1. Introduction

Nepheline syenites and their volcanic equivalents are part of the alkaline silica-undersaturated rock suites which are often associated with carbonatites (e.g., Keller, 1984a, b; Worley et al., 1995; Woolley, 2001; Halama et al., 2005). Melts which are parental to syenites and carbonatites are usually generated by low degrees of partial melting of lithospheric mantle (e.g., Bell and Simonetti, 2010; Yaxley et al., 2022). The formation of carbonatites and their spatial proximity to silicate magmatites is explained by different models.

There is still an open discussion whether silicate and carbonatitic melts are cogenetic. Classical models on the common occurrence of both rock types are: I. The formation of two independent primary melts of a highly oxidized carbonate-metasomatized lithospheric mantle (e.g., Wallace and Green, 1988; Ying et al., 2004; Braunger et al., 2018). II. A primary melt from a metasomatized lithospheric mantle that fractionates and further decouples a carbonatitic melt by liquid immiscibility (e.g., Gittins, 1989; Halama et al., 2005; Brooker and Kjasgaard, 2011). Gittins and Mitchell (2023), however, state that, as yet no reliable experimental work is available that proves that nephelinitic melts evolve to such an extent that the separation of immiscible silicate and carbonate melts can take place. Furthermore, the different rheologies of both melts favor different intrusion levels. III. The possible formation of basic secondary syenites and ijolites through the metasomatic influence of carbonatitic melts on solid and partly acidic rocks (Vasyukova and Williams-Jones, 2022). In addition, Anenburg and Walters (2024) describe silica contamination from silica-oversaturated wall rocks by carbonatite melts behind the formation of silicocarbonatite-like assemblages. IV. Small carbonate accumulations in nepheline-bearing lava caused by melted crustal limestones, i.e., anatectic carbonatites (Gozzi et al., 2014).

Alkaline rock suites sometimes display both silica oversaturated and silica undersaturated trends (Bonin and Giret, 1984), which makes it possible for nepheline syenites to occur in close relationship with silica-saturated granitoids. The majority of the world’s rare earth elements (REEs) and Nb resources are concentrated in carbonatites and alkaline-silicate rocks (Chakhmouradian and Williams, 2004; Atanasova et al., 2017; Anenburg et al., 2021; Beard et al., 2023). Therefore, syenites and carbonatites have increasingly come into the focus of exploration, either for alumina and alkalis for ceramic glass, enamels, fillers, paints, papers, plastics, and foam rubber or critical raw materials such as REEs, Nb, Ta, P, Zr and F (e.g., Chakhmouradian and Wall, 2012; Jones et al., 2013; Marks and Markl, 2017; Stoppa et al., 2019; Anenburg et al., 2021; Beard et al., 2023). More recently, selected nepheline syenites have been studied because of their potential as alternative potash sources in fertilizers (e.g., Chiwona et al., 2020). Some of the world’s first-class REEs and Nb deposits and resources are related to these rock associations (e.g., Anenburg et al., 2021; Beard et al., 2023). Prominent examples are the nepheline-syenites of Norra Kärr, Sweden, the largest European deposit of heavy REEs (Atanasova et al., 2017), and the Nb-Ta-rich, pyrochlore-bearing nepheline-syenites from Motzfeld, Greenland (Finch et al., 2019). An example of co-existence of syenites and carbonatites is Bayan Obo, China, the world’s largest production facility for light REEs (Yang et al., 2019).

The present study explores the use of minor   (1-5 vol.%) and trace (<1 vol.%) minerals as tracers to investigate the possible existence of a buried or eroded carbonatite in the vicinity of the Proterozoic San José del Guaviare Syenite (SJGS) in southeastern Colombia. Secondly, the study also focuses on assessing the abundance of critical elements such as Nb and REEs and their mineralogy in the SJGS. Although the SJGS has so far been reported as isolated outcrops and studied petrographically, geochronologically, and geochemically (whole rock) (Arango et al., 2011; Franco-Victoria et al., 2018; Larrota Rincón and Concha, 2018; Amaya López et al., 2021), there are no in-depth studies on mineral chemistry yet.

2. Geological setting

2.1. The Amazonian Craton and its surroundings

The Amazonian Craton represents the largest and most diverse example of Precambrian continental crustal growth within the South American continent (Engler, 2009). It is subdivided into numerous Neoarchean to Proterozoic domains (Fig. 1) with ages ranging from >2.6 Ga to ca. 0.95 Ga (Cordani and Teixeira, 2007). Texeira et al. (2019) described the craton as a sequence of accreted continental slices of Paleo- to Mesoproterozoic continental and oceanic arcs with one or two Archean cores in the present northeast. These crustal domains were partially affected by the Brasiliano/Pan-African orogeny (peak ca. 0.61 Ga; Rodriguez-Corcho et al., 2021).

