Occurrences and chemical compositions of ultrapotassic mafic dyke rocks from Skallevikshalsen and Rundvågshetta, Lützow-Holm Complex, East Antarctica

Abstract

Melanocratic dyke rocks were found at Skallevikshalsen and Rundvågshetta in the Lützow-Holm Complex. The rocks are commonly holocrystalline and aphyric, and consist mainly of alkali feldspar, biotite, augite, quartz, apatite, and titanite, with minor amounts of plagioclase. Almost all the mineral compositions tended to be homogeneous, whereas the composition of apatite sometimes showed intragranular heterogeneity in the F and Cl components. The melanocratic dyke rocks at Skallevikshalsen have an ultrapotassic mafic composition and resemble the compositional features of the lamproitic to minettic dyke rock previously found at Innhovde, located in the western region of the Lützow-Holm Complex. The dyke rocks at Rundvågshetta are considered to be a mixture of ultrapotassic mafic magma and high-Cl/F intermediate to felsic magma. Considering their occurrence and the results of Rb-Sr mineral dating, the time of intrusion was just after the metamorphism of the Lützow-Holm Complex, and the igneous activity was thought to have been caused by the collision between East and West Gondwana.

INTRODUCTION

The history of metamorphism at the continental margin and crustal collision areas reveals the crustal formation and development. The subsequent igneous activity suggests that mantle activity is associated with the crustal development. In the later period of the Caledonian orogenic at Scotland, intrusive activity of minette, a kind of ultrapotassic lamprophyre, was common and the lithospheric mantle was involved in the generation of magmas associated with the orogenic movement (Canning et al., 1996). Prelević et al. (2004) reported ultrapotassic volcanic rocks formed by mixing between dacite derived from orogenic activity and lamproite originating from the mantle, which is related to the intracontinental post-collisional collapse and lithosphere delamination followed the closure of Vardar Tethys at the end of the Mesozoic. During the convergence of continental crust accompanying the collision of East and West Gondwana (Supplementary Fig. S1; Supplementary Figs. S1-S2 are available online from https://doi.org/10.2465/jmps.221201), Owada et al. (2013) reported igneous activity accompanied by the minette intrusions at the Sør Rondane Mountains, eastern Dronning Maud Land, East Antarctica. They considered that the activity originated from the orogenic movement associated with the collision and stated that the enriched mantle was involved in the formation of minette. Similarly, igneous activity with mafic magma after large-scale orogenic movements is generally associated with enhanced mantle activity related to intracontinental tectonic settings, post-orogenic collapse, post-dating convergent tectonics, and active marginal processes (Mitchell and Bergman, 1991). In general, magma enriched with incompatible elements is generated in such areas, and the enriched mantle is often involved in such igneous activity. Therefore, verifying the influences, results, and contributions of metamorphism and orogenic activity from such igneous activity, particularly from the chemical compositions, constituent minerals, and their characteristics is possible.

The Lützow-Holm Complex (LHC) of Dronning Maud Land, East Antarctica is considered to have metamorphosed between 600-520 Ma (e.g., Dunkley et al., 2020, and reference therein, Kitano et al., 2021, 2023; Mori et al., 2023), and is also located in a part of the convergent zone of East and West Gondwana (Fig. S1). In a field survey during the 52nd Japanese Antarctic Research Expeditions (JARE-52), a few melanocratic dykes that cut across the regional structures were found in Skallevikshalsen and Rundvågshetta of the LHC (Fig. 1a). Their petrogenesis has not been considered so far; however, detailed geochemistry on these dykes will contribute to our understanding of the formation and development processes of the continental crust after the high-grade metamorphism in the LHC. In this study, we present the petrology and chemical characteristics of the dyke rocks, and discuss the mechanisms of crustal formation in the LHC.

Figure 1. Location map in the LHC and geological map of Skallevikshalsen. (a) Location map of Skallevikshalsen, Rundvågshetta (related to Figure 3), and related exposures in the LHC. (b) Geological map of Skellevikshalsen (simplified from Yoshimura et al., 2004), showing locality of melanocratic dyke rock with sampling code.

BACKGROUND

Geological setting of the LHC

The LHC of the Dronning Maud Land, situated west of the Rayner Complex and east of the Yamato-Belgica Complex of East Antarctica, is a high-grade metamorphic terrane within the East Antarctic Shield, where small coastal exposures are found between 45° and 37°E (Fig. S2). Beginning in 1956, JARE conducted detailed surveys of its geology and tectonics and identified various types of metamorphic rocks, granites, pegmatites, and intermediate-to-mafic rocks that intruded during different stages of tectonism, especially during and after peak metamorphism. The metamorphic grade of the LHC increases from the upper amphibolite facies in the NE of LHC to the granulite facies in the SW, with a thermal maximum at Rundvågshetta (Hiroi et al., 1991) (Fig. S2). Petrological studies have shown that the LHC experienced metamorphism with a clockwise P-T path and isothermal decompression after the thermal peak (e.g., Motoyoshi et al., 1989; Hiroi et al., 1991; Mori et al., 2023) from 600-520 Ma (Shiraishi et al., 1994; Hokada and Motoyoshi, 2006; Shiraishi et al., 2008; Dunkley et al., 2020; Kitano et al., 2021, 2023; Mori et al., 2023). Previous studies have suggested the possibility of ultrahigh-temperature (UHT) metamorphism at Rundvågshetta (Motoyoshi et al., 2006; Yoshimura et al., 2008). According to a recent study, the LHC is subdivided in to the Innhovde Suite (INH, 1070-1040 Ma) composed mainly of felsic orthogneiss; the Rundvågshetta Suite (RVG, 2520-2470 Ma), mostly felsic orthogneiss with minor mafic and metasedimentary gneisses; the Skallevikshalsen Suite (SKV, 1830-1790 Ma), composed of felsic to mafic orthogneiss with abundant dolomitic marbles, calc-silicates and other metasediments; the Langhovde Suite (LHV, 1100-1050 Ma), mostly felsic orthogneiss with minor mafic and calc-silicate gneisses; the East Ongul Suite (EOG, 630 Ma), with various orthogneisses and metasediments; and the Akarui Suite (AKR, 970-800 Ma) with diverse orthogneisses and paragneisses, from southwest to northeast, based on origin of protolith defined by U-Pb zircon ages (Dunkley et al., 2020). In the subdivisions, the oldest crustal components of the LHC lie in the southern part of Lützow-Holm Bay and consist of late Neoarchean and Paleoproterozoic protoliths to charnockites and enderbites that dominate the Rundvågshetta and Skallevikshalsen Suites. This older domain is surrounded by gneisses and granulites with late Mesoproterozoic to early Neoproterozoic protolith ages, including the Innhovde and Langhovde suites (Dunkley et al., 2020). Recently, some metamorphic analyses and chronological studies have been carried out on the outcrops of the NE LHC, and the detailed geological background related to continental evolutions of the LHC is being clarified in this region (e.g., Kitano et al., 2021, 2023; Baba et al., 2023; Mori et al., 2023).

Post-peak metamorphic basic intrusions in the LHC

Late-to post-tectonic magmatism was previously found in some exposures to the LHC. Most of these are granitic emplacements and are common in the NE to middle regions of the LHC. Among them, A-type granite may have been generated by partial melting during UHT metamorphism (e.g., Hiroi et al., 2019; Carvalho et al., 2023). However, some mafic to intermediate dyke rocks that discordantly intruded into the surrounding gneisses, were found in the Akebono Rock (Hiroi et al., 1986), Cape Hinode (Yanai and Ishikawa, 1978), Skallen (Yoshida et al., 1976; Osanai et al., 2004), Rundvågshetta (Motoyoshi et al., 1986), and Botnneset (Shiraishi and Yoshida, 1987) from the northeast to the southwest of the LHC. Among them, the mafic dykes from the Akebono Rock were basalt to andesite in nature, intruded in the N-S to NW-SE direction, and cut the foliation of the host gneisses. Even though they are thermally metamorphosed, they usually preserve the igneous texture, contain two igneous pyroxenes in many cases and have no evidence of mylonitization (Hiroi et al., 1986). The metamorphosed andesite was not enriched in alkali compositions (Na₂O: 3.35 wt%, K₂O: 0.10 wt%, and Rb: 1.47 ppm), and other dyke rocks also contain very little Rb (1.71 and 4.13 ppm) (Hiroi et al., 1986). Dyke rocks from Cape Hinode were modified to amphibolite, which is composed mainly of hornblende, biotite, and plagioclase with minor amounts of quartz and potassic feldspar, while preserving the original igneous texture as blastoporphyritic plagioclase partly (Yanai and Ishikawa, 1978). Some of them intruded in the NE-SW direction, others had a N-S trend, and all cut the foliation of host gneisses. In the southern part of Skallen, discordant metabasite dyke occurs cutting the paragneiss-metabasite alternations (Yoshida et al., 1976). It is mesocratic, dark grey with a brownish tint, small-grained, and equigranular rock composed of plagioclase, rhombic and monoclinic pyroxenes, hornblende, biotite, as well as opaque minerals. Previously reported dykes in Rundvågshetta were holocrystalline cpx amphibolites with a dark black tone on the ground, and intruded subvertically along a general N-S direction (Motoyoshi et al., 1986; Ishikawa et al., 1994). In Botnneset, mafic rock dykes were discovered in Austhovde, Vesthovde, and Innhovde (Shiraishi and Yoshida, 1987). These dykes also discordantly intrude into the surrounding gneisses and are metamorphosed. Among them, the melanocratic mafic dykes in Innhovde metamorphosed the ultrapotassic rocks of minette or lamproite affinities (Arima and Shiraishi, 1993). The dyke rock was composed of potassic feldspar, biotite, augite, and apatite with subsequent quartz, plagioclase, titanite, and opaque minerals. Moreover, it had a foliation parallel to the direction of the dyke intrusion, which was the same as that of the dyke and host gneiss in Vesthovde (Shiraishi and Yoshida, 1987). Arima and Shiraishi (1993) reported K-Ar whole rock age of 434.6 ± 21.7 Ma for the mafic dyke rock, and their highly potassic composition was regarded as a manifestation of post-orogenic igneous activity and initial stage of continental rift system linked to the Pan-African Orogeny.

