Correct Nomenclature for the Angadimogar Pluton, Kerala, Southwestern India

Correct nomenclature for the Angadimogar pluton, Kerala, southwestern India

HMRajesh Graduate School of Life and Environmental Sciences, University of Tsukuba, Ibaraki 305-8572, Japan. e-mail: [email protected]

The proper usage of modal composition and geochemical classiﬁcation of granitoids is discussed for assigning a proper nomenclature for the Angadimogar pluton, Kerala, southwestern India. This discussion is mainly aimed at addressing questions concerning the nomenclature of Angadimogar pluton (syenite vs. granite). Modal composition and whole-rock XRD data clearly show that the pluton exposed near Angadimogar is a quartz-syenite and its geochemistry is typical of a ferroan, metaluminous, alkali (A-type) granitoid.

1. Introduction suggest that these bodies are granites based on their normative compositions. They apparently Classification of granitoid rocks involving various question the nomenclature of Angadimogar plu- criteria has long been a subject of frequent debate ton (syenite vs. granite) used by Santosh and Nair and voluminous literature. It can be a confusing (1986). Rajesh (1999) carried out a detailed field, exercise involving the minerals present and their petrographic, geochemical, and isotopic character- relative proportion in the rock (depending largely ization of the Angadimogar pluton. The present on the chemical composition of the magma), the paper is based on this work and intends to demon- texture of the rock (depending largely on the cool- strate that the pluton exposed near Angadimogar ing history of the magma), the colour of the rock is a quartz-syenite. (depending largely on the minerals present and on their grain size), and the chemical composition of the rock. Granitoid classification schemes 2. Study area and field relations have evolved from being genetic (e.g., Chappell and White 1974) and/or tectonic (e.g., Pearce et al The Neoproterozoic granulite-facies terrain of 1984) in nature to non-genetic and non-tectonic southwestern India is intruded by a suite of high-K in nature (e.g., Frost et al 2001). Controversies calc-alkalic plutons (metaluminous to slightly per- surrounding the classification of granitoid rocks aluminous biotite granites and hornblende-biotite are usually due to the usage of different schemes granites) with subsequent widespread alkaline of classification by different workers. Such contro- (A-type) plutons (including quartz syenites) and versies rarely arise from the quality of geochemi- minor garnet-bearing leucogranites (Rajesh 1999, cal data used. This study however, illustrates an 2000, 2003, 2004). These granitoids, emplaced example of misclassification of a granitoid rock between ∼ 640 and 500 Ma (figure 1a), were due to the incorrect application of poor-quality generated in environments of high heat flow geochemical data. and volatile activity, correlative with an exten- Anil Kumar et al (2005) reported geochem- sional tectonic regime (Rajesh 1999, 2003, 2004). ical data from two plutons (Angadimogar and The pluton exposed near Angadimogar (fig- Kumbdaje) from Kerala, southwestern India, and ure 1a), northern Kerala, belongs to this suite,

Keywords. Nomenclature; modal composition; whole-rock XRD; geochemistry; quartz-syenite; A-type granitoid.