Fig. 1. Simplified geotectonic framework of South America highlighting the Amazonian Craton and the Guiana Shield. Modified from Ibáñez-Mejía et al. (2020).

According to Cordani and Teixeira (2007) and Engler (2009), the Amazonian craton is composed of provinces with different ages, starting with Central Amazonian (>2.6 Ga) and the Maroni-Itacaiúnas (2.26-1.95 Ga) domains at the east with four contiguous provinces to the west linked by orogenic events (Fig. 1). In Colombia, Ibáñez-Mejía et al. (2011) described the Putumayo Orogen, a Meso- to Neoproterozoic belt buried under the north Andean foreland basins at the suggested border between the Amazonian Craton and the Guiana Shield. Galvis et al. (1979) described Precambrian igneous and metamorphic rocks of the Mitú Migmatitic Complex, which builds up the crystalline basement of eastern Colombia, as the northwestern part of the Guiana Shield. Priem et al. (1982) reported a Rb-Sr age of 1.78 Ga for biotitic gneisses, suggesting a relationship to the Rio Negro-Juruena Province (1.82-1.60 Ga). Ibáñez-Mejía et al. (2020), however, indicated a discontinuous distribution of U-Pb crystallization ages of these rocks with four main age clusters: 1.99, 1.81-1.72, 1.59-1.50, and 1.41-1.39 Ga, therefore suggesting that the term “Mitú Migmatitic Complex” would now be inadequate and obsolete.

2.2. San José del Guaviare Syenite Unit (SJGS)

Galvis et al. (1979) were the first to describe silica-undersaturated rocks in the SJGS area of southeastern Colombia, found in the La Lindosa mountain range and the isolated hills of Cerritos and El Capricho (Fig. 2). The unit is composed of diverse syenites, like nepheline syenites, nepheline-monzosyenites, and nepheline-bearing alkali feldspar syenites, as well as quartz-bearing syenites, syenogranites, and quartz-rich granitoids (Amaya López et al.,2021). Amaya López et al. (2021) stated that silica-saturated intrusions would be underrepresented in terms of volume and restricted to the southwestern hills (El Capricho and Cerritos; Fig. 2).

Fig. 2. Geological map of the SJGS area, after Amaya López et al. (2021). Blue circles refer to sample localities.

U-Pb zircon ages reported by Arango et al. (2011), Franco-Victoria et al. (2018), and Amaya López et al. (2021) indicate an Ediacaran age (~620-605 Ma) for the SJGS. The primary melts correspond to intraplate anorogenic alkaline magmas intruded into rift-like stretching zones during the fragmentation of Rodinia (Amaya López et al., 2021), when the Iapetus Ocean opened and the northern Amazonian Craton separated from Laurentia and Baltica (Condie, 2011; Cawood and Pisarevsky, 2017).

The isotope signature of the products indicates a mantle origin and silica-saturated melts, resulting in syenogranites and quartz-syenites with evidence of crustal assimilation (Amaya López et al.,2021). The intrusions are hosted in the Mesoproterozoic metamorphic basement of the Guaviare Complex described by Maya et al. (2018), which is locally affected by contact metamorphism producing hornfels aureoles, well exposed in the southern Cerritos hill (Fig. 2). Syenites of this unit are overlain unconformably by the Cretaceous sedimentary sequence of the San José Formation in the La Lindosa mountain range and by Neogene mudstones of the La Caja Formation (Arango et al., 2011).

3. Methods

Eight thin sections from El Capricho and two from Las Delicias were studied with a petrographic microscope. Of these, seven were examined by electron probe microanalyzer (EPMA) and scanning electron microscopy (SEM) with energy-dispersive spectroscopy (EDS).

Coated thin sections were analyzed using a JEOL JXA-8200 EPMA. Natural and synthetic silicates, oxides, and phosphates were used as reference materials. For the major elements, natural reference materials from Astimex Standards Ltd (MINM25-53) were used: almandine-rich garnet for Fe, Mg, Si, Al, Ta, Th, U; albite for Na, plagioclase for Ca; bustamite for Mn; celestine for Sr; tugtupite for Cl. The REEs were calibrated with element phosphates, Ba, and Ti as silicates from the Smithsonian Institution/Washington. For major elements, F, and Cl, the X-ray lines Ka1 were analyzed; La1 lines for Sr, La, Y, Ce, Eu, Gd, and Ta; Lb1 lines for Pr and Nd; and the Ma1 line for Th and U. Raw X-ray intensities were corrected using a ZAF (atomic number, absorption, fluorescence) correction procedure. Matrix corrections were performed using the software supplied by JEOL.   An acceleration voltage of 20 kV was used for REEs, while 15 kV were used for all the other elements. A beam current of 15/20 nA with a beam size of 1-15 µm was used. The counting time was 10-30 s on peaks for major elements and 50 s for REEs and other trace elements with 5 to 15 s on background. Na, K, and F, as least-stable components were analyzed first during the analytical process. To reduce migration, counting times were reduced to 6 s on peaks and 3 s on backgrounds. Total iron shown as FeO was calculated to Fe²⁺ cations in the mineral formula. Fe³⁺ was calculated by charge balance or other procedures (see table captions). In addition, detailed images and a first semi-quantitative analysis were provided by a JEOL JSM-651 equipped with an Oxford INCA x-act spectrometer.