GEOLOGY AND OCCURRENCES OF MELANOCRATIC DYKE ROCKS

Skallevikshalsen

The Skallevikshalsen is an exposure located in the south-western part of the LHC within the granulite facies zone, approximately 70 km southwest of Syowa Station on Ongul Islands (Fig. 1a). According to Dunkley et al. (2020), this is a representative site of the SKV Suite. Yoshida et al. (1976) reported the occurrence of orthogneiss (brown gneissose granodiorite and garnet gneissose granite), paragneiss, marble associated with skarn, and minor amounts of metabasite lenses as well as layers within the orthogneiss and paragneiss in Skallevikshalsen. They also reported an overturned synform with an ENE-WSW axis, which is considered an overturned recumbent anticline (or nappe) (Fig. 1b). Open to gentle folds with steep axial planes and wavelengths of several hundred meters are developed on the northeastern and northwestern shores of Skallevikshalsen. In addition, minor folds also developed well, and their orientations were either parallel or discordant with those of the abovementioned folds. The banding, foliation, or gneissosity of metamorphic rocks are also largely parallel to the synform; however, partly undergo tectonic disturbances. Crenulation and mineral lineations with gentle east or west plunges nearly parallel to the axis of minor folds of the close-to-open type are well developed in association with the planar structures (Yoshida et al., 1976). Two ductile deformation events can be recognized from the fold interference patterns (Kawakami and Ikeda, 2004). Yoshimura et al. (2004) grouped the lithologies into orthopyroxene felsic gneiss, garnet-sillimanite gneiss, garnet-biotite gneiss, and crystalline limestone, with subordinate garnet-spinel-sillimanite gneiss and mafic gneiss. Moreover, estimated metamorphic conditions of 770-940 °C and 0.65-1.1 GPa for garnet-biotite gneiss and 780-960 °C and 0.6-1.1 GPa for mafic gneiss were also organized, which is consistent with independent estimates of peak metamorphic conditions reported from metacarbonate rocks (850-870 °C; Mizuochi et al., 2010). Based on the metamorphic conditions and mineral occurrences, Yoshimura et al. (2004), Kawakami and Motoyoshi (2004), and Kawakami et al. (2016) suggested that metamorphic rocks experienced partial melting during high-grade metamorphism. Kawakami et al. (2016) reported that the activity of fluids containing Cl is involved in partial melting.

Although the intrusive rocks were less voluminous than the widespread metamorphic rocks, the emplacement of the melanocratic dyke was recognized during field surveys in the northwest of Skallevikshalsen (Fig. 1b). The dyke was 10-50 cm width intrusive to 20 m length (Fig. 2a), dipped steeply to the east, oriented mostly to the N-S direction, and cut the foliation of the surrounding host gneisses sharply. The boundary between the dyke and surrounding host gneisses had no reaction zones or textural features. The dyke rarely contains fragments of surrounding rocks as xenoliths.

Figure 2. Occurrences of melanocratic dyke rocks at the LHC. (a) Melanocratic dyke rock from Skallevikshalsen (TM11010406). The dyke with width around 20 cm has intruded in the surrounding metamorphic rocks cutting their gneissosity. (b) Melanocratic dyke rock from central Rundvågshetta (TM11012302). The dyke with about 40 cm thickness (partly wider) has intruded in the surrounding metamorphic rocks cutting their gneissosity. (c) Melanocratic dyke rock from northern Rundvågshetta (TM11012407). A dyke with 10 cm thickness is about to branch out while cutting through the gneiss. (d) Melanocratic dyke and the previously intruded black-colored dyke that cut by the melanocratic dyke from central Rundvågshetta (north of TM11012506). This black-colored dyke is composed of cpx-amphibolite (Motoyoshi et al., 1986). (e) A distant view of the south from the point of TM11012306. A dyke at TM11012506 can be seen on the surface of cliff in the foreground. Black dykes that dip to the west (upper left to lower right in the photo) are cpx-amphibolite, while those that dip to the east (upper right to lower left in the photo) are melanocratic dyke described in this study. The extension of this dyke can also be found on the cliffs in the back. (f) Melanocratic rock produced by being incorporated into pegmatite at Rundvågshetta (TM11012605). A melanocratic rock is recognized in the black part of the upper part of the dyke, and its fragments are also incorporated into the pegmatite. (g) Enlarged photo of the pegmatite (TM11012605). Melanocratic rock is partially present as fragments in pegmatite.

Rundvågshetta

The Rundvågshetta is an exposure spreading approximately 2.5 km north-south and 2.5 km east-west, on the coastal side of Lützow-Holm Bay, and about 30 km southwest of Skallevikshalsen (Fig. 1a). According to Dunkley et al. (2020), this is the type area for the RVG Suite. At this locality, several types of well-layered metamorphic rocks mainly garnet-sillimanite gneisses, garnet-biotite gneisses, and pyroxene gneisses with subordinate calc-silicate rocks are distributed (Fig. 3). Pyroxene gneisses were charnockitic and the most common lithology, whereas intercalated garnet-biotite and garnet-sillimanite gneisses predominated in the northern area. The orthopyroxene + sillimanite ± quartz assemblages revealed high T and P conditions of peak metamorphism in the granulites of Rundvågshetta. Yoshimura et al. (2008) reported the occurrence of sapphirine + quartz associations within garnet porphyroblasts in the garnet-orthopyroxene-sillimanite granulites from Rundvågshetta as evidence of ultrahigh-temperature (UHT) metamorphism. According to the results, they suggested a clockwise P-T trajectory with steep isothermal decompression near the peak UHT metamorphism, based on the occurrence of orthopyroxene + sillimanite + quartz and peak metamorphic temperature of about 1000-1100 °C obtained by ternary feldspar thermometry in common gneisses in this area. In addition, Yoshimura et al. (2008) suggested that these rocks may have partially melted during UHT metamorphism.

Figure 3. Geological map of Rundvågshetta (simplified from Ishikawa et al., 1994) showing locality of melanocratic dykes with sampling codes.

Most metamorphic rocks exhibit a mineral alignment, defined by sillimanite, orthopyroxene, hornblende, and quartz. General foliation of the gneisses trends WNW-ESE dipping 50-70° southward in the central and southern areas. In contrast, Ishikawa et al. (1994) reported that regional strikes were disturbed in the northern area because of low-strain ductile deformation, represented by E-W extension and N-S compression after peak metamorphism (Ishikawa et al., 1994). Moreover, they showed that mineral fabrics formed at peak metamorphism imply regional ductile deformation during peak metamorphism, which was characterized by the formation of a WNW-ESE subhorizontal mineral lineation and the development of high-strain structures such as isoclinal buckling folds and boudinage. They revealed that the regional high-strain deformation had ceased before the reactions formed cordierite-bearing symplectic intergrowth, which was common in decompression. The last structural event in Rundvågshetta is represented by the emplacement of felsic pegmatites (Motoyoshi et al., 1986; Ishikawa et al., 1994).

Although intrusive rocks were less voluminous than widespread metamorphic rocks, a few emplacements of cpx amphibolite and other melanocratic dykes were recognized in field surveys at Rundvågshetta (Figs. 2d and 2e). The cpx amphibolite intruded as a dyke with low-strain deformation and was influenced by the emplacement of the subsequent felsic pegmatite (Motoyoshi et al., 1986). The dykes are dark colored and are composed mainly of clinopyroxene, hornblende, plagioclase, and opaque minerals, with minor amounts of potassic feldspar, biotite, and quartz. They intrude into the N-S trending subvertical fractures perpendicular to the mineral lineations exhibited by most metamorphic rocks (Motoyoshi et al., 1986; Ishikawa et al., 1994). In contrast to the peak metamorphic high-strain deformation, the post-peak metamorphic structural evolution was characterized by low-strain and localized deformations such as gentle folding, emplacement of cpx amphibolite, and local shearing in the N-S compressional and E-W extensional fields.

Two melanocratic dykes were also found in Rundvågshetta during the field survey; one of them extended over 2 km through from north to south of the central area of Rundvågshetta while forming an echelon arrangement, and another extended over 200 m, at least on the western side of the area (Fig. 3), although they were partly covered by snow or rubble stones. The dykes were a few to 50 cm thick, dipping steeply to the east, were largely NS to NNE-SSW striking, and sharply cut the foliation of the surrounding host gneisses (Figs. 2b-2d). Such intrusive conditions are consistent with previously reported cpx amphibolites (Motoyoshi et al., 1986; Ishikawa et al., 1994), although the melanocratic dykes cut the cpx amphibolites (Figs. 2d and 2e). Part of the melanocratic dyke rocks was subsequently disturbed and replaced by felsic pegmatite (Figs. 2f and 2g).

PETROGRAPHY OF MELANOCRATIC DYKE ROCKS

The melanocratic dyke rocks from Skallevikshalsen and Rundvågshetta are commonly holocrystalline and aphyric. They consisted mainly of alkali feldspar, biotite, quartz, apatite, titanite, and with minor amounts of plagioclase (Fig. 4). Augite is usually present, except in some dykes. Alkali feldspar is mainly orthoclase and sometimes perthitic. The grain size is approximately 0.1-1 mm; however, it occasionally grows up to 3 mm. The biotite has a grain size of 0.5-2 mm and is sometimes occur as coarse grains up to 5 mm. They tend to be arranged linearly parallel to the intrusive direction of the dykes. Augite is granular to columnar, and their grain size is mostly between 0.1-0.5 mm, and occasionally coarse up to 2 mm. Augite is often surrounded by biotite and is sometimes filled with fine biotite along the cleavage (Figs. 4a and 4b). Apatite is granular and idiomorphic, with a grain size mainly 0.1-1 mm, and can grow to 3 mm. Titanite is granular and subhedral; therefore, the corners of the crystals are often rounded. Its grain size is approximately 0.1-0.5 mm. The quartz and plagioclase are xenomorphic, and the grain size is approximately 0.1-1 mm. In addition, zircon and monazite was rarely found as accessory minerals. Based on the mineral assemblages, the melanocratic rocks were distinguished from the cpx amphibolite dykes (Motoyoshi et al., 1986). The crystals often grow in parallel arrays in the intrusive direction of the dyke. At Rundvågshetta, the melanocratic rock from the northern end of the dyke (TM11012605E. It will be abbreviated as 12605E in the text from now on. The same applies for other sample codes below) was enriched with quartz (Fig. 4f).