J. Earth Syst. Sci. 115, No. 2, April 2006, pp. 239–248 © Printed in India. 239 240 HMRajesh

Figure 1. (a) Generalized geologic map of Kerala, southwestern India, showing the approximate location of granitoids and anorthositic gabbro mentioned in the text. Available ages of granitoids are also given. Inset is a shaded relief image of southern India showing the states of Kerala and Tamil Nadu. (b) Photograph illustrating the pink syenite exposed near Angadimogar, northern Kerala. (c) Modal composition (this study) and normative compositions (reproduced from Anil Kumar et al 2005) of the Angadimogar samples in a QAP plot. The three components, Q (quartz), A (alkali (Na–K) feldspar), and P (plagioclase), were recalculated from the modal amounts to sum to 100 per cent (as recommended by IUGS). The plagioclase ratio (100 × P/(A + P )) is also useful to name the rock, as the non-horizontal divisions in the QAP diagram are lines of constant plagioclase ratio. and is a quartz-syenite composed dominantly minor quartz, to a medium- to coarse-grained of medium- to coarse-grained K-feldspar, with rock rich in pink K-feldspar, with fine-grained minor amounts of quartz, plagioclase and uni- hornblende, to a medium-grained rock rich in formly distributed ferromagnesian minerals (fig- grey feldspar, with finer grained hornblende. The ure 1b). Greenish amphibole constitutes the syenite is locally cut by thin (< 2 cm) amphibole- principal mafic constituent with minor biotite. bearing veinlets. Pegmatites and aplites cut across This is in contrast to other syenites in south- the syenite exposures and attest to the presence of western India, where pyroxene constitutes the late-stage fluids. main mafic mineral. The syenitic body shows The syenite shows an intrusive relationship with variations from a coarse-grained rock rich in the surrounding light coloured, medium- to coarse- pink feldspar, with large hornblende grains and grained (sometimes migmatitic) hornblende-biotite Correct nomenclature for Angadimogar pluton 241 gneiss/biotite-hornblende gneiss, often carrying Rajesh 2005), with a dominant peak for feldspar boudins and bands (concordant as well as dis- (figure 2). The syenite shows a general allotrio- cordant) of amphibolites. These ortho-gneisses morphic texture and consists predominantly of range in composition from alkali-feldspar-granite perthitic orthoclase (68–75%), normally zoned pla- with less than 6% modal biotite to tonalite with gioclase (6–9%), and a variety of mafic minerals: up to 10% biotite. Hornblende-rich granitoid is amphibole (potassian-ferro edenitic hornblende to much more common than biotite-rich granitoid. ferro edenite to ferro actinolite (in the nomencla- Dark greenish-grey coloured, medium grained, ture of Giret et al 1980)), clinopyroxene (diop- orthopyroxene-bearing quartzo-feldspathic charno- side (in the nomenclature of Morimoto 1988)), ckite is exposed in the northwestern and southeast- and biotite (annite) (Rajesh 1999). In some sam- ern parts of the study area. This rock also contains ples acmite was found. Quartz occurs as intersti- hornblende and biotite, with occasional garnet. tial grains. Magnetite is the main opaque phase and is generally intergrown with ilmenite. Other accessory phases include zircon, allanite, apatite, 3. Petrography titanite, calcite, and epidote. For plutonic rocks with less than 90% (by volume) mafic (dark) minerals, the rock name is determined 4. Geochemistry from the relative proportions of quartz, alkali feldspars, plagioclase feldspars, and felspathoids Santosh and Nair (1986), Rajesh (1999) and (nepheline, leucite, etc.). In any