4. Petrography

The studied samples are all classified as nepheline-syenites, composed of alkali feldspar (29-65 vol.%), nepheline (15-35 vol.%), biotite (2-9 vol.%), amphibole (0-18 vol.%), and clinopyroxene (0-10 vol.%). The minor minerals are titanite (0-8 vol.%), calcite (0-3 vol.%), cancrinite (0-4 vol.%), and sodalite  (0-3 vol.%). Trace minerals with <1 vol.% are restricted to individual samples and consist of pyrochlore, columbite, euxenite, apatite, britholite, ancylite-Ce, wöhlerite, rhodochrosite, strontianite, fluorite, zircon, magnetite, ilmenite, and pyrite.

All rocks are hololeucocratic to leucocratic with colors that vary from white to gray. They are massive and phaneritic. In general, crystal sizes range from <1 mm to 4.5 mm but some samples present a coarser-grained and inequigranular texture. Samples SNSJG 2, LC 1, and LC 2 (see figure 2 for locations) show alkali feldspar tabular crystals of up to ~2 cm. Some samples (SNSJG 4 and 5a from El Capricho and LC 1 and LC 2 from Las Delicias) display an inequigranular texture with oriented alkali feldspar phenocrysts (up to 1.5 cm), suggesting a sub-solidus magmatic flow orientation in a stress field (see Stein, 2000).

Whereas nepheline-syenites from El Capricho have feldspar abundance ranging from 55 to 60 vol.% (K-feldspar>Albite) and nepheline between 15 and   35 vol.%, the Las Delicias samples contain ~30 vol.% feldspars, ~35 vol.% nepheline, and have higher concentrations of clinopyroxene and amphibole. In all samples biotite is present with ~2-9 vol.%, independent of the sampling area.

In order to determine the importance of minor and trace minerals in the relation of Nb- and REE-rich syenites to a possible carbonatite occurrence, detailed description of the composition and fabric of major, minor, and trace minerals is necessary.

K-feldspar (Kfs) occurs either as orthoclase or microcline (Fig. 3A-E). Microcline often forms rims around orthoclase, indicating a second generation of Kfs, as shown in figure 3B. The transformation is often continuous or due to dynamic or static recrystallization (Fig. 3B). Both minerals show frequent perthitic exsolutions with cord-, patch- or vein-like morphology (Fig. 3A-E). Orthoclase and microcline appear mostly with lath-shaped forms. Microcline shows the typical tartan twinning (Fig. 3A, B) and orthoclase Carlsbad twinning (Fig. 3C). Inclusions of nepheline and magnetite are frequently observed (Fig. 3D, E), but pyrochlore and other trace minerals are common inclusions too. Kfs itself also occurs as inclusions in nepheline and apatite. The composition of Kfs, independent of microcline or orthoclase lattice, is very limited with Or_93-95 Ab_5-7 An_0-0.1 (Supplementary Table S1). Unlike nepheline, Kfs remains nearly unaffected by kaolinite.

Fig. 3. Photomicrographs and BSE images of some SJGS samples showcasing the main constituent minerals and some accessory minerals of relevance. A. Subhedral microcline (Mcl) with perthitic exsolutions (pert), albite (Ab), biotite (Bt), cancrinite (Ccn), and nepheline (Ne). Sample UNG 1762, crossed nicols. B. Orthoclase (Or) with partial tartan twinning and a rim of microcline (recrystallized K-feldspar) near to calcite (Cc). Sample SNSJG 1, crossed nicols. C. Oriented orthoclase with sodalite (Sdl) in the center. Sample LC 2, crossed nicols. D. Microcline with perthitic exsolutions and magnetite (Mgt) inclusions. Albite rims formed around nepheline inclusions. Sample SNSJG 6, BSE image. E. Nepheline with a kaolinite (Kln) rim as inclusion in microcline. Sample SNSJG 5, BSE image. F. Reaction of nepheline and biotite to form cancrinite. Sample UNG 1762, crossed nicols.

Albite (Ab_94-99 An_0.3-6 Or_0.4-0.5; Supplementary Table S1) occurs as corona (rims) around nepheline (Fig. 3D) and in Kfs or as individual grains (Fig. 3A, 3E, 4A-D), commonly in contact with nepheline, clinopyroxene, microcline, amphibole, and locally with primary calcite (Fig. 3A).