Figure 4. Photomicrograph of representative melanocratic dyke rocks from Skallevikshalsen (a) and Rundvågshetta (b)-(f) in the LHC. Plane polarized light. (a) TM11010406A, (b) TM11012302C, (c) TM11012204A, (d) TM11012501A, (e) TM11012407A, and (f) TM11012605E. All scale is 1 mm. When accompanied by augite, the constituent minerals are generally fine-grained (a), (b), and (c), but sometimes they are coarse-grained (d). Dyke rocks free from pyroxene tend to have coarse-grained (e) and (f). Afs, alkali feldspar; Aug, augite; Bt, biotite; Pl, plagioclase; Qtz, quartz; Ttn, titanite; Opq, opaque mineral.

ANALYTICAL PROCEDURES

Representative minerals in the melanocratic dyke rocks were observed using a JEOL JSM-5800LV secondary electron microanalyzer (SEM) and their compositions were analyzed using a JEOL JXA-8530F wavelength-dispersive electron microprobe (WDS) at Kyushu University. For compositional analyses using the JEOL JXA-8530F, the counting times for all elements were 20 s for the peak position and 10 s for the background on each side of the peak. The operating conditions were 15 kV and 6 nA and natural minerals were used as standards.

Whole-rock samples of the dykes were analyzed for the composition of major elements and minor elements (Sc, V, Cr, Ni, Cu, Zn, Rb, Sr, Y, Zr, Nb, Ba, and Pb); they were determined by X-ray fluorescence spectrometry (XRF: Rigaku-Primus IV) at Kyushu University following calibration techniques using glass discs (Nakada, 1985; Nakada et al., 1985). Fluorine and Cl contents of the whole-rock samples were analyzed using the same XRF system with pressed discs, as described by Nakada (1987). Whole-rock samples were crushed (200-500 g), powdered using a ball mill, and then pulverized in an agate mortar. For pegmatite, a 5 kg sample (12605C) was prepared to avoid compositional deviation due to the heterogeneous nature of the coarse grains. They were divided into five groups (each weighing 1 kg) and pulverized. Five groups of powdered samples were subjected to XRF analysis, twice for each group. After 10 analyses of the pegmatite fractions, their compositions varied very little from the average composition (Supplementary Table S1; Supplementary Tables S1-S2 are available online from https://doi.org/10.2465/jmps.221201). Hence the average composition was adopted as the representative of pegmatite. Some lanthanoids compositions (e.g., La, Ce, Nd, Sm, Eu, Gd, Dy, Er, and Yb) were determined for some of the dyke rock samples by inductively coupled plasma atomic emission spectrometry (ICP-AES; Seiko-SPS 1200AR) after decomposition by HF and HNO₃ and conditioning with diluted HNO₃ at Kyushu University, following the methods of Miyamoto et al. (2004). The REE composition of pegmatite was obtained from the sample that had the closest major element composition to the average values. The full lanthanoids were analyzed using inductively coupled plasma mass spectrometry (ICP-MS, Agilent 7900) for a few samples.

For geochronological research, the biotite and felsic fractions were separated from crushed rock samples 10406H from Skallevikshalsen and 12302C from Rundvågshetta. The fractions were concentrated by magnetic separation using a Frantz isodynamic separator and gravity separation with tetrabromoethane. The mineral separates were further purified manually under a binocular microscope. For the Rb-Sr isotopic analysis, 100-200 mg of each mineral fraction was used. The fractions were cleaned with acetone and rinsed with purified water prior to decomposition. The samples were digested with HNO₃-HF in a vial and heated to dryness. Conventional isotope dilution methods were applied to determine the Rb and Sr concentrations in the mineral fractions and in their original rock samples for dating by spiking them with ⁸⁷Rb and ⁸⁴Sr. The decomposed samples of each fraction were dissolved in a mixture of hydrochloric acid and oxalic acid and passed through a DOWEX 50W-X8, 200 mesh cation exchange resin to separate Rb and Sr by following the method of Miyamoto et al. (2004). The Sr isotope compositions as well as the total Rb and Sr concentrations were determined by TIMS using a Finnigan MAT-262 system at Okayama University. The strontium standard NBS987 yielded ⁸⁷Sr/⁸⁶Sr = 0.71023 ± 0.00001 (2σ). The relative analytical errors for Rb and Sr concentrations were 2 and 1%, respectively. The contamination levels of Rb and Sr were less than 1 × 10⁻¹⁰ and 1 × 10⁻¹⁰ g per sample, respectively. The decay constant used for the age calculations was 1.42 × 10⁻¹¹ year⁻¹ for ⁸⁷Rb (Steiger and Jäger, 1977). Rb-Sr isochrons were determined using IsoplotR (Vermeesch, 2018).

RESULTS

Mineral compositions analyzed by electron microscope

The average compositions of the constituent minerals in the representative dyke rocks from Skallevikshalsen and Rundvågshetta are shown in Table S2. Alkali feldspar in the dyke rock from Skallevikshalsen (10406A) is potassic with an 83% Or component, and those from Rundvågshetta have 80-86% Or components that rarely decrease to 76% Or component. The augite in the dyke rock from Skallevikshalsen showed an average composition of #Mg [= Mg/(Mg + Fe)_mol] = 0.75, and that from Rundvågshetta showed a composition of #Mg = 0.74-0.82. Biotite in the dyke rock from Skallevikshalsen had an average composition with #Mg = 0.68, TiO₂ = 4.08 wt%, and Cl/F = 0.21. In contrast, the average #Mg of biotite in those from Rundvågshetta varies from 0.53 to 0.75, TiO₂ = 4.63-5.57 wt% and Cl/F = 0.06-1.81 among the dyke rock samples; however, they were almost the same value within each rock samples. The #Mg of biotite coexisting with augite is relatively high (= 0.64-0.75) and such biotite shows Cl/F = 0.06-0.38 (e.g., 10406A, 12302C, 12204A). However, the #Mg of biotite in samples free from augite is low (= 0.53-0.61) and such biotite shows higher Cl/F ratios (= 0.48-1.81) (e.g., 12407A, 12605E). Although the number of measurements is small because they occur as minor phase in rock samples, plagioclase in dyke rocks from Rundvågshetta showed 22-25% An components and rarely increased to 44%.

Apatite sometimes shows intragranular heterogeneity in the F and Cl components, although the major elemental compositions tend to be homogeneous. Most of the apatite in the melanocratic dyke rocks were a solid solution between fluorapatite and chlorapatite and are devoid of hydroxyl component. Almost all apatite has only F-rich domains (>90% fluorapatite component) (e.g., 10406A, 12204A, and 12302C) (Figs. 5a and 5b), whereas apatite in a few rocks from Rundvågshetta (12407A and 12605E) is composed of an inner Cl-rich domain and an outer F-rich portion (Figs. 5c and 5d). Despite their euhedral outline, the F-rich portion exhibited a texture that exhibits a dissolution-reprecipitation of the Cl-rich domain.

Figure 5. Representative ClKα image with PKα image (inset) for the apatite in the melanocratic rocks from Skallevikshalsen and Rundvågshetta. (a) TM11010406, (b) TM11012302C, (c) TM11012407A, and (d) TM11012605E. The circles in the figures are the points analyzed by WDS, and the attached numbers indicate the concentration ratios of Cl/F (wt%) and the stoichiometric Cl/F/OH ratios calculated from analytical results. The fact that the inner F-rich part is found in the apatite in Figure 5d may be due to the infiltration of fluorine along the cracks of the apatite and the alteration of the inner domains of the apatite.

Whole rock compositions

The analytical results for the major and minor element compositions of the melanocratic dyke rocks and related rocks from the LHC are listed in Table 1. Harker diagrams for the dyke rocks are shown in Figures 6 and 7. The melanocratic dyke rocks from Skallevikshalsen and Rundvågshetta have alkalic and mafic compositions (Figs. 6a and 6e) and are categorized as minette to lamproite, with ultrapotassic compositions of K₂O >3 wt%, MgO >3 wt%, and K₂O/Na₂O > 2 (Foley et al., 1987). In addition, on the CaO-Al₂O₃ discrimination diagram for ultrapotassic igneous rocks after Foley et al. (1987), the compositions of the melanocratic dyke rocks from Rundvågshetta are classified into Group I, which is grouped into representative lamproite (Fig. 6h). In contrast, the compositions of the dyke rocks from Skallevikshalsen plot in the transitional area between Groups I and III (i.e., Group IV) in the same discrimination diagram. Because of their characteristic compositions, these rocks should be considered representative of the origin of ultrapotassic mafic rocks, including their relationship with the formation conditions.