classification Anil Kumar et al (2005) presented geochemical scheme, boundaries between classes are set arbi- data on the Angadimogar pluton. Representative trarily; however, if the boundaries can be placed major element data from these studies are given in close to natural divisions or gaps between classes, table 1. The syenite has high K2O (up to 6 wt%). they will seem less random and subjective, and In terms of aluminium saturation index (ASI; such standards can facilitate universal understand- Shand 1943) (which is better than A/CNK (molar ing. The most commonly used scheme was devised Al2O3/(CaO + Na2O+K2O)) in defining peralka- by the International Union of Geological Sciences line/metaluminous/peraluminous compositions, as (IUGS) and is based on Streckeisen (1976) (fig- it takes into account the presence of apatite, and ure 1c). IUGS recommends that if the mineral is expressed as the molecular ratio of Al, Ca, P, Na mode cannot be determined (or difficult to deter- and K), Angadimogar syenite samples are clearly mine) as is often the case for volcanic rocks, then a metaluminous (ASI < 1 and molecular Na + K < chemical classification (e.g., total alkali versus sil- molecular Al) similar to quartz-syenites worldwide ica diagram of Le Bas et al (1986)) is better. More and different from alkali-feldspar-syenites which discussions on this topic can be found in Le Bas are mostly peralkaline (e.g., Eby 1990; Nédélec et al and Streckeisen (1991) and Le Maitre et al (2002). 1995; Rajesh 2003). In such rocks (where ASI < 1 Modal analyses (obtained by point count- and molecular Na + K < molecular Al) there is ing) of representative samples from the Angadi- likely to be excess Ca after aluminium has been mogar pluton plotted on a Streckeisen (1976) accommodated in the feldspars. As a result, meta- QAP plot indicates that they range in composi- luminous rocks contain calcic phases such as horn- tion from quartz-syenite to quartz-alkali-feldspar- blende and augite but lack either muscovite or syenite (figure 1c). Whole rock XRD analyses were sodic ferromagnesian phases. This clearly accounts carried out as part of this study on powdered for the mineralogy of the Angadimogar syenite. rock samples from six representative plutons (three On the other hand, if ASI > 1, then the granites, one anorthositic gabbro, and two syen- rock is corundum-normative and peraluminous. ites, including the Angadimogar pluton) across This means that the rock has more Al than southwestern India, using a Philips PW3020 model can be accommodated in feldspars and that it X-ray diffractometer (housed at Fukuoka Univer- must have another aluminium phase present. For sity of Education, Japan), using Ni-filtered, CuKα weakly peraluminous rocks this phase may be radiation and a scanning speed of 1◦ min−1.All aluminous biotite, but for strongly peraluminous the representative whole-rock XRD plots of south- granites the phase can be muscovite, cordierite, western Indian granites (e.g., Ezhimala, Ambala- garnet or an Al2SiO5 polymorph. All the ‘gran- vayal, Pathanapuram granites (see figure 1a)) show ite’ samples given in Anil Kumar et al (2005) are clear peaks for the main granitic minerals quartz strongly peraluminous (ASI: 1.24–2.25 (Angadi- and feldspar (figure 2). On the other hand, Angadi- mogar); 1.78 (Kumbdaje); see table 1). But the mogar syenite whole-rock XRD pattern is simi- mineralogy of both Angadimogar and Kumbdaje lar to that of Puttetti syenite (Rajesh 2003) and plutons mentioned in Anil Kumar et al (2005) (qtz- Perinthatta anorthositic gabbro (see figure 1a; kfs-plg-bt-pyx-hbl-opaques) includes no Al2SiO5 242 HMRajesh

