points plotting off the mixing line to common Pb, which are not The Upper Unit is not the youngest part of the pluton as included in mean magmatic age calculations. Analyses with anomalously suggested by Vogt & Flower (1989) but is in fact the oldest high uranium contents (not included in mean magmatic age calculations) lithology with a mean magmatic age of 7.85 Æ 0.08 Ma. This is are not shown. Uncertainties are 1ó; weighted mean ages are reported as the first recorded magmatic episode (Pulse 1). 95% confidence limits. Zircons from the Middle Unit range in age from 7.69 Æ 0.07 56 M. COTTAM ET AL.

Table 2. Ion microprobe U–Pb zircon analyses of inherited zircons

Sample spot U (ppm) Th (ppm) 204/206 % error 207Pb/206Pb % error 206Pb/238U % error Age Error Æ (Ma)* (Ma)†

Upper Unit CS021-1.2 282 146 0.000199 39.5 0.050394 1.8 0.093743 1.7 242 3 CS021-2.2 257 254 0.000052 43.2 0.124594 0.4 0.857335 2.0 1896 20 CS021-13.1 102 42 – – 0.045264 6.2 0.026295 1.7 180 3 CS021-17.2 1579 316 0.000076 99.0 0.049668 1.6 0.018429 4.1 49 1 CS021-19.1 61 25 0.001091 41.6 0.061051 3.3 0.099810 2.1 289 4 CS021-23.1 184 110 – – 0.112315 1.1 0.741135 1.3 1658 18 Middle Unit (western plateau) CS018-2.1 189 142 0.047755 3.5 0.034497 1.1 102 1 CS018-3.1 533 184 0.000217 49.5 0.053919 2.4 0.023424 1.8 68 1 CS018-4.1 153 141 0.000634 40.8 0.052761 3.5 0.037452 2.6 103 1 CS018-6.2 156 117 0.000593 37.8 0.054089 2.9 0.053495 1.5 138 2 CS018-7.2 93 89 – – 0.054909 4.5 0.036341 3.1 105 2 CS018-12.1 97 57 0.005684 47.6 0.080352 7.4 0.009213 2.9 25 1 CS018-18.1 267 325 0.000489 44.7 0.052408 4.1 0.015204 2.4 41 1 CS018-23.1 991 385 0.000023 99.0 0.050193 1.0 0.041208 6.9 123 2 CS018-10.2 198 130 0.000191 79.9 0.052104 2.4 0.042833 2.7 121 1 CS018-16.2 619 382 0.000039 99.0 0.049340 1.1 0.063382 1.6 173 2 CS018-20.2 441 338 0.000256 38.3 0.051025 1.3 0.056993 3.7 187 2 CS020-6.1 327 234 – – 0.048954 2.6 0.041453 1.8 113 1 CS020-7.2 1189 558 À0.000062 60.9 0.049565 1.2 0.053706 1.8 140 1 Middle Unit (eastern plateau) CS072-14.2 238 136 0.000196 56.6 0.052501 2.1 0.045437 2.6 142 1 CS072-16.2 170 135 0.000480 40.8 0.052439 2.6 0.040786 1.9 114 1 CS072-19.2 435 349 0.000158 72.5 0.050567 1.9 0.030833 3.0 121 1 CS072-22.1 418 231 0.000587 53.3 0.053572 2.7 0.016557 5.3 43 1 Lower Unit CS054-3.1 260 111 0.000072 54.1 0.056256 1.1 0.176941 1.5 469 7 CS054-4.1 83 37 0.001785 33.3 0.087591 3.3 0.042086 3.3 108 2 CS054-5.1 69 178 À0.000057 60.9 0.070595 1.3 0.376629 2.2 970 11 CS054-7.1 89 69 0.000793 35.4 0.052836 4.3 0.067108 1.2 182 2 CS054-8.1 233 115 0.000540 40.6 0.046909 2.9 0.037406 3.1 105 1 CS054-16.1 133 159 À0.000403 67.9 0.047509 3.5 0.047564 2.3 130 2 Aplite dyke CS007-6.2 70 61 0.000210 99.0 0.048417 5.6 0.044557 3.0 115 2 CS007-10.2 94 84 0.000599 78.6 0.047903 4.6 0.048319 1.9 122 2

* Based on 207Pb-corrected 206Pb/238U age. † Errors are 1ó. to 7.44 Æ 0.09 Ma. The two hornblende granite samples from the The porphyritic Lower Unit yields ages of 7.32 Æ 0.09 and western plateau have crystallization ages of 7.69 Æ 0.07 and 7.22 Æ 0.07 Ma. The microgranite dyke intruding country rock 7.64 Æ 0.11 Ma. An aplite dyke that intruded these granites gives an age of 7.38 Æ 0.08 Ma, within error of the two Lower yields an age of 7.58 Æ 0.07 Ma. These ages are interpreted to Unit ages. These three ages represent a further magmatic episode record one complete intrusive pulse terminating with a fractio- (Pulse 4). The younger ages of the Lower Unit rule out the nated aplitic end product (Pulse 2). Less than 3 km away, the two suggestion that it was emplaced first, as a shell into which the hornblende granite samples from the eastern plateau yield young- Middle and Upper Units were intruded (Jacobson 1970). Field er ages of 7.46 Æ 0.08 to 7.44 Æ 0.09 Ma. These ages may observations of porphyritic granite intruded locally