Table 1. Whole-rock composition of melanocratic dyke rocks and pegmatite from Skallevikhalsen and Rundvågshetta, in LHC

Locatiom	Skallevikshalsen			Rundvågshetta
	TM110104 06A	TM110104 06G	TM110104 06H	TM110121 02A	TM110121 03A	TM110121 03F	TM110121 04A	TM110122 01A
(wt%)
SiO2	48.89	44.74	45.09	50.15	55.35	55.70	53.60	53.57
TiO2	2.86	3.19	3.18	3.20	2.65	2.75	3.02	3.03
Al2O3	10.90	11.27	11.21	10.48	10.81	11.31	10.94	10.92
Fe2O3	7.06	6.80	6.42	10.10	5.84	6.70	6.18	6.77
MnO	0.09	0.08	0.08	0.08	0.07	0.04	0.06	0.08
MgO	7.33	8.68	8.71	7.18	6.72	5.99	6.72	7.93
CaO	7.61	8.36	8.41	5.39	5.49	5.10	5.79	5.62
Na2O	0.89	0.83	0.82	0.51	1.09	0.88	1.26	1.14
K2O	7.98	8.48	8.39	5.53	6.86	5.76	6.41	5.13
P2O5	4.06	4.87	4.91	3.59	3.72	3.35	3.83	3.56
Total	97.67	97.30	97.22	96.20	98.59	97.57	97.80	97.75
(wt%)
F	0.17	0.19	0.19	0.08	0.17	0.12	0.12	0.14
Cl	0.12	0.13	0.12	0.23	0.11	0.28	0.13	0.18
(ppm)
Sc	11.13	8.27	10.97	9.37	8.84	14.96	6.46	8.60
V	127.92	146.92	148.53	137.93	103.58	121.20	103.86	106.59
Cr	239.00	303.52	298.03	260.16	237.56	197.86	215.85	302.54
Ni	166.08	237.00	233.56	145.81	167.38	42.53	178.05	200.29
Cu	u.d.	u.d.	u.d.	1.94	u.d.	8.38	u.d.	11.33
Zn	127.03	120.92	113.98	148.53	105.44	210.00	91.95	115.76
Rb	295.16	313.51	305.12	361.89	519.04	223.40	270.31	324.46
Sr	1643.71	2073.17	2262.09	818.62	1458.25	1019.38	1874.90	928.53
Y	50.70	50.80	49.88	56.53	49.54	54.49	41.02	38.90
Zr	176.99	95.45	162.25	479.58	315.34	320.55	107.13	54.38
Nb	15.79	10.79	14.40	26.94	27.99	28.86	19.69	16.94
Ba	2047.54	2091.30	2149.80	1482.24	1166.46	1899.41	1234.96	1738.53
Pb	8.72	7.78	8.66	10.42	20.75	14.13	11.25	7.01
(ppm)
La	130.53	130.95		124.63	194.80	84.72		143.12
Ce	283.91	286.08		252.45	409.15	193.71		339.25
Pr
Nd	127.92	139.67		123.11	185.46	103.53		168.39
Sm	24.19	25.24		25.75	29.99	22.53		26.73
Eu	5.69	5.92		3.96	5.89	2.75		4.27
Gd	14.15	14.53		14.55	15.48	14.10		13.94
Tb	2.45	2.58				2.37
Dy	7.51	7.52		6.39	6.15	7.22		5.59
Ho	1.43	1.06			1.01	1.33
Er	3.04	2.71		2.31	1.94	3.19		1.87
Tm	0.30	0.47				0.46
Yb	2.25	1.76		1.88	1.30	2.61		1.20
Lu

Major and minor elements with F and Cl compositions were analyzed XRF system (Rigaku Primus IV), and lanthanoid compositions were analyzed by using ICP-AES system (Seiko-SPS 1200AR) mainly and ICP-MS system (Agilent 7900) for some samples (TM11012204A, 2501A, and 2506B from Rundvågshetta). Composition of pegmatite from Rundvågshetta (TM11012605C) is average of 10 analytical data (Supplementary Table S1). As a reference, the composition of ultrapotassic mafic rock from Innhovde (TM11020201A), which was collected during geological survey at JARE-52, was also analyzed and listed.

Locatiom
	TM110122 01B	TM110122 02A	TM110122 02J	TM110122 03A	TM110122 04A	TM110122 05A	TM110122 05C	TM110123 02C
(wt%)
SiO2	57.06	53.98	53.12	51.92	52.54	53.61	55.18	52.61
TiO2	2.87	2.81	2.86	2.75	2.52	2.62	2.80	2.68
Al2O3	10.31	10.87	11.26	10.85	11.30	11.40	11.15	11.35
Fe2O3	7.23	8.06	6.70	9.46	7.95	5.23	6.64	4.92
MnO	0.05	0.08	0.06	0.08	0.07	0.05	0.06	0.05
MgO	6.41	7.47	7.22	7.16	7.52	7.72	7.23	7.01
CaO	4.40	5.11	6.51	5.74	6.18	6.20	6.11	6.00
Na2O	0.69	1.21	1.26	0.93	1.17	1.08	1.02	1.14
K2O	6.06	6.52	6.79	7.00	6.81	7.26	6.80	7.21
P2O5	2.98	3.11	4.10	3.86	3.99	4.18	4.02	3.90
Total	98.05	99.22	99.88	99.74	100.05	99.33	101.01	96.85
(wt%)
F	0.10	0.10	0.13	0.11	0.14	0.14	0.11	0.11
Cl	0.18	0.17	0.06	0.21	0.13	0.01	0.18	0.00
(ppm)
Sc	10.95	9.73	6.42	5.61	8.09	9.45	5.10	4.39
V	114.45	122.75	113.16	145.46	98.03	104.66	107.38	85.83
Cr	213.48	240.33	250.78	258.26	293.62	249.28	235.66	225.33
Ni	177.98	188.36	184.37	170.11	206.71	226.72	200.51	209.32
Cu	8.21	4.83	22.12	1.29	4.29	9.38	9.64	u.d.
Zn	110.37	144.24	88.64	124.60	121.65	86.54	112.59	67.17
Rb	363.32	496.26	265.01	431.11	437.66	240.27	310.22	374.90
Sr	603.17	685.69	2414.30	835.30	1447.84	3525.49	1732.02	3561.48
Y	49.34	46.48	43.42	52.60	47.15	35.39	43.20	39.58
Zr	581.03	102.28	87.34	60.19	54.71	65.46	227.91	91.88
Nb	45.17	19.16	12.17	15.01	7.09	11.74	22.80	11.48
Ba	1948.47	1542.96	1145.34	1614.79	1171.55	1142.89	1242.98	1141.50
Pb	13.81	14.61	7.62	20.01	8.38	6.48	13.37	5.41
(ppm)
La		111.58	107.90	134.41	134.63	113.38	109.80	122.91
Ce		222.01	245.71	274.55	292.15	250.71	252.45	259.55
Pr					35.76
Nd		105.78	130.86	128.80	150.15	126.45	128.94	140.85
Sm		20.50	25.82	24.52	24.82	24.94	24.58	27.02
Eu		3.65	4.86	3.53	5.20	4.83	5.08	5.52
Gd		11.00	12.41	12.95	14.63	13.88	13.01	14.67
Tb					1.45			1.99
Dy		4.02	5.51	4.92	5.40	5.11	5.20	5.05
Ho					0.76			1.11
Er		1.66	1.57	2.09	1.42	1.31	1.76	1.64
Tm					0.17			0.21
Yb		1.13	1.10	1.49	0.94	0.90	1.06	0.93
Lu					0.10

Major and minor elements with F and Cl compositions were analyzed XRF system (Rigaku Primus IV), and lanthanoid compositions were analyzed by using ICP-AES system (Seiko-SPS 1200AR) mainly and ICP-MS system (Agilent 7900) for some samples (TM11012204A, 2501A, and 2506B from Rundvågshetta). Composition of pegmatite from Rundvågshetta (TM11012605C) is average of 10 analytical data (Supplementary Table S1). As a reference, the composition of ultrapotassic mafic rock from Innhovde (TM11020201A), which was collected during geological survey at JARE-52, was also analyzed and listed.

Locatiom							Innhovde	Pegmatite (Rundvågshetta)
	TM110123 04A	TM110124 07A	TM110125 01A	TM110125 06B	TM110126 05B	TM110126 05E	TM110202 01A	TM11012605C
(wt%)
SiO2	52.67	53.53	57.32	54.90	56.58	62.19	52.69	70.32
TiO2	2.82	3.02	3.03	2.81	2.87	2.78	1.15	0.59
Al2O3	10.89	10.21	10.00	9.97	10.34	8.38	11.09	14.24
Fe2O3	5.14	11.75	8.79	8.44	8.60	10.93	6.83	2.47
MnO	0.05	0.11	0.08	0.06	0.06	0.07	0.11	0.02
MgO	7.03	6.43	6.00	6.17	6.37	5.82	7.40	1.10
CaO	6.01	4.71	4.15	4.36	4.42	1.96	8.04	2.31
Na2O	1.24	0.69	0.98	0.92	0.96	0.19	1.12	2.65
K2O	6.75	5.40	6.19	5.96	6.12	4.88	6.68	3.95
P2O5	3.87	3.38	2.99	3.10	3.16	1.62	2.95	0.05
Total	96.48	99.21	99.53	96.69	99.48	98.82	98.06	97.70
(wt%)
F	0.12	0.05	0.10	0.08	0.08	0.05	0.09	0.04
Cl	0.13	0.37	0.17	0.18	0.20	0.25	0.14	0.05
(ppm)
Sc	6.85	15.70	3.28	4.38	8.74	15.18	23.86	1.40
V	106.42	137.25	138.17	120.72	131.79	245.36	90.03	47.72
Cr	212.81	231.37	200.03	204.07	207.65	244.94	207.38	6.99
Ni	208.24	143.59	145.19	178.74	178.28	68.21	156.20	15.44
Cu	u.d.	5.14	12.83	2.43	15.44	205.51	1.57	40.31
Zn	83.08	206.99	138.66	125.28	130.02	185.02	120.87	39.72
Rb	283.88	466.50	406.10	324.66	329.08	410.24	217.40	182.57
Sr	1939.66	321.02	579.66	631.64	623.80	35.77	1555.92	373.75
Y	38.33	75.67	48.35	52.88	55.01	134.89	50.67	14.44
Zr	76.81	590.00	463.75	344.09	368.83	1730.89	120.88	131.30
Nb	15.28	38.49	39.51	31.07	31.64	71.54	9.31	10.98
Ba	1290.33	1378.57	1761.47	1609.14	1695.32	2139.12	8002.86	1018.14
Pb	10.24	12.75	10.69	7.64	11.67	17.98	20.13	21.23
(ppm)
La	121.83	122.57	128.07	137.57		202.69	86.16	44.70
Ce	250.01	251.28	297.62	306.49		349.24	192.20	73.29
Pr			31.03	38.37
Nd	129.84	115.94	127.82	161.51		129.60	89.82	21.63
Sm	25.93	24.82	19.82	26.14		22.97	18.62	2.92
Eu	4.89	3.21	3.58	4.77		1.04	4.46	0.81
Gd	13.77	14.17	13.77	16.75		16.41	12.51	2.86
Tb	1.83		1.33	1.81				0.30
Dy	4.81	7.43	5.63	7.43		10.53	7.66	0.69
Ho			0.89	1.29				0.13
Er	1.64	3.67	2.05	2.95		5.47	3.15	0.29
Tm	0.23		0.28	0.38
Yb	1.04	3.40	1.31	1.86		4.47	2.34	0.31
Lu			0.23	0.28

Major and minor elements with F and Cl compositions were analyzed XRF system (Rigaku Primus IV), and lanthanoid compositions were analyzed by using ICP-AES system (Seiko-SPS 1200AR) mainly and ICP-MS system (Agilent 7900) for some samples (TM11012204A, 2501A, and 2506B from Rundvågshetta). Composition of pegmatite from Rundvågshetta (TM11012605C) is average of 10 analytical data (Supplementary Table S1). As a reference, the composition of ultrapotassic mafic rock from Innhovde (TM11020201A), which was collected during geological survey at JARE-52, was also analyzed and listed.

