Journal of the Geological Society, London, Vol. 147, 1990, pp. 603-606, 3 figs 2 tables. Printed in Northern Ireland SHORT PAPER evidence of Precambrian inheritance (Parrish & Tirrull989). Earlier pre-collision granites within theKarakoram Age of crystallization and cooling of the batholith include the Muztagh Tower unit (Fig. 1) composed K2 gneiss in the Baltoro Karakoram of biotite and hornblende-rich foliated granodiorites, which gave three K-Ar hornblende ages spanning 82-75 f 3 Ma M.P. SEARLE', R. R. PARRISH', (Searle et al. 1989), and the Hushegneiss, SE of the Baltoro R.TIRRUL** & D.C. REX3 area, which has a U-Pb zircon age of 145 f 5 Ma and two 'Department of Earth Sciences, Oxford University, 40Ar-39Arages of 203 f 0.6 Ma and 204 f 1.4 Ma (Searle et Parks Road, Oxford OX1 3PR al. 1989). Further west,hornblende-bearing granodiorites 'Geological Survey of Canada, 601 Booth Street, from the Hunza plutonic unit gave a U-Pb age of 95 f 4 Ottawa, Canada KlA OE8 (LeFort et al. 1983) and similar granites at the Darkot Pass Department of Earth Sciences, Leeds University, gave a Rb-Sr isochron age of 111 f 6 (Debon et al. 1987). Leeds, LS2 9JT These pre-collision granites of the Karakoram batholith all have calc-alkaline geochemical affinities and have been interpretedas Andean-type granitesalong thesouthern continental margin of the Asian plate, related to the Themountains of K2 (8611 m)and Broad Peak (8047111) in the northward subduction of Tethyan oceanic crust (LeFort et Baltoro (northernPakistan) are composedof Karakoram al. 1983; Debon et al. 1987; Rex et al. 1988). plagioclase-hornblendeand biotite-hornblende-K-feldspar ortho- gneisses and Fphiholite-facies paragneisses,intruded by garnet- K2 gneiss. The K2 gneiss in northernmost Pakistan is not biotite-muscovite-tourmaline leucograniticveins. A U-Pb zircon part of the Karakoram batholith sensu stricto but outcrops age of 115-1u) Ma was obtained on an orthogneiss from the south along a wide belt NE of the batholith, separated bylow- face of K2. 40Ar-J9Aranalysis on hornblende yields a plateau age of grademetasediments. These metasedimentswere termed 90.6 f 1.8 Ma, consistent with a mid-Cretaceous phase of magma- the Doksamsequence and the Biange gneiss by Desio & tism, concomitant with early subduction-related components of the Zanettin (1970), and are metamorphosedequivalents of Karakoram batholith (Muztagh Tower unit, Hunza plutonic unit). the Carboniferous black shales and Permo-Triassic carbo- We interpret the K2 gneiss as representing a culmination of mid- nates exposed along the Gasherbrum Range. crustal along a discontinuousbut wide zone north of the rocks The gneiss includes both orthogneisses and para- Karakoram batholith. K2 gneisses which appear to be in fault-bounded relationship Recent mapping around the upper Baltoro, Godwin-Austin with the adjacentsediments. The paragneisses include and Savoia glaciers in the central Karakoram has identified clinopyroxene-hornblende bearing psammites and garnet- very large-scale culminations of mid-crustal gneisses (K2 diopside bearingmarbles in the amphibolite facies, and gneiss) surrounded by upper crustalsedimentary rocks of were subdivided by Desio & Zanettin (1970) who des- theGasherbrum sedimentaryseries (Fig. 1). The regional cribed them as the Falchan gneiss. We include these rocks geology of the Baltoro glacier area has been described by with the orthogneisses as the K2 gneiss because theyap- Desio & Zanettin (1970) andmore recentlySearle et al. pear to be interbanded on the west flank of Broad Peak (1986, 1989), and Rex et al. (1988). This paper describes the and the SW spur of K2. The orthogneisses include K2 gneiss in thenorthern Karakoram terrane and we plagioclase-hornblende gneiss and biotite-hornb1ende-K- interpret its crystallization and cooling history from a study feldspar gneiss, which are stongly foliated and intruded by of regional structural constraints, U-Pb and 40Ar-39Ar,K-Ar late leucogranitic dykes containing garnet, muscovite, geochronology. biotite, tourmaline, plagioclase, K-feldspar and quartz. K- The northern Karakoram terrane lies north of the main Ar ages from mica separatesfrom the dykesrange from Karakoram axial batholith which stretches at least 500 km 70-58Ma (Searle et al. 1989). The leucogranitedykes from the Hindu Kush mountains across northern Pakistan post-date an early deformation event in the K2 gneiss but into the Siachen glacier areanorth of Ladakh.The are clearly offset by later shear zones. Gasherbrum sedimentaryseries includes Carboniferous Southeast of K2-Broad Peak, intrusions of quartz black shales overlain by thick Permain-Jurassic/Lower diorite porphyry have been mapped on the south flank of Cretaceous carbonates with minor sandstones, conglomer- BroadPeak