as dykes represent a second intrusive pulse within the Middle Unit (Pulse within the Middle Unit (Jacobson 1970) are entirely consistent 3). However, given the uncertainties associated with these mean with its younger age. ages (1–1.5%) it is not possible to distinguish these two pulses based solely on age data. Recent studies have shown that zircon Duration and rates of emplacement growth in silicic magmas may predate eruption by several 100 ka, biasing U–Pb mean ages towards older ages (Simon et al. These data show that the Kinabalu granite was incrementally 2008). For the young ages reported here, even relatively short assembled by several discrete episodes of magmatism. The entire zircon residence times of c. 200 ka would significantly increase pluton was emplaced and crystallized within a period of less than errors associated with U–Pb mean ages and further preclude 800 ka. The ages from the Middle Unit provide some constraint distinction between pulses. on the duration of an individual magma pulse. The oldest Field observations suggest that further dating, with an in- hornblende granite on the western plateau has an age of creased spatial resolution and precision, would probably identify 7.69 Æ 0.07 Ma and the fractionated aplite intruding it has an additional intrusive pulses within the Middle Unit. Observations age of 7.58 Æ 0.07 Ma. Considering the associated errors, these of a chilled contact between light and dark hornblende granite ages constrain the duration of this pulse to less than 250 ka, and within the Middle Unit (e.g. Jacobson 1970; this study) support it could possibly have been as brief as 30 ka. multiple intrusive pulses. Complex cross-cutting dyke sets may Such time scales necessitate rapid magma ascent and emplace- reflect the fractionated end products of several intrusive pulses. ment via fracture propagation and dyking (e.g. Clemens & PULSED EMPLACEMENT OF THE MT KINABALU GRANITE 57

Mawer 1992; Petford 1996) rather than slower diapiric ascent known if the bottom of the granite is exposed in it. Tilting and (Miller & Paterson 1999). Field observations of sharp magmatic post-intrusive extensional faulting (Fig. 7) may have increased contacts suggest minimal interaction between the products of the apparent vertical thickness of the pluton. If the roof of the each pulse, implying that each was fully crystallized prior to granite were just above the present summit of the mountain, and subsequent pulses. the body is assumed to extend no deeper than the lowest observed outcrop on the flanks, the granite would have a thickness broadly consistent with the power-law prediction. Morphology and internal structure

The new data presented here are inconsistent with earlier Melt protoliths interpretations (Fig. 7) of the Kinabalu granite as a composition- ally zoned, steep-sided diapir-like pluton (e.g. Jacobson 1970; The inherited zircons in the Middle Unit are predominantly Early Vogt & Flower 1989). The oldest ages of the Upper Unit coincide Cretaceous and have no ages greater than Jurassic. In contrast, with the highest elevations of the western plateau, and the the Upper Unit contains the oldest zircon ages, which are youngest ages of the Lower Unit are found at the edges of the Proterozoic, and there are no Cretaceous zircon cores. The body at lower elevations. The age and field relationships can be different crustal inheritance patterns suggest a number of differ- explained if the Kinabalu pluton is a subhorizontally layered ent sources. body comprising multiple granite sheets (Fig. 7). This is The Proterozoic zircons within the Upper Unit necessitate an consistent with recent models of plutons (e.g. Evans et al. 1993; extremely old protolith. The thinned South China continental McCaffrey & Petford 1997; Petford & Clemens 2000), which crust subducted beneath northern