Figure 6. Comparison of major element compositions for investigated rocks. (a) Total alkali-silica (TAS) classification diagram. The chemical classification followed Le Maitre et al. (1989). (b)-(g) Variation diagrams for (b) TiO₂, (c) Al₂O₃, (d) Fe₂O₃, (e) MgO, (f) CaO, and (g) P₂O₅ versus SiO₂ (wt%) for investigated rocks. (h) CaO-Al₂O₃ discrimination diagram for ultrapotassic igneous rocks after Foley et al. (1987). The legend is common to all diagrams. Sample codes are attached partly to representative samples for recognition of each other (e.g., 12201A, TM11012201A; 12407A, TM11012407A; 12605E, TM11012605E).

Figure 7. Variation diagrams for (a) Zr, (b) Ni, (c) Rb, (d) Sr, (e) Nb, (f) Y, (g) Ba (ppm), and (h) Cl/F ratios versus SiO₂ (wt%) for investigated rocks. The legend is common to all diagrams. Sample codes are attached partly to representative samples for recognition of each other (e.g., 12201A, TM11012201A; 12407A, TM11012407A; 12605E, TM11012605E).

The composition of the rocks from Skallevikshalsen were almost the same, except for a few elements (e.g., SiO₂, MgO, P₂O₅, Ni, and Sr), probably because the rocks were derived from the same magma source with no effect of segregation and contamination. However, the major element oxide contents of the melanocratic dyke rocks from Rundvågshetta range widely from mafic to relatively felsic: 50-62 wt% SiO₂, 8-6 wt% MgO, 11-5 wt% total Fe as Fe₂O₃, 7.3-4.9 wt% K₂O (Table 1 and Fig. 6). The dyke possesses most of the characteristics ascribed to near-primitive mantle melts, with #Mg of up to 74 and high Cr and Ni contents (up to 300 and 200 ppm, respectively). In addition, TiO₂ (2.5-3.2 wt%) and P₂O₅ (1.6-4.2 wt%) contents tend to be higher and Al₂O₃ content is lower (11-8 wt%) than that of common igneous rocks. F and Cl contents and their ratios (Cl/F values) also differed among the dyke rock samples (Fig. 7h). In contrast, the pegmatite from Rundvågshetta (12605C) is felsic, and has a dacitic-to-rhyolitic composition (Fig. 6).

The contents of trace elements, specifically those of incompatible elements, in the melanocratic dyke rocks from Skallevikshalsen and Rundvågshetta were generally high (Table 1 and Fig. 7). Normalized incompatible trace element patterns generally display a gentle downward trend to the right, with enrichment in Ba and Rb more than 160 and 350 times that of the primitive mantle, respectively (Fig. 8a), and high values of large ion lithophile element-to-high field strength element ratios (LILE/HFSE). In the spidergram, troughs are evident for Nb, Ti, Sr, and Zr, whereas peaks of variable size are present for P (Fig. 8a). The melanocratic dyke rocks from Skallevikshalsen and Rundvågshetta have similar REE patterns, with downward trends to the right, and some samples show slightly negative Eu anomalies. The trace element concentration of the pegmatite from Rundvågshetta was lower than that of the dyke rocks. However, both the spidergram and REE pattern of the pegmatite show downward trends to the right. In the REE pattern, negative Eu anomaly are scarce for the pegmatite.

Figure 8. (a) Spidergram and (b) REE pattern for investigated rocks. Estimated C1 chondrite and primitive mantle abundances are from Sun and McDonough (1989).

Rb-Sr isotope compositions

The Rb and Sr contents and Sr isotopic compositions of the biotite and felsic minerals (mainly orthoclase) selected from sample 10406H collected at Skallevikshalsen and its whole-rock composition are shown in Table 2. Their ⁸⁷Rb/⁸⁶Sr and ⁸⁷Sr/⁸⁶Sr values are aligned on the isochron diagram (Fig. 9), and the age showed 490.4 ± 2.5 Ma (2σ) [IR = 0.706803 ± 0.000016 (2σ)] calculated from the slope of the alignment using error calculation method of IsoplotR (Vermeesch, 2018). In contrast, the Rb/Sr ratios and Sr isotopic compositions of biotite, felsic minerals (mainly orthoclase) and whole rock composition selected from 12302C collected at Rundvågshetta (Table 2) were also aligned on the isochron diagram (Fig. 9). The slope of the alignment showed 506.6 ± 0.8 Ma (2σ) [IR = 0.705103 ± 0.000013 (2σ)]. Both are defined from only three points; hence, the MSWD values are high (230 and 2500, respectively). However, the purity of the selected biotite is high, resulting in a high Rb/Sr value and small errors for estimated ages; hence, the results are meaningful in terms of age of intrusion in spite of high MSWD values.

Table 2. Analytical data for whole rock and mineral concentrates of melanocratic mafic dyke rocks TM11010406H and TM11012302C

	Rb (ppm)	Sr (ppm)	87Rb/86Sr	87Sr/86Sr
TM11010406H (Skallevikshalsen)
Whole Rock (WRS)	305.12	2262.09	0.3905	0.709366 ± 0.000009
Felsic Fractions (FFS)	170.40	5605.00	0.08801	0.707476 ± 0.000008
Biotite (BtS)	446.29	446.46	2.793	0.726305 ± 0.000009
TM11012302C (Rundvågshetta)
Whole Rock (WRR)	374.90	3561.48	0.3047	0.706860 ± 0.000009
Felsic Fractions (FFR)	153.28	10139.56	0.04375	0.705683 ± 0.000009
Biotite (BtR)	480.27	102.67	13.67	0.802629 ± 0.000009

Error given are ±2σ.

Figure 9. Rb-Sr isochron diagram for melanocratic dyke rocks from Skallevikshalsen (TM11010406H) and Rundvågshetta (TM11012302C). Letters identifying points refer to sample code in Table 2. For TM11010406H, a line with age of 490.4 ± 2.5 Ma with an initial (⁸⁷Sr/⁸⁶Sr) = 0.706803 ± 0.000016 is defined. For TM11012302C, a line with age of 506.6 ± 0.8 Ma with an initial (⁸⁷Sr/⁸⁶Sr) = 0.705103 ± 0.000013 is defined.

DISCUSSION

Intrusion timing of the melanocratic dyke

Major structures of metamorphic rocks, such as gneissosity, were formed during peak metamorphism at Rundvågshetta (Motoyoshi et al., 1986; Ishikawa et al., 1994). The melanocratic rock dykes are undeformed and clearly cut the main gneissosity found in the surrounding metamorphic rocks of Skallevikshalsen and Rundvågshetta. In addition, the melanocratic dykes cut across the cpx-amphibolite that intruded after the peak metamorphism (Fig. 2d). Therefore, the timing of dyke intrusion is considered to postdate the peak metamorphism at the Rundvågshetta. Similarly, the melanocratic dyke at Skallevikshalsen is thought to have intruded after the peak metamorphism. In contrast, some melanocratic dyke fragments were present in the pegmatite at Rundvågshetta (Fig. 2g). Therefore, the melanocratic dyke rocks may have been emplaced before pegmatite activity.

The melanocratic dyke rocks are holocrystalline, indicating that the dykes crystallized completely after intrusion into the metamorphic rocks. Biotite often grows parallel to the dyke margins. The dyke rocks are holocrystalline due to slow cooling after their intrusion.

The Rb-Sr isotope compositions of the melanocratic rocks showed 490.4 ± 2.5 Ma for 10406H from Skallevikshalsen, and 506.6 ± 0.8 Ma for 12302C from Rundvågshetta (Fig. 9); these ages limit the youngest timing of intrusion. Given that these are Rb-Sr mineral isochron ages and that the main associated mineral is biotite, these ages are thought to indicate a relatively late stage of slow cooling after the dyke intrusion (Spear, 1993). These dykes may have intruded relatively early after peak metamorphism. Therefore, the dykes may have experienced complete crystallization because of the high temperature conditions prevailing in the surrounding crust.

Correspondence between heterogeneity of apatite composition and variations of rock composition

The apatite in the melanocratic dyke rocks shows idiomorphic crystal forms under microscope. Compositionally they are fluorine-rich with a low Cl content was found in the melanocratic dyke rocks (Figs. 5a and 5b); however, relatively Cl-rich apatite with outer F-dominant portions was found in some dyke rocks from Rundvågshetta (e.g., 12407A and 12605E) (Figs. 5c and 5d). These F-rich portions appeared to have partly overgrown around the Cl-rich apatite. This indicates that the Cl-rich apatite crystallized before, and partially altered its composition in the F-rich environment. Such occurrences may be due to the assimilation of Cl-rich apatite-bearing materials (magma or rocks) by F-rich magma or local contamination of the materials by F-rich fluids.

The compositions of these two samples show high Cl/F ratios (7.0 and 4.8 for 12407A and 12605E, respectively) and a few other samples with high Cl/F ratios (>2) are observed, although almost of other rocks show low Cl/F ratios (e.g., 0.7 and 0.0 for 10406A and 12302C, in which only fluorapatite has been observed). The dyke rocks with high Cl/F values were relatively felsic in composition, whereas those with low Cl/F values were mafic (Table 1). Kawakami et al. (2016) reported the activity of high-salinity fluids at the LHC; however, the occurrence of dykes, especially those leaving a well-defined boundary and no reacting texture between dykes and surrounding host rocks, makes it difficult to imagine that any fluid alters the composition of the rocks from mafic to felsic locally. Considering the occurrence of the apatite and the compositional characteristics of the rocks, magma with a mafic composition mixed with magma with a relatively felsic composition or assimilated felsic rocks is plausible.