Figure 2. Representative whole-rock XRD plot of Angadimogar syenite compared with that of Ezhimala, Ambalavayal, and Pathanapuram granites, Puttetti syenite, and Perinthatta anorthositic gabbro, southwestern India. The granites show clear peaks for the main granitic minerals quartz and feldspar, while the syenites and anorthositic gabbro show a dominant peak for feldspar. Peaks of other identiﬁed minerals are not shown for clarity.

polymorph to characterize the strongly peralumi- and Hamner 1981; Nabelek et al 1992; Inger and nous nature. One reason for this can be the poor Harris 1993; Searle et al 1997), charnockite mas- quality of geochemical data used. sifs from southern India (Rajesh and Santosh 2004 Let me address the quality of geochemical data and references therein), hbl-bt gneiss/bt-hbl gneiss given in Anil Kumar et al (2005). At a glance from central and northern Kerala (Nambiar et al the first thing that strikes a granitoid geochemist, 1992; Rajesh 1999), sodic and potassic garnet- is the extremely low Na2O content (0.23–0.65) of biotite gneiss, garnet-biotite-sillimanite gneiss, and all the rock types from the study area (7 gran- cordierite gneiss from southern Kerala (Chacko ite samples from Angadimogar pluton, 1 aplite et al 1992), and mafic dykes from southern India sample, 3 gneiss samples and 1 charnockite sam- (Radhakrishna et al 1995) (figure 3). It should be ple from Angadimogar area, and 1 granite sam- noted that quartz syenites with hornblende as the ple from Kumbdaje pluton; see table 1) (figure 3). major mafic mineral (syenites from East Green- The Na2O content of these samples were com- land, Russia and Antarctica) have similar Na2O pared with Na2O contents of granites and syenites contents with the Angadimogar syenite samples, from southwestern India (Rajesh 1999, 2000, 2003, while quartz syenites with pyroxene as the major 2004), typical A-type granitoids (including quartz- mafic mineral (other syenites from southwestern syenites) given in Eby (1990), quartz-syenites from India and Madagascar) have lower Na2Ocontents. East Greenland, syenites (quartz-syenite to alkali- A common feature of all these quartz-syenite stud- feldspar-syenite) from Russia, quartz-syenites from ies is their modal per cent of quartz (average 10%; Antarctica, and quartz-syenites from Madagascar sometimes reaching up to ∼19%). This is compa- (see Rajesh 1999, for the references and com- rable to modal contents used in this study as well pilation), typical peraluminous granites (Strong as Santosh and Nair (1986). Correct nomenclature for Angadimogar pluton 243