Borneo during the Early have been shown to grow incrementally as a number of dyke-fed Miocene collision (Hall & Wilson 2000; Hutchison et al. 2000) horizontal sheets (e.g. Petford et al. 2000) that become younger is a potential source. Little is known of the basement age of this downwards (Michel et al. 2008). crust. It must have formed part of the South China margin before Recent geophysical studies have shown that many small Oligocene to Miocene opening of the South China Sea, and granite plutons are emplaced at shallow levels as tabular consequently is part of the Cathaysian South China block. A laccoliths (e.g. Vigneresse 1990; Evans et al. 1993; McCaffrey & Proterozoic basement is probable (Sewell et al. 2000; Li et al. Petford 1997; Ameglio & Vigneresse 1999; Vigneresse et al. 2001). Gravity data indicate that the Kinabalu pluton was 1999; Petford & Clemens 2000; Petford et al. 2000). Their intruded directly above the position of maximum crustal thick- dimensions may follow a power-law relationship (McCaffrey & ness in northern Borneo, reflecting the subduction of South Petford 1997). With a major axis of c. 16 km and a minor axis of China continental crust and collisional thickening (Hutchison et c. 10 km the power-law relationship predicts a maximum thick- al. 2000; Milsom & Holt 2001). Therefore we suggest that the ness for the Kinabalu granite of 1.82 km. Based on the observed Upper Unit represents the melting of the thinned South China height difference between the Kinabalu summit and the bottom continental crust deep beneath Kinabalu. of Low’s Gully, the deep chasm that cuts through the pluton, the There are two possible sources for the Mesozoic zircons that Kinabalu granite is at least 2 km thick. However, neither the roof dominate inheritance within the Middle Unit (Table 2). The nor the base of the Kinabalu body is seen. Low’s Gully remains subducted South China continental crust beneath Sabah is a dangerous and almost unvisited (Connaughton 1996) and it is not potential source. A belt of granitic rocks of Jurassic and Early

Fig. 7. Simplified cross-sections of the Mt Kinabalu region drawn along the line X–Y shown in Figure 2. (a) Redrawn from Jacobson (1970) based on original plutonic interpretations. (b) New interpretation of the Kinabalu granite as a sheeted laccolith- like body. 58 M. COTTAM ET AL.

Cretaceous age is known to have extended through SE China Regional mapping demonstrates that the Kinabalu granite is in (Jiang et al. 2006; Zhou et al. 2006) and potentially continued contact with serpentinized ultrabasic basement rocks along its SW towards the region involved in the collision. However, the western margin (Jacobson 1970). We suggest that the relatively absence of the Proterozoic ages similar to those of the Upper weak serpentinized rocks may have acted in a similar fashion to Unit suggests that the Middle Unit was generated without the shales, with ductile deformation restricting upward propagation involvement of South China continental crust. An alternative of feeder dykes and forcing lateral emplacement. Dehydration of source is basement rocks known from other parts of Sabah. Early serpentinites and the associated increase in pore fluid pressures Cretaceous ages have been reported from ophiolitic and probable could aid this process. Contact metamorphism at the Kinabalu arc plutonic rocks exposed in other parts of Sabah (Hall & contact has been observed in both ultrabasic and Crocker Wilson 2000; Hutchison 2005). This hypothesis is supported by Formation rocks. This may imply that the level at which the geochemical studies (Chiang 2002), which suggest generation of granite was emplaced was influenced by the thrust contact the Kinabalu hornblende granites from deep fluid-absent melting between the serpentinite and the Crocker Formation. of arc crust (Chiang 2002). Inherited Mesozoic core ages suggest that the Lower Unit has a similar source to the Middle Unit. The Neoproterozoic and Summary Palaeozoic ages hint at the involvement of older crustal material, The U–Pb age data reported here provide important constraints possibly as a result of mixing with previously intruded material on the emplacement and origin of the Kinabalu granite. Several or mixing of melts at depth. There are too few data to speculate separate pulses of intrusion can be recognized between further. 