Possibility of mixing between mafic magma and felsic material

The composition of the dyke rocks varies from mafic to relatively felsic (SiO₂ = 45-49 wt% and MgO = 8.7-7.3 wt% for Skallevikshalsen, and SiO₂ = 50-62 wt% and MgO = 7.9-5.8 wt% for Rundvågshetta). In Figure 6, plots of major element oxides against SiO₂ indicate that the compositions of the melanocratic dyke rock samples are located on the apparent continuation of the trends between relatively mafic and felsic compositions. The decrease in the mafic components, Al, Ca, and P, with an increase in SiO₂ in the melanocratic dyke in Rundvågshetta (Fig. 6) may be explained by the contemporaneous fractionation of plagioclase, augite, and apatite. However, it is difficult to explain the inhomogeneity in apatite found in some dyke rocks (Figs. 5c and 5d) by the fractionation. The compositional variations in the melanocratic dyke rocks should be considered as a mixture of mafic magma and felsic materials.

To determine whether this compositional change was the result of mixing, we examined the possibility only with the compositions of the dyke rocks of Rundvågshetta, which is a representative suite of dyke rocks with larger sample populations and contains inhomogeneous apatite. Particularly, we consider how the composition of the mixture change between two endmembers, after setting the composition of 12201A with the highest MgO content (MgO = 7.9 wt%) as the mafic endmember and the composition of 12605E with the lowest MgO content (MgO = 5.8 wt%) as the less mafic endmember. Incidentally, 12201A is an SiO₂-poor, K₂O and F-rich rock, whereas 12605E is the most SiO₂-rich and F-poor rock (Table 2). When felsic magma such as 12605E is mixed with potassic mafic magma such as 12201A, the apatite in the felsic magma is possibly replaced by the surrounding F-rich domains. The magmatic mixing process was validated as a simple binary hyperbola, with major and trace element relationships between the mafic and felsic compositions plotted against Zr (Fig. 10). Almost all compositions of the dyke rocks from Rundvågshetta plot on or close to these hyperbolas. This implies that the composition of the dyke rocks can be explained simply by a mixture of two endmembers. The wide range of mixing proportions indicates that magma mixing, rather than crustal contamination, might have operated. This is because the assimilation of solid felsic rocks would result in lower melting potential and solidification of the entire system before such high proportions of the crust could be mixed in (Conticelli and Peccerillo, 1992; Peccerillo, 1992, 1995; Conticelli, 1998). If the surrounding rocks are assimilated, they are expected to be preserved as xenoliths, because xenoliths were observed in the dyke at Skallevikshalsen. Recent studies on melt inclusions also suggest the presence of A-type granitic magma during UHT metamorphism (Carvalho et al., 2023).

Figure 10. Selected major element oxide and trace element ratios versus Zr as an index of magma mixing. The curve in the figures show the variation of composition by mixing between assumed two compositional endmembers (the composition of TM11012201A and the composition of TM11012605E). The legend is common to all diagrams. The attached numbers indicate the ratio (%) of influence of compositions of TM11012201A in the mixture. The selection of elements was based on Prelević et al. (2004). The inset in the SiO₂/Zr versus Zr (ppm) diagram shows a linear trend. This relation is used to plot the hyperbola in the SiO₂/Zr versus Zr (ppm) diagram. Sample codes are attached partly to representative samples for recognition of each other (e.g., 12201A, TM11012201A; 12407A, TM11012407A; 12605E, TM11012605E).

In contrast, the compositions of the pegmatite from Rundvågshetta, which is more felsic, often deviated from the mixing curve (Fig. 10). Moreover, because pegmatite would have been emplaced after the melanocratic dyke rock activity, pegmatites are unlikely endmembers or products of mixing. Felsic endmembers should still be considered as the origin of dyke rocks during the mixing process.

Origin of ultrapotassic mafic magma and its physical properties

The mafic endmembers of the melanocratic dyke rocks from Rundvågshetta have ultrapotassic compositions. Ultrapotassic mafic rocks with equivalent compositions have also been found at Skallevikshalsen, as described above, and reported at Innhovde (Arima and Shiraishi, 1993). In addition, they are characterized by abundant trace elements, both incompatible elements and also compatible elements such as Ni and Cr (Table 1 and Fig. 7). This corresponds to the compositional characteristics of mantle-derived melts such as the lamproite or minette (Jaques et al., 1984, 1986; Foley et al., 1987; Mitchell and Bergman, 1991). They are generally associated with intra-continental tectonic settings, or post-orogenic collapses, post-dating convergent tectonics, and active margin processes (Mitchell and Bergman, 1991). In the latter setting, ultrapotassic mafic rocks may be associated with calc-alkaline felsic magma; therefore, the mixing of melts with contrasting compositions might be expected in this environment. Furthermore, Prelević et al. (2004) mentioned that lamproite melts may be especially reactive and hybridized with felsic magmas with high normative quartz contents owing to their high alkali contents and considerable under-saturation in Al₂O₃. Melanocratic mafic rocks from the LHC have high F content (Table 1), which is expected to reduce the viscosity and density of the melt, act as a volatile component, and produce a reactive potential (Burnham, 1979; Foley et al., 1986a, 1986b). Therefore, ultrapotassic mafic magma, as an endmember of the melanocratic dyke, has a high hybridization potential and is considered to have easily mixed with relatively felsic magma to form a suite of dykes at Rundvågshetta.

The LHC is interpreted as a complex that exists in the suture zones of East and West Gondwana (Fig. S1), and its active period is 600-520 Ma (e.g., Dunkley et al., 2020, and reference therein). Owada et al. (2013) found ultrapotassic mafic rocks as unmetamorphosed minettes intrusions from the igneous activity at Sør Rondane Mountains in East Antarctica (Fig. S1) and concluded that they were products that postdating the suturing event in the Pan-African suture zone between East and West Gondwana. They concluded that the minette magma was derived from an enriched mantle source (Owada et al., 2013). Ultrapotassic mafic rocks in the LHC may also have been generated as part of the igneous activity associated with the Pan-African Orogeny related to the East and West Gondwana collisions (Arima and Shiraishi, 1993).

Origin of felsic magma as another endmember of the melanocratic dyke rocks

We must consider the origin of another endmember with a relatively felsic composition of melanocratic dyke rocks in Rundvågshetta. In the mixing calculation mentioned above, the composition of 12605E was set as the end member, and although the rock showed a relatively high SiO₂ content equivalent to andesite to dacite (SiO₂ = 62 wt%) and was characteristically poor in Al (Al₂O₃ = 8.38 wt%). The rock is moderately mafic (MgO = 5.82 wt%, Fe₂O₃ = 10.93 wt%), relatively low in CaO (CaO = 1.96 wt%) and ultrapotassic (K₂O = 4.88 wt%, K₂O/Na₂O > 25) composition (Table 1), with a character of high Zr content and relatively low Ni and Sr contents in trace elements (Figs. 7 and 8a), and a strong Eu anomaly in the REE pattern (Fig. 8b). Considering the occurrence of apatite, 12605E rock may have already diluted its features from the felsic endmember by mixing with ultrapotassic mafic magma corresponding to a lamproitic to minettic composition. Thus, felsic magma with Al-, Mg-, Fe-, and Ca-poor compositions is regarded as another endmember of the mixture. Composition of pegmatite (12605C) show a high Al content (average Al₂O₃ = 14.24 wt%), and extremely low Ti and P contents (average TiO₂ = 0.59 wt%, and P₂O₅ = 0.05 wt%), and is outside the mixing curve (Fig. 10); hence, it is not a candidate for the endmember. The existence of such Al-poor magma is difficult to imagine in general igneous rocks; however, low-Ca boninite (Yajima and Fujimaki, 2001) may be a possible candidate for mixing. If magma was generated from enriched mantle rather than depleted mantle under the environment of generation of the low-Ca boninite seen in Ogasawara, which was thought to have formed from depleted mantle (Yajima and Fujimaki, 2001), it might have become rich in Mg, Fe, and trace elements while maintaining low Al. Boninite may have become Zr-enriched during its formation; this feature is common in the composition of 12605E. If plagioclase is crystallized and fractionated from the product, Al₂O₃ and CaO will be further reduced, SiO₂ will increase in the melt, and the characteristics of the major elements may become similar to those of 12605E. Alternatively, fractionation of Al-bearing minerals such as anorthite and the high-pressure phase of pyroxene from such boninitic magma may have modified its composition to felsic and Al-poor, although it is difficult to assume that the composition of magma becomes Al-poor only by the fractionation of usual mafic minerals. Thus, the partial melting of the enriched mantle related to the igneous activity associated with the East and West Gondwana collisions might have produced relatively felsic magmas corresponding to this endmember because of differences in volatile components and the degree of melting, although the generated magma might have a more mafic composition than expected.