andaround Gasherbrum IV and 111, which ates and shales. Ash-flow tuffs have also been recognized have an identical geochemical signature to the K2 ortho- associated with the Lower Cretaceous Savoia Fm. gneiss and we presume that the two are related. limestones along theeast bank of the Godwin-Austin U-Pb geochronology. A 4 kg sample of K2 orthogneiss glacier. Along the northern margin of the Baltoro plutonic from the SSE ridge of K2 was collected approximately unit near Concordia (Fig. l), the Carboniferous black shales 200 m NE of K2 base camp on the upper Godwin-Austin have been thermally metamorphosed to andalusite-bearing glacier. The sample is a strongly foliatedhornblende- hornfels along a narrow contact metamorphic aureole (Mitre biotite granodioriticorthogneiss with zircon, allanite, and sequence). apatiteas the main accessory minerals. U-Pbdating of In the Baltoro glacier region the Karakoram batholith is zircon was carried out using the methods outlined by Par- dominantlycomposed of theBaltoro plutonicunit with rish et al. (1987): data are given in Table 1 and plotted on compositions ranging from monzogranite to garnet two-mica a concordia diagram (Fig. 2). Some of the zircon fractions leucogranite, associated with numerous pegmatite-aplite were abraded (Krogh, 1982) to removeany potential sec- dykes. AU-Pb zircon age of 21 f 0.5 Madates the ondary Pb loss, although this failed to improve the con- crystallization age of this unit,and the zircons show cordance of the data. Five analyses of high quality zircon plotin a small * Deceased. region bounded between 117 and 123 Ma on the concordia 603 Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/147/4/603/4890401/gsjgs.147.4.0603.pdf by guest on 26 September 2021 604 SEARLE M. P. ET AL. SavolaFm Lamprophyre dykes L Khalkhal Fm. m Tulls. Porphyrys Quartz diorites m Aghll Fm IGasherbrum m sedimentary Doksam sequence serles UrdokFm I*'\'21 Muztagh Tower unlt Shaksgam Fm Baltoro Fm JMU/ Masherbrumcomplex Fig. 1. Geological map of the Baltoro galcier area includingK2, Broad Peak and the Gasherbrum Range. Geochronological data are taken from Parrish & Tirrul (1989) Searle et al. (1989) and the present work. diagram, but do not define either a line or a quasi-linear faceted, and typical of igneous zircons, Although one array. Their '"'Pb/'"Pb ages rangefrom 124 to180Ma. could argue for extreme Pb loss and a much older age, this Thistype of zircon data is not amenableto rigorouses- is not consistent with the rock's history, which is not charac- timation of age uncertainty in itsage interpretation.The terized by excessively high temperature metamorphism. scatter is related to small amounts of older zircon in- heritance in most or all analyses, with the age of inherited 40Ar-39Argeochronology. Full experimentaldetails of the zircons being variable andindeterminate. Nevertheless, irradiationand argon analytical methods used at Leeds the visible evidence of cores in some grains and the rela- University are found in Parsons et al. (1988). Hornblendes tively tight cluster of dataare mostconsistent with this from samples of K2 orthogneiss and Muztagh Tower gneiss being an igneous rock 115-120 Ma old.The zircon mor- have been analysed and results are shown in Fig. 3. phologies areeuhedral tosubhedral, elongate, simply The low temperature steps of the Ar-Ar analysis of the Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/147/4/603/4890401/gsjgs.147.4.0603.pdf by guest on 26 September 2021 SHORT PAPER Table 1. U-Pb analytical data for zircon, K2 gneiss U Pb t 2o"Pbl msPb t mPh/206pb Sample Size pm* wt. (mg) (ppm) (wm) 2MPb5 (%) mPh/z38U v mPb/z35UmPb/mPb n age, (Ma)** 1 +149n 1 15910.3926 27.79 10980 5.0 0.01838 (0.09) 0.1234(0.11)0.04869(0.04) 133.1f2.0 2 +149a 0.1759 1678 29.47 11490 4.1 0.018@3(0.11) 0.1251 (0.13) 0.04856 (0.04) 126.8f2.0 3 +149a 0.1635 1227 21.78 7253 6.20.01848 (0.07) 0.1236(0.10) 0.04851 (0.05) 124.2 k 2.4 4 -149+105n 0.171 1506 26.07 10930 4.2 0.01839 (0.07) 0.1241(0.10) 0.04893 (0.04) 144.2 k 2.0 5 -149+ lO5a 0.2543 1405 24.82 12010 5.0 0.01859 (0.06) 0.1274(0.09) 0.04968 (0.04) 180.0 f 1.8 *a, abraded; n, needles; analysed zircons were generally clear and had rare faint evidence of cores. t Weighing error = 0.002 mg. t Radiogenic Pb. 0 Measured ratio, corrected for spike and fractionation; fractionation was 0.12 f 0.036%/amu. 11 Total common Pb in analysis corrected for fractionation andspike. Corrected for blankPb and U, common Pb(Stacey-Kramers 116Ma Pb: M6Pb/Z04Pb= 18.53 f 0.54;m7pb/~b= 15.62 f 0.8); errors in parentheses are 1 standard error of the mean in percent. Pb blanks were about 20 pg. '*Corrected for blank and common Pb, errors are 2 standard errors of the mean. steps there is evidence thatthe rock was subjected to a ,lg.