7.85 Æ 0.08 and 7.22 Æ 0.07 Ma. Magmatic pulses lasted no The different crystallization ages and sources for the Upper longer than 250 ka but could possibly be as short as 30 ka. The and Middle Units are consistent with the observation that these oldest ages coincide with the highest elevations; the youngest two units cannot be linked by closed-system fractional crystal- ages are found at lower elevations around the edge of the body. lization (Vogt & Flower 1989). The Upper Unit represents an Based on these age data and field observations the units older and distinct intrusive pulse (Pulse I) derived by melting of previously regarded as the central biotite granodiorite, main a different source. In contrast, the Middle and Lower Units can hornblende granite and marginal porphyritic facies are renamed be linked by fractionation. the Upper, Middle and Lower Units respectively. Finally, the Cenozoic ages in the Upper and Middle Units Our data suggest that the Kinabalu pluton is a sheeted suggest melting of arc rocks formed during Eocene to Early laccolith (Fig. 7) incrementally assembled over a period of less Miocene subduction of the proto-South China Sea. than 800 ka. These data add to a growing body of evidence that small granitic plutons are incrementally assembled over relatively Emplacement model short time scales by the intrusion of multiple granitic sheets that young downwards (e.g. McCaffrey & Petford 1997; Petford & The new data provide compelling evidence to reinterpret the Clemens 2000; Michel et al. 2008). Kinabalu granite as a sheeted laccolith (Fig. 7), comprising The Upper Unit biotite granodiorite is the oldest part of the multiple dyke-fed sheets. The Upper Unit was emplaced first, pluton and requires a source that includes old zircons, which is sourced by melting of the deep continental crust thrust beneath most likely to be subducted attenuated continental crust of the Sabah following orogenic thickening. Intrusion of the Upper Unit South China margin. Inherited zircon ages from the Middle and may have heated the middle crust, leading to further melting. Lower Units imply melting of the crystalline basement exposed The intra-oceanic arc crystalline basement at this level provided today in part of Sabah with little or no contribution from South the source for the hornblende granites of the Middle and Lower China crust. Cenozoic inherited zircon ages probably indicate Units, which were intruded sequentially beneath the Upper Unit. some contribution of arc material formed during subduction of The porphyritic nature of the Lower Unit is attributed here to the Proto-South China Sea. textural coarsening of feldspar crystals (e.g. Vernon 1986; The Late Miocene age of the Kinabalu pluton indicates that it Higgins 1999); large crystals coarsening at the expense of is not a subduction product (compare Vogt & Flower 1989), as dissolved smaller crystals. This phenomenon suggests that the subduction terminated in the Early Miocene (Hall & Wilson Lower Unit was intruded before the Middle Unit had fully 2000; Hutchison et al. 2000; Hall 2002). The granite melting cooled, with the remaining heat from the latter buffering the occurred more than 10 Ma after the Early Miocene collision. magma’s temperature at the K-feldspar liquidus for a substantial time (Higgins 