A more plausible scenario is that the endmember may have been the product of the partial melting of crustal material associated with UHT metamorphism. Partial melting during the UHT metamorphism to form felsic melt has been reported (Hiroi et al., 2019, 2023): it has been suggested that the composition of the melt produced at that time is A-type granite-like with weakly peraluminous to weakly peralkaline affinity, ferroan character, high alkali contents, high K/Na and Ga/Al, and low Ca, Ba, Sr, and H₂O concentrations (Carvalho et al., 2023). When biotite melts at high temperatures, it crystallizes orthopyroxene to produce melt, and the higher the Ti and F contents of biotite, the longer the stable condition of the biotite extends to higher temperatures (Tareen et al., 1995). Furthermore, when biotite enriched with Ti and F coexists with sillimanite, specifically when the starting material is like pelitic rocks, melting results in cordierite crystallization, and at a higher temperature (about 1100 °C), sapphirine is produced together with the melt. In addition, when H₂O-undersaturated, the melting causes crystallization of sapphirine and orthopyroxene coexisting with fluorobiotite and incongruently forms a melt (Tareen et al., 1998). In their synthetic experiments, the melts produced were rich in K₂O and SiO₂ and relatively poor in Al₂O₃ compared with the starting material. Furthermore, Rundvågshetta has experienced UHT metamorphism, and sapphirine coexist with orthopyroxene in pelitic metamorphic rocks (Motoyoshi et al., 1986; Yoshimura et al., 2008). During the incongruent melting, F tends to remain in the solid phase such as phlogopite (Tareen et al., 1995, 1998; Markl and Piazolo, 1998; Motoyoshi and Hensen, 2001). F content in biotite has been reported high at Rundvågshetta (Yoshimura et al., 2008); hence, a small amount of F in the melt is expected. In addition, Cl is likely to be released from the solid phase during the partial melting in high-grade metamorphism (Markl and Piazolo, 1998; Kawakami et al., 2016); therefore, the generated melt is expected to be enriched with Cl components. The sample 12605E also had a higher Cl/F ratio than the other samples, and the apatite core had a high Cl/F value and contained almost no OH components (Fig. 5d). When the melt formed by such incongruent melting is the endmember and feldspar remains in the original rock as its restitic component during melting, the strong Eu anomaly in the REE pattern and the low Ca and Sr contents can be explained. In addition, the sampling point of 12605E was closest to the distribution of Grt-Sil and Grt-Bt gneisses in Rundvågshetta (Fig. 4), which experienced partial melting during UHT metamorphism (Yoshimura et al., 2008; Hiroi et al., 2019); therefore, the chemical composition of 12605E might have become the most felsic in the dyke. Finally, the generated melt produced by partial melting accompanying the UHT metamorphism not only formed leucosomes in the metamorphic rocks but also formed dykes by mixing with the ultrapotassic mafic magma supplied from the mantle during the collision of the East and West Gondwana.

Either the enriched mantle may have partially melted or the melt may have been generated by the partial melting of pelitic rocks during UHT metamorphism. In any case, collisions between eastern and western Gondwana seem to have been involved in the formation of the felsic magma. Ultrapotassic mafic magma has also been reported to have occurred at the Sør Rondane Mountains, eastern Dronning Maud Land, East Antarctica due to collisions between East and West Gondwana (Owada et al., 2013). However, this study aids in understanding that collisions between East and West Gondwana also generated ultrapotassic mafic magma at the LHC. The generation of ultrapotassic mafic magma may be universal in crustal collision situations over time (Canning et al., 1996; Prelević et al., 2004). As discussed above, the genesis of igneous rocks with compositional characteristics different from those of common igneous rocks should be investigated by considering their occurrences, chemical compositions, and properties. In this study, we clarified the relationship between mafic igneous rocks and the collision between the East and West Gondwanan continents based on their common compositional features and timing. The genesis of felsic igneous rocks remains unclear; however, the finding of ultrapotassic dykes with Al-poor felsic compositions in Rundvågshetta are likely genetically related to UHT metamorphism with partial melting at the LHC. To investigate this possibility and consider the relationship with metamorphism in more detail, it is necessary to consider differences in the degree of metamorphism and compare a wide range of igneous rock compositions in the future.

CONCLUSIONS

Melanocratic rock dykes were found at Skallevikshalsen and Rundvågshetta in the LHC during JARE-52. The dyke rocks at Skallevikshalsen had an ultrapotassic mafic composition enriched with minor elements. They resemble the compositional features of the lamproitic to minettic dyke rock previously found at Innhovde, western LHC. The dyke rocks from the Rundvågshetta dykes are a mixture of ultrapotassic mafic magma with lamproitic to minettic compositions and intermediate to felsic magma with high Cl/F and low Al₂O₃ composition. Considering their occurrence and the results of Rb-Sr mineral dating, the time of the intrusion was immediately after the peak metamorphism of the LHC, and the cause of the igneous activity was thought to be the collision between East and West Gondwana.

ACKNOWLEDGMENTS

We express our sincere thanks to the members of JARE-52 and crew of the icebreaker ‘Shirase’. We would like to thank the member of West Japan East Antarctic Research Group, especially Prof. H. Ishizuka, T. Kawasaki, Y. Osanai, M. Owada, T. Shimura, A. Kamei, and Drs. N. Nakano and T. Adachi, for their constructive discussions. In them, we acknowledge the useful comments and advise from Prof. M. Owada during data analysis. We thank Prof. M. Satish-Kumar for editing and two anonymous reviewers, who gave us useful advices to revise the text. We would like to thank Editage (www.editage.com) for English language editing. We also received help from Ms. K. Shiraga in correcting the text. Finally, we would also like to thank Prof. T. Hokada, main proposer of this special issue, for giving us the opportunity to publish this thesis. TM’s research was supported by MEXT KAKENHI Grant Number JP19540484. D.Dunkley’s research was supported financially by an OPUS grant UMO2021/43/B/ST10/03161 from the National Science Centre, Poland.

SUPPLEMENTARY MATERIALS

Supplementary Figures S1-S2 and Tables S1-S2 are available online from https://doi.org/10.2465/jmps.221201.