Figure 3. Na2O vs. SiO2 plots illustrating the compositional variance of the data presented in Anil Kumar et al (2005) in comparison with representative data from a range of rock types from southern India and elsewhere. The dotted line in each ﬁgure represents the upper limit of the data given in Anil Kumar et al (2005). None of the rock types have the extremely lower Na2O content as the samples given in Anil Kumar et al (2005). 244 HMRajesh 5 O 2 will only increase the 5 O 1.00 2 Rajesh (1999) 0.86 0.92 0.90 0.90 9.86 9.72 10.65 11.02 0.090.68 0.13 0.47 0.08 0.35 0.09 0.36 1.56 1.60 1.16 1.20 0.410.97 0.53 0.99 0.20 0.97 0.29 0.99 5.850.42 7.01 3.89 0.30 0.14 5.24 0.09 6.874.55 6.05 5.27 6.39 5.42 5.97 6.25 0.19 0.28 0.17 0.19 0.424.04 0.38 5.610.18 0.16 3.22 0.13 0.16 3.14 0.06 0.08 0.85 3.16 3.86 3.79 1.72 1.17 0.42 0.89 99.03 100.67 99.07 101.40 27.2758.95 31.10 51.12 32.40 54.69 36.53 49.96 61.8918.34 62.27 18.23 64.24 17.79 65.33 18.53 Syenite Syenite Syenite Syenite R95091G R95091B R95091F R95091E . MgO) + (2005). Instead a column titled ‘Others’ was given; MALI – Modiﬁed Alkali Lime Index. et al O–CaO; Fe-number – total Fe/(total Fe 2 K + O 2 0.311.83 0.12 1.470.13 0.16 2.48 0.02 0.44 1.16 0.02 0.15 59.7418.86 62.88 18.86 63.46 17.84 65.85 16.31 or LOI data was given in Anil Kumar 5 O 2 Representative major element data of the Angadimogar syenite from Santosh and Nair (1986) and Rajesh (1999). The major element data and ∗ 3 3 5 2 O 6.73 6.35 6.43 6.12 2 O O 2 O 5.17 5.05 5.60 3.68 O 2 2 2 2 Table 1. data was used in the calculation of ASI for dataSample from No. Anil Kumar et al (2005), as it was not available. Inelements any case, addition AM-1/2(wt%) of P SiO AM-1/4 AM-1/7MnO AM-1/9 Total 0.07 0.04q 99.04 0.05di 99.24 0.05 *Others 99.72 – No P 4.76 99.45 1.25 0.59 1.73 2.01 10.37 4.33 normative compositions on rock types from the Angadimogar and Kumbdaje areas from Anil Kumar et al (2005) are reproduced for comparison. No P K P ap 0.31 0.05 0.05 0.35 il 0.60 0.23 0.31 0.84 TiO Fe Fe-number 0.94 0.83 0.92 0.84 Al FeOLOI 3.09 1.69 0.32 1.29 0.33 2.26 0.50 0.32 ASI 0.90 0.99 0.92 0.89 MgOCaO 0.32 2.47 0.64 