1999). Migration of interstitial melt during this The industrial member companies of the SE Asia Research Group period, enhancing coarsening along fluid flow channels, may Consortium provided financial support for our work. The authors thank explain field observations of megacryst-rich zones within the J. Nias (Sabah Parks), F. Tongkul (Universiti Sabah) and the Lower Unit exposed near Mesilau (Fig. 1). Conversely, the staff of Sabah Parks for facilitating our work in the Kinabalu National absence of megacrysts within the Middle Unit suggests that the Park. M. Fanning, B. Armstrong, S. Mehrtens and G. Lister are thanked Upper Unit was fully cooled before subsequent emplacement. for assistance with SIMS U–Pb work at ANU. M. Thirlwall and A. Beard Sills and laccoliths emplaced at shallow depths have been assisted with geochemical analyses. C. Hutchison has provided helpful widely reported to preferentially intrude along shale horizons discussion over many years. (e.g. Corry 1988). The ductile response of such layers limits the ability of the feeder dykes to propagate upwards via vertical fracturing, forcing the magma to intrude sideways (Thomson & References Schofield 2008). Such effects may be enhanced by the volatiliza- Ameglio, L. & Vigneresse, J.L. 1999. Geophysical imaging of the shape of tion of pore fluids upon contact with the magma, dramatically granitic intrusions at depth: a review. In: Castro, A., Fernandez, C. & Vigneresse, increasing pore fluid pressures (Bjorlykke 1993), reducing the J.L. (eds) Understanding Granites: Integrating New and Classical Techniques. Geological Society, London, Special Publications, 168, effective strength of the rock and thus aiding horizontal intrusion 39–54. (Thomson & Schofield 2008). Arculus, R.J. 1987. The significance of source versus process in the tectonic PULSED EMPLACEMENT OF THE MT KINABALU GRANITE 59

controls of magma genesis. Journal of Volcanology and Geothermal Re- Classical Techniques. Geological Society, London, Special Publications, 168, search, 32, 1–12. 207–219. Balaguru, A. & Nichols, G. 2004. Tertiary stratigraphy and basin evolution, Hutchison, C.S. 1996. The ‘Rajang Accretionary Prism’ and ‘Lupar Line’ southern Sabah (Malaysian Borneo). Journal of Asian Earth Sciences, 23, problem of Borneo. In: Hall, R. & Blundell, D. J. (eds) Tectonic 537–554. Evolution of SE Asia. Geological Society, London, Special Publications, 106, Balaguru, A., Nichols, G.J. & Hall, R. 2003. Tertiary stratigraphy and basin 247–261. evolution of southern Sabah: implications for tectono-stratigraphic evolution Hutchison, C.S. 2005. Geology of North-West Borneo. Elsevier, Amsterdam. of Sabah, Malaysia. Geological Society of Malaysia Bulletin, 47, 27–49. Hutchison, C.S., Bergman, S.C., Swauger, D.A. & Graves, J.E. 2000. A Barbarin, B. 1999. A review of the relationships between granitoid types, their Miocene collisional belt in north Borneo: uplift mechanism and isostatic origins and their geodynamic environments. Lithos, 46, 605–626. adjustment quantified by thermochronology. Journal of the Geological Bateman, P.C., Clark, L.D., Huber, N.K., Moore, J.G. & Rinehart, C.D. 1963. Society, London, 157, 783–793. The Sierra Nevada Batholith; a synthesis of recent work across the Central Imai, A. & Ozawa, K. 1991. Tectonic implications of the hydrated garnet Part. US Geological Survey, Professional Papers, 414-D. peridotites near Mt Kinabalu, Sabah, East Malaysia. Journal of Southeast Bellon, H. & Rangin, C. 1991. Geochemistry and isotopic dating of the Cenozoic Asian Earth Sciences, 6, 431–446. volcanic arc sequences around the Celebes and Sulu seas. In: Silver, E.A., Ingram, G.M., Chisholm, T.J., Grant, C.J., Hedlund, C.A., Stuart-Smith, P. Rangin, C., von Breymann, M., et al. (eds) Proceedings of the Ocean & Teasdale, J. 2004. Deepwater North West Borneo: Hydrocarbon Drilling Program, Scientific Results, 124. Ocean Drilling Program, College accumulation in an active fold and thrust belt. Marine and Petroleum Station, TX, 321–338. Geology, 21, 879–887. Bjorlykke, K. 1993. Fluid-flow in sedimentary basins. Sedimentary Geology, 86, Jacobson, G. 1970. Gunung Kinabalu area, Sabah, East Malaysia. Malaysia 137–158. Geological Survey, Report, 8. Sarawak Government Press, Kuching. Black, L.P., Kamo, S.L., Allen, C.M., Aleinikoff, J.N., Davis, D.W., Korsch, Jiang, Y.