REFERENCES

• Arima, M. and Shiraishi, K. (1993) Geochemical characteristics of metamorphosed high K/Na dykes in eastern Queen Maud Land, Antarctica: ultrapotassic igneous activity linked to Pan African orogeny. Proceedings of NIPR Symposium on Antarctica Geosciences, 6, 103-115.
•  Baba,  S.,  Nantasin,  P.,  Kamei,  A.,  Kitano,  I., et al (2023) Counter-clockwise P-T history deduced from kyanite-bearing pelitic gneiss in Tenmondai Rock, Lützow-Holm Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 118, S001.
• Burnham, C.W. (1979) The importance of volatile constituents. In The Evolution of the Igneous Rocks; Fiftieth Anniversary Perspectives (Yoder, H.S., Jr Ed.). Princeton University Press, Princeton, NJ, 439-482.
•  Canning,  J.C.,  Henney,  P.J.,  Morrison,  M.A. and  Gaskarth,  J.W. (1996) Geochemistry of late Caledonian minettes from northern Britain; implications for the Caledonian sub-continental lithospheric mantle. Mineralogical Magazine, 60, 221-236.
•  Carvalho,  B.B.,  Bartoli,  O.,  Cesare,  B.,  Satish-Kumar,  M., et al (2023) Revealing the link between A-type granites and hottest melts from residual metasedimentary crust. Geology, 51, 845-849.
•  Conticelli,  S. (1998) The effect of crustal contamination on ultrapotassic magmas with lamproitic affinity: mineralogical, geochemical and isotope data from the Torre Alfina lavas and xenoliths, Central Italy. Chemical Geology, 149, 51-81.
•  Conticelli,  S. and  Peccerillo,  A. (1992) Petrology and geochemistry of potassic and ultrapotassic volcanism in central Italy: petrogenesis and inferences on the evolution of the mantle sources. Lithos, 28, 221-240.
•  Dunkley,  D.J.,  Hokada,  T.,  Shiraishi,  K.,  Hiroi,  Y., et al (2020) Geological subdivision of the Lützow-Holm Complex in East Antarctica: From the Neoarchean to the Neoproterozoic. Polar Science, 26, 100605.
•  Foley,  S.F.,  Taylor,  W.R. and  Green,  D.H. (1986a) The role of fluorine and oxygen fugacity in the genesis of the ultrapotassic rocks. Contributions to Mineralogy and Petrology, 94, 183-192.
• Foley, S.F., Taylor, W.R. and Green, D.H. (1986b) The structural role of fluorine in magmatic systems with application to the genesis of lamproites. Proceedings of the Eighth Australian Geological Convention; Geological Society of Australia; Earth Resources in Time and Space. Geological Society of Australia, Sydney, N.S.W., 212-213.
•  Foley,  S.F.,  Ventturelli,  G.,  Green,  D.H. and  Toscani,  L. (1987) The ultrapotassic rocks: Characteristics, classification and constraints for petrogenetic models. Earth Science Reviews, 24, 81-134.
• Hiroi, Y., Shraishi, K. and Sasaki, K. (1986) Explanatory text of geological map of Akebono Rock, Antarctica. Antarctic Geological Map Series, Sheet 16 (with explanatory text 6p), NIPR, Tokyo.
• Hiroi, Y., Shiraishi, K. and Motoyoshi, Y. (1991) Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In Geological Evolution of Antarctica (Thomson, M.R.A., Crame, J.A. and Thomson, J.W. Eds.). Cambridge University Press, Cambridge, 83-87.
•  Hiroi,  Y.,  Hokada,  T.,  Kato,  M.,  Yanagi,  A., et al (2019) Felsite-nanogranite inclusions and three Al₂SiO₅ polymorphs in the same garnet in ultrahigh-temperature granulites from Rundvågshetta, Lützow-Holm Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 114, 60-78.
•  Hiroi,  Y.,  Hokada,  H.,  Adachi,  T.,  Shiraishi,  K., et al (2023) Nanogranitoid inclusions with grandidierite in mafic granulite from Austhovde, Lützow-Holm Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 118, S002.
•  Hokada,  T. and  Motoyoshi,  Y. (2006) Electron microprobe technique for U-Th-Pb and REE chemistry of monazite, and its implications for pre, peak- and post metamorphic events of the Lützow-Holm Complex, East Antarctica. Polar Geosciences, 19, 118-151.
• Ishikawa, M., Motoyoshi, Y., Fraser, G.L. and Kawasaki, T. (1994) Structural evolution of Rundvågshetta region, Lützow-Holm Bay, East Antarctica. Proceedings of the NIPR Symposium on Antarctic Geosciences, 7, 69-89.
•  Jaques,  A.L.,  Webb,  A.W.,  Fanning,  C.M.,  Black,  L.P., et al (1984) The age of the diamond-bearing pipes and associated leucite lamproites of the West Kimberley region, Western Australia. BMR Journal of Australian Geology and Geophysics, 9, 1-7.
• Jaques, A.L., Boxer, G.L., Lucas, H. and Haggerty, S.E. (1986) Mineralogy and petrology of the Argyle lamproite pipe, Western Australia. Proceedings of the Fourth International Kimberlite Conference. Geological Society of Australia, Sydney, N.S.W., 48-50.
•  Kawakami,  T. and  Ikeda,  T. (2004) Timing of ductile deformation and peak metamorphism in Skallevikshalsen, Lützow-Holm Complex, East Antarctica. Polar Geoscience, 17, 1-11.
•  Kawakami,  T. and  Motoyoshi,  Y. (2004) Timing of attainment of the spinel + quartz coexistence in garnet-sillimanite leucogneiss from Skallevikshalsen, Lützow-Holm Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 99, 311-319.
•  Kawakami,  T.,  Hokada,  T.,  Sakata,  S. and  Hirata,  T. (2016) Possible polymetamorphism and brine infiltration recorded in the garnet-sillimanite gneiss, Skallevikshalsen, Lützow-Holm Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 111, 129-143.
• Kitano, I., Hokada, T., Baba, S., Kamei, A., et al. (2021) U-Pb zircon geochronology of high-grade metamorphic rocks from outcrops along the Prince Olav Coast, East Antarctica. Abstracts with Programs of the 12th Symposium on Polar Science. National Institute of Polar Research.
•  Kitano,  I.,  Hokada,  T.,  Baba,  S.,  Kamei,  A., et al (2023) Zircon geochronology of high-grade metamorphic rocks from outcrops along the Prince Olav Coast, East Antarctica: Implications for multi-thermal events and regional correlations. Journal of Mineralogical and Petrological Sciences, 118, S009.
• Le Maitre, R.W., Bateman, P., Dudek, A., Keller, J., et al. (1989) A classification of igneous rocks and glossary of terms. Blackwell, Oxford.
•  Markl,  G. and  Piazolo,  S. (1998) Halogen-bearing minerals in syenites and high-grade marbles of Dronning Maud Land, Antarctica: monitors of fluid compositional changes during late-magmatic fluid-rock interaction processes. Contribution to Mineralogy and Petrology, 132, 246-268.
• Mitchell, R.H. and Bergman, S.C. (1991) Petrology of Lamproites. New York, Plenum Press, 447.
•  Miyamoto,  T.,  Yoshimura,  Y.,  Sato,  K.,  Motoyoshi,  Y., et al (2004) Occurrences of metamorphosed ultramafic rock and associating rocks in Howard Hills, Enderby Land, East Antarctica: An evidence of partial melting on their geochemical and isotopic characters. Polar Geosciences, 17, 88-111.
•  Mizuochi,  H.,  Satish-Kumar,  M.,  Motoyoshi,  Y. and  Michibayashi,  K. (2010) Exsolution of dolomite and application of calcite-dolomite solvus geothermometry in high-grade marbles: An example from Skallevikshalsen, East Antarctica. Journal of Metamorphic Geology, 28, 509-526.
•  Mori,  Y.,  Hokada,  T.,  Miyamoto,  T. and  Ikeda,  T. (2023) Metamorphic age and pressure-temperature conditions recorded in a sillimanite-garnet-bearing pelitic gneiss from Niban-nishi Rock of Niban Rock on the Prince Olav Coast, eastern Dronning Maud Land, East Antarctica: Evidence for Tonian metamorphism. Journal of Mineralogical and Petrological Sciences, 118, S005.
• Motoyoshi, Y., Matsueda, H., Matsubara, S., Sasaki, K. and Moriwaki, K. (1986) Geological map of Rundvågskollane and Rundvågshetta, Antarctica. Antarctic Geological Map Series, Sheet 24 (with explanatory text 25p), Natl. Inst. Polar Res., Tokyo.
• Motoyoshi, Y., Matsubara, S. and Matsueda, H. (1989) P-T evolution of the granulite-facies rocks of the Lützow-Holm Bay region, East Antarctica. In Evolution of Metamorphic Belts (Daly, J.S., Cliff, R.A. and Yardley, B.W.D. Eds.). Geol. Soc. London, Spec. Pub., 43, 325-329.
•  Motoyoshi,  Y. and  Hensen,  B.J. (2001) F-rich phlogopite stability in ultra-high-temperature metapelites from the Napier Complex, East Antarctica. American Mineralogist, 86, 1404-1413.
• Motoyoshi, Y., Hiroi, Y., Satish-Kumar, M. and Hokada, T. (2006) Timing of UHT metamorphic event at Rundvågshetta, Lützow-Holm Complex, East Antarctica. Abstracts of the 113th annual meeting of the Geological Society of Japan, 139 (O-193).
•  Nakada,  S. (1985) X-ray fluorescence analysis of trace elements in silicate rocks using fused disk samples. Science Reports, Department of Geology, Kyushu University, 14, 117-127 (in Japanese with English abstract).
•  Nakada,  S. (1987) X-ray fluorescence determination of trace elements in silicate rocks (part 2). Science Reports, Department of Geology, Kyushu University, 15, 37-44 (in Japanese with English abstract).
•  Nakada,  S.,  Yanagi,  T.,  Maeda,  S.,  Fang,  D. and  Yamaguchi,  M. (1985) X-ray fluorescence analysis of major elements in silicate rocks. Science Reports, Department of Geology, Kyushu University, 14, 103-115 (in Japanese with English abstract).
• Osanai, Y., Toyoshima, T., Owada, M., Tsunogae, et al. (2004) Geological map of Skallen (revised edition), Antarctica. Antarctic Geological Map Series, Sheet 39 (with explanatory text 31p), Natl. Inst. Polar Res., Tokyo.
•  Owada,  M.,  Kamei,  A.,  Horie,  K.,  Shimura,  T., et al (2013) Magmatic history and evolution of continental lithosphere of the Sør Rondane Mountains, eastern Dronning Maud Land, East Antarctica. Precambrian Research, 234, 63-84.
•  Peccerillo,  A. (1992) Potassic and ultrapotassic rocks: compositional characteristics, petrogenesis and geologic significance. Episodes- Journal of International Geoscience, 15, 243-251.
• Peccerillo, A. (1995) Mafic calcalkaline to ultrapotassic magmas in central-southern Italy; constraints on evolutionary processes and implications for source composition and conditions of magma generation. In Proceedings of the Symposium on the Physics and the Chemistry of the Upper Mantle (Cordani, U.G. and Svisero, D.P. Eds.). Academia Brasileira de Ciencias, Rio de Janeiro, 171-189.
•  Prelević,  D.,  Foley,  S.F.,  Cvetković,  V. and  Romer,  R.L. (2004) Origin of minette by mixing of lamproite and dacite magmas in Veliki Majdan, Serbia. Journal of Petrology, 45, 759-792.
• Shiraishi, K. and Yoshida, M. (1987) Geological map of Botnneset, Antarctica. Antarctic Geological Map Series, Sheet 25 (with explanatory text 18p), Natl. Inst. Polar Res., Tokyo.
•  Shiraishi,  K.,  Ellis,  D.J.,  Hiroi,  Y.,  Fanning,  C.N., et al (1994) Cambrian orogenic belt in East Antarctica and Sri Lanka: Implications for Gondwana assembly. Journal of Geology, 102, 47-65.
• Shiraishi, K., Dunkley, D.J., Hokada, T., Fanning, C.M., et al. (2008) Geochronological constraints on the Late Proterozoic to Cambrian crustal evolution of eastern Dronning Maud Land, East Antarctica: a synthesis of SHRIMP U-Pb age and Nd model age data. In Geodynamic evolution of East Antarctica: A key to the East-West Gondwana Connection (Satish-Kumar, M., Motoyoshi, Y., Osanai, Y., Hiroi, Y. and Shiraishi, K. Eds.). Geological Society, London, Special Publications, SP308, 21-67.
• Spear, F.S. (1993) Metamorphic phase equilibria and pressure-temperature-time paths. Mineral Soc. Amer., Washington D.C., 799.
•  Steiger,  R.H. and  Jäger,  E. (1977) Subcommission on geochronology: convention on the use of decay constants in geo- and cosmo-chronology. Earth and Planetary Science Letters, 36, 359-362.
• Sun, S.-S. and McDonough, W.F. (1989) Chemical and isotope systematics of oceanic basalts: Implication for mantle composition and processes. In Magmatism in the Ocean Basins (Saunders, A.D. and Norry, M.J. Ed.). Blackwell, Oxford, 313-345.
•  Tareen,  J.A.K.,  Keshava Prasad,  A.V.,  Basavalingu,  B. and  Ganesha,  A.V. (1995) The effect of fluorine and titanium on the vapour-absent melting of phlogopite and quartz. Mineralogical Magazine, 59, 566-570.
•  Tareen,  J.A.K.,  Keshava Prasad,  A.V.,  Basavalingu,  B. and  Ganesha,  A.V. (1998) Stability of F-Ti-phlogopite in the system phlogopite-sillimanite-quartz: an experimental study of dehydration melting in H₂O-saturated and undersaturated conditions. Mineralogical Magazine, 62, 373-380.
•  Vermeesch,  P. (2018) IsoplotR: a free and open toolbox for geochronology. Geoscience Frontiers, 9, 1479-1493.
•  Yajima,  K. and  Fujimaki,  H. (2001) High-Ca and low-Ca boninites from Chichijima, Bonin (Ogasawara) archipelago. Japanese Magazine of Mineralogical and Petrological Sciences, 30, 217-236 (in Japanese with English abstract).
• Yanai, K. and Ishikawa, T. (1978) Geological map of Cape Hinode, Antarctica. Antarctic Geological Map Series, Sheet 11 (with explanatory text 10p), Natl. Inst. Polar Res., Tokyo.
• Yoshida, M., Yoshida, Y., Ando, H., Ishikawa, T. and Tatsumi, T. (1976) Geological map of Skallen, Antarctica. Antarctic Geological Map Series, Sheet 9 (with explanatory text 16p), Natl. Inst. Polar Res., Tokyo.
•  Yoshimura,  Y.,  Motoyoshi,  Y.,  Miyamoto,  T.,  Grew,  E.S., et al (2004) High-grade metamorphic rocks from Skallevikshalsen in the Lützow-Holm Complex, East Antarctica: Metamorphic conditions and possibility of partial melting. Polar Geoscience, 17, 57-87.
• Yoshimura, Y., Motoyoshi, Y. and Miyamoto, T. (2008) Sapphirine + quartz association in garnet: implication for ultrahigh-temperature metamorphism at Rundvågshetta, Lützow-Holm Complex, East Antarctica. In Geodynamic Evolution of East Antarctica: A Key to the East-West Gondwana Connection (Satish-Kumar, M., Motoyoshi, Y., Osanai, Y., Hiroi, Y. and Shiraishi, K. Eds.). Geological Society, London, Special Publications, SP308, 377-390.