1.79 0.32 1.57 0.64 2.47 Rock type Syenite Syenite Syenite Syenite C.I.P.W. norm Na or 30.95 30.17 33.35 21.94 Others MALI 9.43 9.61 10.46 7.33 Reference Santosh and Nair (1986) aban 47.45 6.06 54.32 8.13 54.84 3.30 52.24 6.22 valueofASI.MALI–Na Major C hy 3.10 2.01 Correct nomenclature for Angadimogar pluton 245 7.03 0.57 13.14 − (2005) et al . 0.23 0.12 0.69 0.071.10 0.38 2.30 0.05 0.80 0.25 0.05 1.30 0.27 0.20 0.40 0.90 0.10 1.30 0.70 0.97 2.10 0.00 1.20 1.50 1.30 0.50 69.1016.59 71.21 14.83 70.11 15.73 76.80 13.14 79.10 11.53 79.25 11.53 79.10 79.20 10.97 11.06 75.20 13.46 76.20 13.14 78.40 12.45 59.50 13.40 78.30 13.87 (Continued) ∗ 3 3 5 2 O 0.65 0.54 0.63 0.32 0.33 0.37 0.38 0.33 0.43 0.41 0.37 0.23 0.51 2 O O 2 O 4.61 5.73 3.30 5.78 2.35 4.10 2.12 4.75 3.32 3.68 4.38 1.94 2.43 O 2 2 2 2 P K Fe-numberC.I.P.W. norm 0.94 0.93 0.80 0.82 0.76 0.95 0.86 0.98 0.86 0.83 0.70 0.61 0.74 Al TotalASIMALI 98.21 1.92 3.85 99.83 1.24 3.58 99.12 98.82 1.76 1.55 99.51 1.64 5.41 97.30 1.74 0.73 97.95 97.20 1.94 3.98 99.47 2.07 1.19 99.93 1.79 4.80 99.92 2.25 2.82 99.59 1.80 2.64 99.74 1.60 3.41 0.70 1.78 Major elements (wt%) Table 1. Sample No.Rock typeReference A1 ‘Granite’ ‘Granite’SiO ‘Granite’ A2 ‘Granite’ Gneiss ‘Granite’FeO A3MnO GneissMgO ApliteCaO ‘Granite’Na A4 ‘Granite’ Gneiss 4.11LOI Charnockite A7 0.14Others ‘Granite’ 0.27 1.41 2.13 A8 0.12 0.16 2.69 4.29 A9 0.09q 1.10or 2.38 A10 0.58ab 0.01an 0.13C A11 Anil 2.76 0.69 Kumar 0.05di 0.86 0.57 A13 1.95 41.10 0.01 27.24 0.03 5.24 2.14 A17 0.49 38.04 0.03 6.90 33.91 0.35 0.79 8.06 0.02 1.31 4.71 43.98 A20 13.24 0.02 19.46 3.17 0.28 2.75 0.06 51.00 5.24 11.86 0.53 33.92 A21 0.93 2.82 60.60 6.73 0.07 13.90 2.62 3.31 0.56 56.70 0.96 1.45 24.46 0.01 5.20 2.62 9.66 0.41 63.72 12.78 1.34 7.91 4.90 58.80 0.12 3.14 2.48 28.35 5.02 54.73 9.20 5.51 19.46 3.14 6.34 1.27 0.05 55.68 0.44 2.62 0.28 5.71 21.68 2.37 55.14 5.20 26.13 3.66 4.69 21.06 7.44 11.12 3.66 7.18 5.81 59.22 3.14 6.62 14.45 4.39 29.53 2.10 11.59 4.19 6.12 hyilap 8.09 0.46 4.66 0.30 9.66 1.37 1.22 0.15 6.95 0.76 1.92 0.15 4.46 0.48 0.23 0.15 6.84 0.46 1.40 0.76 2.58 0.15 19.26 1.82 3.61 TiO Fe 246 HMRajesh