-H., Jiang, S.-Y., Zhao, K.-D. & Ling, H.-F. 2006. Petrogenesis of R.J. & Foudoulis, C. 2003. TEMORA 1: a new zircon standard for Late Jurassic Qianlishan granites and mafic dykes, Southeast China: Phanerozoic U–Pb geochronology. Chemical Geology, 200, 155–170. implications for a back-arc extension setting. Geological Magazine, 143, Chiang, K.K. 2002. Geochemistry of the Cenozoic igneous rocks of Borneo and 457–474. tectonic implications. PhD thesis, University of London. Kasama, T., Akimoto, H., Hada, S. & Jacobson, G. 1970. Geology of the Mt. Claoue´-Long, J.C., Compston, W., Roberts, J. & Fanning, C.M. 1995. Two Kinabalu area, Sabah, Malaysia. Journal of Geosciences, Osaka City Carboniferous ages: a comparison of SHRIMP zircon dating with conven- University, 13, 113–148. tional zircon ages and 40Ar/39Ar analysis. In: Berggren, W.A., Kent, D.V., Kirk, H.J.C. 1968. The igneous rocks of Sarawak and Sabah. Malaysia Geological Aubry, M.-P. & Hardenbol, J. (eds) Geochronology, Time Scales and Survey, Bulletin, 5. Sarawak Government Press, Kuching. Global Stratigraphic Correlation. SEPM (Society for Sedimentary Geology) Koopmans, B.N. 1967. Deformation of the metamorphic rocks and the Chert Special Publications, 54,3–21. Spilite Formation in the southern part of the Darvel Bay area, Sabah. Clemens, J.D. & Mawer, C.K. 1992. Granitic magma transport by fracture Malaysia Geological Survey, Borneo Region, Bulletin, 8, 14–24. propagation. Tectonophysics, 204, 339–360. Le Maitre, R.W. 1989. A Classification of Igneous Rocks and Glossary of Terms. Coleman, D.S., Gray, W. & Glazner, A.F. 2004. Rethinking the emplacement Blackwell Scientific, Oxford. and evolution of zoned plutons: Geochronologic evidence for incremental Leong, K.M. 1974. The geology and mineral resources of the Upper Segama assembly of the Tuolumne Intrusive Suite, California. Geology, 32, 433–436. Valley and Darvel area, Sabah, Malaysia. Malaysia Geological Society, Collenette, P. 1958. The geology and mineral resources of the Jesselton– Borneo Region, Memoir, 4 (revised). Sarawak Government Press, Kuching. Kinabalu area, North Borneo. Malaysia Geological Survey, Borneo Region, Li, Z.X., Li, X.H., Wang, J., et al. 2001. South China in Rodinia—An update. Memoir, 6. Sarawak Government Press, Kuching. Gondwana Research, 4, 685–686. Collenette, P. 1965. The geology and mineral resources of the Pensiangan and Ludwig, K.L. 2001a. Using Isoplot/EX, v2.49, A geochronological toolkit for Upper area, Sabah. Malaysia Geological Survey, Memoir, 12. Microsoft Excel. Berkeley Geochronological Center Special Publication, 1a. Sarawak Government Press, Kuching. Ludwig, K.R. 2001b. Squid v1.02, A user manual. Berkeley Geochronological Connaughton, R.M. 1996. Descent into Chaos: The Doomed Expedition to Low’s Center Special Publication, 2. Gully. Brassey’s, London. Maniar, P.D. & Piccoli, P.M. 1989. Tectonic discrimination of granitoids. Corry, C.E. 1988. Laccoliths: Mechanisms of Emplacement and Growth. Geologi- Geological Society of America Bulletin, 101, 635–643. cal Society of America, Special Papers, 220. Matzel, J.E.P., Bowring, S.A. & Miller, R.B. 2006. Time scales of pluton Dhonau, T.J. & Hutchison, C.S. 1966. The Darvel Bay area, East Sabah, construction at differing crustal levels: Examples from the Mount Stuart and Malaysia. Malaysia Geological Survey Borneo Region, Annual Report for Tenpeak intrusions, North Cascades, Washington. Geological Society of 1965, 141–160. America Bulletin, 118, 1412–1430. Encarnacion, J. & Mukasa, S.B. 1997. Age and geochemistry of an ‘anorogenic’ McCaffrey, K.J.W. & Petford, N. 1997. Are granitic intrusions scale invariant? crustal melt and implications for T-type granite petrogenesis. Lithos, 42, Journal of the Geological Society, London, 154,1–4. 1–13. Michel, J., Baumgartner, L., Putlitz, B., Schaltegger, U. & Ovtcharova, Evans, D.J., Rowley, W.J., Chadwick, R.A. & Millward, D. 1993. Seismic M. 2008. Incremental growth of the Patagonian Torres del Paine laccolith reflections from within the Lake District Batholith, Cumbria, Northern over 90 k.y. Geology, 36, 459–462. England. Journal of the Geological Society, London, 150, 1043–1046. Miller, R.B. & Paterson, S.R. 1999. In defence of magmatic diapirs. Journal of Franke, D., Barckhausen, U., Heyde, I., Tingay, M. & Ramli, N. 2008. Structural Geology, 21, 1161–1173. Seismic images of a collision zone offshore NW Sabah/Borneo. Marine and Milsom, J. & Holt, R. 2001. Discussion of a Miocene collisional belt in north Petroleum Geology, 25, 606–624. Borneo: uplift mechanism and isostatic adjustment quantified by thermo- Frost, B.R., Barnes, C.G., Collins, W.J., Arculus, R.J., Ellis, D.J. & Frost, chronology. Journal of the Geological Society, London, 158, 396–400. C.D. 2001. A geochemical classification for granitic rocks. Journal of Morley, C.K. & Back, S. 2008. Estimating hinterland exhumation from late Petrology, 42, 2033–2048. orogenic basin volume, NW Borneo. Journal of the Geological Society, Glazner, A.F., Bartley, J.M., Coleman, D.S., Gray, W. & Taylor, R.Z. 2004. London, 165, 353–366. Are plutons assembled over millions of years by amalgamation from small Mrazec, M.L. 1915. Les plis diapirs et le diapirisme en ge´ne´ral. Comptes Rendus magma chambers? GSA Today, 14, 4–11. des Se´ances, Institut Ge´ologique de Roumanie, 4, 226–270. Hall, R. 2002. Cenozoic geological and plate tectonic evolution of SE Asia and Muir, R.J., Ireland, T.R., Weaver, S.D. & Bradshaw, J.D. 1994. Ion microp- the SW Pacific: computer-based reconstructions, model and animations. robe U–Pb zircon geochronology of granitic magmatism in the Western Journal of Asian Earth Sciences, 20, 353–431. Province of the South Island, New Zealand. Chemical Geology, 113,171– Hall, R. & Wilson, M.E.J. 2000. Neogene sutures in eastern Indonesia. Journal of 189. Asian Earth Sciences, 18, 781–808. Newton-Smith, J. 1967. Bidu-Bidu Hills area, Sabah, East Malaysia. Geological Hall, R., van Hattum, M.W.A. & Spakman, W. 2008. Impact of India–Asia Survey of Malaysia, Borneo Region, Bulletin, 4. collision on SE Asia: The record in Borneo. Tectonophysics, 451, 366–389. Nicolesco, C.P. 1929. Anticlinaux diapirs se´dimentaires, volcaniques et plutoni- Hazebroek, H.P. & Tan, D.N.K. 1993. Tertiary tectonic evolution of the NW ques. Bulletin de la Socie´te´ Ge´ologique de France, 29, 21–24. Sabah Continental Margin. Bulletin of the Geological Society of Malaysia, Nockolds, S.R. 1954. Average chemical compositions of some igneous rocks. 33, 195–210. Geological Society of America Bulletin, 65, 1107–1032. Higgins, M.D. 1999. Origin of megacrysts in granitoids by textural coarsening: a Omang, S.A.K. & Barber, A.J. 1996. Origin and tectonic significance of the crystal size distribution (CSD) study of microcline in the Cathedral Peak metamorphic rocks associated with the Darvel Bay Ophiolite, Sabah, Granodiorite, Sierra Nevada, California. In: Castro, A., Fernandez, C. & Malaysia. In: Hall, R. & Blundell, D. J. (eds) Tectonic Evolution of SE Vigneresse, J.L. (eds) Understanding Granites: Integrating New and Asia. Geological Society, London, Special Publications, 106, 263–279. 60 M. COTTAM ET AL.