Figure 4. (a) Q’ vs. ANOR plot for the Angadimogar syenite samples (ﬁlled symbols; this study), and ‘granite’ samples from Anil Kumar et al (2005) (open symbols). Inset shows the outlier samples from Anil Kumar et al (2005). (b) Normative compositions of granite samples given in Anil Kumar et al (2005) compared with similar compositions of 7659 granitic rocks from around the world. Note that about 50% of the analyses fall within only about 5% of the diagram (analyses compiled and plotted courtesy of Roger W Le Maitre, Tasmania).

None of the rock types given in figure 3 have without an Al2SiO5 polymorph to account for the such extremely lower Na2O content as the samples peraluminous composition. given in Anil Kumar et al (2005). Even typical peraluminous granites do not have such low Na2O.In the absence of data on analytical accuracy and pre- 5. Normative compositions and the cision in the Anil Kumar et al (2005) paper, the nomenclature of Angadimogar pluton only way I can explain the extremely low and similar Na2O content of granite, gneiss, charnockite, Anil Kumar et al (2005) calculate normative com- and aplite samples analysed by them is poor data positions from their geochemical data and plot quality. This low Na2O combined with the high it in the QAP diagram of Streckeisen (1976) Al2O3 content translates to a corundum normative (see figure 1c). It is quite clear from figure 1(c) composition and hence a peraluminous rock, but that plotting normative compositions and modal Correct nomenclature for Angadimogar pluton 247 compositions of the same pluton in the QAP plot can lead to a significant difference in nomenclature. Streckeisen and Le Maitre (1979) put forward the Q’ – ANOR method, which is approximately equal to that of the modal QAP diagram of IUGS, for usage with normative compositions. In the Q’ – ANOR plot (figure 4a) samples from this study plot in the syenite field, while most of the ‘granite’ samples from Anil Kumar et al (2005) plot out of the fields covered by the diagram (see inset in figure 4a). This uncertainty of the normative compositions of the ‘granite’ samples is further visible in figure 4(b), where they are compared with normative compositions of 7659 granitic rocks from around the world (compiled and plotted by Le Maitre).

6. Proper classiﬁcation of the Angadimogar pluton

Granitoid classification is problematic as the same mineral assemblage (quartz-feldspar- ferromagnesian minerals) can be achieved by a number of processes. A major drawback of the IUGS classification scheme is that it ignores compositional variations apart from those that affect the feldspar abundances. Because of this com- plexity, petrologists have relied upon geochemical classifications to distinguish between various types of granitoids. This study utilizes the geochemi- ∗/ ∗ cal classification of Frost et al (2001) based on Figure 5. Fe (Fe +MgO) vs. SiO2 and Na2O+K2O–CaO vs. SiO2 plots for the Angadimogar syenite samples, and Fe-number, modified alkali lime index, and ASI. ‘granite’ samples from Anil Kumar et al (2005). Accordingly Angadimogar syenite is ferroan, alkalic, and metaluminous, typical of alkali (A-type) ∼ granitoids (figure 5). Data from Anil Kumar et al belongs to the 640–500 Ma granitoid magmatic activity along southwestern India (Rajesh 1999, (2005) subjected to similar geochemical discrim- 2000, 2003, 2004). It has all the geochemi- ination suggests that the Angadimogar pluton cal characteristics of alkali (A-type) granitoids. is magnesian to ferroan, calcic, and peralumi- Similar quartz-syenites have been reported from nous (Kumbdaje is magnesian, calcic, and per- different continental fragments (Arabian shield aluminous) (figure 5). Interestingly, Frost et al (Harris 1985), Madagascar (Nédélec et al 1995), (2001) points out that magnesian, calcic, and East Antarctica (Zhao et al 1995)), placed adjacent peraluminous compositions are shown by dior- to India in a supercontinent framework. ites, quartz diorites, and tonalites, while there is The absence of any mineral (muscovite, no known granitoid rock with ferroan, calcic, and cordierite, corundum, andalusite, tourmaline, peraluminous compositions. topaz, garnet, cordierite, sillimanite) characteristic of peraluminous rock does not correlate with the 7. Concluding remarks peraluminous compositions presented for samples from the Angadimogar pluton (as well as Kumb- Experimental results indicate that the partial melts daje pluton) by Anil Kumar et al (2005). The mafic generated deep in the crust are closer to syen- mineralogy (hornblende, clinopyroxene, biotite) of ite than granite in composition (e.g., Huang and Angadimogar syenite is clearly characteristic of Wyllie 1975). Low initial 87Sr/86Sr isotope ratio a metaluminous rock. Thus the syenite–granite (0.7032 ±0.0008; Santosh and Nair 1986; Rajesh nomenclature issue raised by Anil Kumar et al 1999) and estimated (using Al-in-Hbl geobarome- (2005) for the Angadimogar pluton (as well as try; Rajesh 1999) pressure conditions of ∼ 8kbar Kumbdaje pluton) is just an artifact of using ques- for the Angadimogar syenite translates to a deep tionable geochemical data, and the improper usage crustal depth. The Angadimogar quartz-syenite of normative compositions in the QAP diagram. 248 HMRajesh