Pearce, J.A., Harris, N.B.W. & Tindle, A.G. 1984. Trace element discrimination Tera, F. & Wasserburg, J. 1972. U–Th–Pb Systematics in three Apollo-14 diagrams for the tectonic interpretation of granitic rocks. Journal of basalts and problem of initial Pb in lunar rocks. Earth and Planetary Science Petrology, 25, 956–983. Letters, 14, 281–304. Petford, N. 1996. Dykes or diapirs? Transactions of the Royal Society of Thomson, K. & Schofield, N. 2008. Lithological and structural controls on the Edinburgh: Earth Sciences, 87, 105–114. emplacement and morphology of sills in sedimentary basins. In: Thomson, Petford, N. & Clemens, J.D. 2000. Granites are not diapiric! Geology Today, 16, K. & Petford, N. (eds) Structure and Emplacement of High-Level Magmatic 180–184. Systems. Geological Society, London, Special Publications, 302, 31–44. Petford, N., Cruden, A.R., McCaffrey, K.J.W. & Vigneresse, J.L. 2000. Tongkul, F. & Chang, F.K. 2003. Structural geology of the Neogene Granite magma formation, transport and emplacement in the Earth’s crust. Maliau Basin, Sabah. Geological Society of Malaysia Bulletin, 47, Nature, 408, 669–673. 51–61. Rangin, C., Jolivet, L. & Pubellier, M. 1990. A simple model for the tectonic Twist, D. & Harmer, R.E.J. 1987. Geochemistry of contrasting siliceous magmatic evolution of southeast Asia and Indonesia region for the past 43 m.y. Bulletin suites in the Bushveld Complex—genetic aspects and implications for de la Socie´te´ Ge´ologique de France, 8, 889–905. tectonic discrimination diagrams. Journal of Volcanology and Geothermal Reinhard, M. & Wenk, E. 1951. Geology of the colony of North Borneo. Research, 32, 83–98. Geological Survey Department of the British Territories in Borneo (Bulletin), 1. Vernon, R.H. 1986. K-feldspar megacrysts in granites—phenocrysts, not porphyr- Roberts, M.P. & Clemens, J.D. 1993. Origin of high-potassium, calc-alkaline, I- oblasts. Earth-Science Reviews, 23, 1–63. type granitoids. Geology, 21, 825–828. Vigneresse, J.L. 1990. Use and misuse of geophysical data to determine the shape Sandal, S.T. 1996. The Geology and Hydrocarbon Resources of Negara Brunei at depth of granitic intrusions. Geological Journal, 25, 249–260. Darussalam. Syabas, Bandar Seri Begawan. Vigneresse, J.L., Tikoff, B. & Ameglio, L. 1999. Modification of the regional Sewell, R.J., Chan, L.S., Fletcher, C.J.N., Brewer, T.S. & Zhu, J.C. 2000. stress field by magma intrusion and formation of tabular granitic plutons. Isotope zonation in basement crustal blocks of southeastern China: Evidence Tectonophysics, 302, 203–224. for multiple terrane amalgamation. Episodes, 23, 257–261. Vogt, E.T. & Flower, M.F.J. 1989. Genesis of the Kinabalu (Sabah) granitoid at a Simon, J.I., Renne, P.R. & Mundil, R. 2008. Implications of pre-eruptive subduction–collision junction. Contributions to Mineralogy and Petrology, magmatic histories of zircons for U–Pb geochronology of silicic extrusions. 103, 493–509. Earth and Planetary Science Letters, 266, 182–194. Williams, I.S. 1998. U–Th–Pb geochronology by ion microprobe. In: McKibben, Sperber, C.M. 2009. The thermotectonic development of Mount Kinabalu, Sabah, M.A., Shanks, W.C., III & Ridley, W.I. (eds) Applications of Microanaly- Malaysia: Constraints from low-temperature thermochronology. PhD thesis, tical Techniques to Understanding Mineralizing Processes. Reviews in University of London. Economic Geology, 7, 1–35. Swauger, D.A., Hutchison, C.S., Bergman, S.C. & Graves, J.E. 2000. Age and Zhou, X., Sun, T., Shen, W., Shu, L. & Niu, Y. 2006. Petrogenesis of Mesozoic emplacement of the Mount Kinabalu pluton. Geological Society of Malaysia granitoids and volcanic rocks in South China: A response to tectonic Bulletin, 44, 159–163. evolution. Episodes, 29, 26–33.

Received 23 February 2009; revised typescript accepted 27 August 2009. Scientific editing by Martin Whitehouse.