Acknowledgements petrologic and geochemical constraints; Contrib. Mineral. Petrol. 110 173–191. Nambiar C G, Bhaskar Rao B, Parthasarathy R and Fed- This work is dedicated to the fond memory of kin V V 1992 Geochemistry and genesis of charnock- Sujith D Prasad, Geologist at Geoservices, who was ites and associated gneisses from northern Kerala, India; more than an inspiration to all the 1994-Nattakom- In: In high-grade metamorphics (ed.) A Barto (Athens: batch-mates. Sandeep Singh and Manoj Pandit Theophrastus Publishers) pp. 187–215. provided perceptive and constructive comments on Nédélec A, Stephens W E and Fallick A E 1995 The Pan- African stratoid granites of Madagascar: alkaline magma- the manuscript. The reviewers are also thanked tism in a post-collisional extensional setting; J. Petrol. for taking the labour of going through the article 36 1367–1391. of Anil Kumar et al (2005). Editorial comments Pearce J A, Harris N B W and Tuttle A C 1984 Trace ele- by Hetu Sheth were valuable. Naomi Tashiro and ment discrimination diagrams for the tectonic interpreta- Teiichi Ueno from the Fukuoka University of tion of igneous rocks; J. Petrol. 25 956–983. Radhakrishna T, Pearson D G and Mathai J 1995 Evolution Education, Japan, helped with whole-rock XRD of Archaean southern Indian lithospheric mantle: a geo- analysis. This is a contribution to the Gondwana chemical study of Proterozoic Agali-Coimbatore dykes; Institute for Geology and Environment (GIGE) Contrib. Mineral. Petrol. 121 351–363. and IGCP 453. Rajesh H M 1999 Characterization and origin of alkaline and calc-alkaline aluminous A-type granitoids from southwestern India: implications for Gondwanaland tectonics; Unpublished DSc thesis Osaka City University, Japan, References 317p. Rajesh H M 2000 Characterization and origin of a com- Anil Kumar A, Gopinathan V, Gopinath K S and Rao V P positionally zoned aluminous A-type granite from South 2005 A review on the nomenclature of Angadimogar and India; Geol. Mag. 137 291–318. Kumbdaje plutons, Kasargod district, Kerala; J. Geol. Rajesh H M 2003 Outcrop-scale silicate liquid immiscibility Soc. India 65 141–146. from an alkali syenite (A-type granitoid)-pyroxenite asso- Chacko T, Kumar G R R, Meen J K and Rogers J J W 1992 ciation near Puttetti, Trivandrum Block, South India; Geochemistry of high grade supracrustal rocks from the Contrib. Mineral. Petrol. 145 612–627. Kerala Khondalite Belt and adjacent massif charnockites; Rajesh H M 2004 The igneous charnockite – high-K alka- Precamb. Res. 55 469–489. licalcic I-type granite – incipient charnockite association Chappell B W and White A J R 1974 Two contrasting gran- in Trivandrum Block, southern India; Contrib. Mineral. ite types; Pacific Geol. 8 173–174. Petrol. 147 346–362. Eby G N 1990 The A-type granitoids: a review of their Rajesh H M 2005 Progressive or continual exsolu- occurrence and chemical characteristics and speculations tion in pyroxenes: an inidcator of polybaric igneous on their petrogenesis; Lithos 26 115–134. crystallization for the Perinthatta anorthositic gabbro, Frost B R, Barnes C G, Collins W J, Arculus R J, Ellis D J northern Kerala, southwestern India; J. Asian Earth Sci. and Frost C D 2001 A geochemical classification for (doi:10.1016/j.jseaes.2004.11.004). granitic rocks; J. Petrol. 42 2033–2048. Rajesh H M and Santosh M 2004 Charnockitic mag- Giret A, Bonin B and Leger J M 1980 Amphibole composi- matism in southern India; J. Earth Sys. Sci. 113 tional trends in over saturated and under saturated alka- 565–585. line plutonic ring-complexes;Can. Mineral. 18 481–495. Santosh M and Nair N G K 1986 Petrogenesis of Angadi- Harris N B W 1985 Alkaline complexes from the Arabian mogar syenite, Kerala and its taphrogenic affiliation; Shield; J. Afr. Earth Sci. 3 83–88. J. Geol. Soc. India 27 494–507. Huang W and Wyllie P J 1975 Melting reactions in the Searle M P, Parrish R R, Hodges K V, Hurford A, system NaAlSi3O8–KAlSi3O8–SiO2 to 35 kbars, dry and Ayres M W and Whitehouse M J 1997 Shisha Parigma with excess water; J. Geol. 83 737–748. leucogranite, South Tibetan Himalaya: field relations, Inger S and Harris N 1993 Geochemical constraints on geochemistry, age, origin, and emplacement; J. Geol. 105 leucogranite magmatism in the Langthan Valley, Nepal 295–307. Himalayas; J. Petrol. 34 345–368. Shand S J 1943 The Eruptive rocks, 2nd edn (New York: Le Bas M J and Streckeisen A L 1991 The IUGS systematics John Wiley) 444p. of igneous rocks; J. Geol. Soc. London 148 825–833. Streckeisen A L 1976 To each plutonic rock its proper name; Le Bas M J, Le Maitre R W, Streckeisen A and Zanettin B Earth Sci. Rev. 12 1–33. 1986 A chemical classification of volcanic rocks based on Streckeisen A L and Le Maitre R W 1979 A chemical approx- the total alkali-silica diagram; J. Petrol. 27 745–750. imation to the modal QAPF classification of the igneous Le Maitre R W, Streckeisen A, Zanettin B, Le Bas M J, rocks; Neu. Jahr. Mineral. 136 169–206. Bonin B, Bateman P, Bellieni G, Dudek A, Efremova S, Strong D F and Hammer S K 1981 The leucogranites of Keller J, Lameyre J, Sabine P A, Schmid R, Sørensen H southern Brittany: origin by faulting, frictional shear- and Woolley A R 2002 Igneous rocks: A classification ing, fluid flux, and fractional melting; Can. Mineral. 19 and Glossary of Terms; 2nd edn, Cambridge Univ. Press, 163–176. 252 p. Zhao J-X, Shiraishi K, Ellis D J and Sheraton J W 1995 Morimoto N 1988 Nomenclature of pyroxenes; Mineral. Geochemical and isotopic studies of syenites from the Mag. 52 535–550. Yamato mountains, East Antarctica: implications for the Nabelek P I, Russ-Nabelek C and Denison J R 1992 Harney origin of syenitic magmas; Geochem. Cosmochim. Acta Peak leucogranite, Black Hills, South Dakota, USA: 59 1363–1382.

MS received 6 July 2005; revised 23 August 2005; accepted 24 August 2005