Fig. 4. Photomicrographs and BSE images of some SJGS samples showcasing the main constituent minerals and some accessory minerals of relevance. A. Anhedral clinopyroxene (Cpx), titanite (Ttn), amphibole (Amp), albite (Ab), and K-feldspar (Kfs). Sample SNSJG 4, crossed nicols. B. Amphibole (Amp) overgrown on clinopyroxene. Sample SNSJG 4, parallel nicols. C. Anhedral clinopyroxene with inclusions of titanite, rimmed by albite and intergrown with nepheline. Sample LC 2, BSE image. D. Anhedral hornblende (Hbl) with relics of clinopyroxene and inclusions of calcite (Cc), cancrinite (Ccn), titanite, and magnetite (Mgt). Sample SNSJG 4, BSE image.

Nepheline is the dominant feldspathoid. It occurs as anhedral to subhedral (hexagonal) crystals (Fig. 5A) and as inclusion in microcline and orthoclase. It contains inclusions of Kfs and rarely magnetite, biotite or calcite. In sample LC 2, large crystals with numerous inclusions of biotite and titanite are present. Inclusions of pyrite are rare. Nepheline is altered to cancrinite or intergrown in symplectites with cancrinite (Fig. 3F). Nepheline is sometimes also altered to kaolinite ranging from kaolinite rims to complete pseudomorphs (Fig. 3E). The chemical composition of nepheline is shown in Supplemantary Table S2.

Fig. 5. Apatite (Ap) and britholite (Brt) in the syenite SJGS rock suite, BSE images. A. Coexisting apatite and britholite between clinopyroxene (Cpx) and nepheline (Ne). Sample LC 2. B. Zoned apatite with a REE-rich rim. Sample LC 1. C. Britholite intergrown with clinopyroxene. Sample LC 2. D. Britholite single crystal between albite (Ab) and clinopyroxene. Cancrinite (Ccn) on the right and K-feldspar (Kfs) below. Sample LC 2. E. Small grain of apatite, overgrown by an apatite matrix, and exsolution of britholite from apatite. Sample LC 1. F. Zoom-in from e, different hues indicate different SiO2 and REE concentration domains.

Sodalite is restricted to a few samples (LC 2 and UNG 1762) and in low concentration. It forms anhedral to euhedral grains with a maximum grain size of ~0.8 mm (Fig. 3C). It contains up to 10 wt% Cl (Supplementary Table S2). It commonly has inclusions of euhedral pyrochlore.

Cancrinite with the general formula (Na,Ca,☐)₈(Al₆Si₆)O₂₄(CO₃,SO₄)₂·2H₂O is formed both as a primary phase and interstitial anhedral crystals associated with calcite, nepheline, and sodalite. More common are polycrystalline aggregates growing at the expense of nepheline (Fig. 3F, 5A), occasionally with calcite aggregates. The analyzed grains contain ~4.7 wt% SO₃ (Supplementary Table S2). Cancrinite is particularly abundant in coarse-grained samples with up to 4 vol.%.

Clinopyroxene occurs in thin sections as intense green crystals (Fig. 4B), that are commonly rimmed by amphibole±titanite (Fig. 4A, B). Additionally, calcite, titanite and magnetite are common inclusions (Fig. 4C). Clinopyroxene is often overgrown by amphibole (Fig. 4B, D). The composition of clinopyroxene depends on the sampled rocks, although all are rich in FeO_tot (total Fe as FeO) and Na₂O with concentrations of 16 to 25 wt% FeO_tot and 3.3 to 4.8 wt% Na₂O, and low in TiO₂ (<0.7 wt%) (Supplementary Table S3). The calculated components indicate that most are aegirine-augite (Supplementary Table S3).

Amphibole is present in most samples. In thin sections they show very dark green to slightly bluish colors (Fig. 4B). Chemically they are rich in FeO, Na₂O, and K₂O (Supplementary Table S4). The amphiboles (general formula A_0-1B₂C₅T₈O₂₂(OH)₂) in the syenites can be classified either as a member of the sodium-calcium amphiboles defined with ^B(Mg, Fe²⁺, Mn²⁺, Li) ≤0.5, ^B(Ca, Na) ≥1, and 0.5≤ ^BNa <1.5, or as a member of the Ca-amphiboles with ^B(Mg, Fe²⁺, Mn²⁺, Li) ≤0.5, ^B(Ca, Na) ≥1, and ^BNa <0.5 (all values in atoms per formula unit; apfu) (Leake et al., 1997, 2004). According to the nomenclature of Leake et al. (1997), the analyzed Na-Ca-amphiboles can be named as potassian taramites (sample LC 1) and the Ca-amphiboles potassian-sodian hastingsites (samples LC 2 and SNSJG 4) (Supplementary Table S4). Both types are chemically and optically very similar. These amphiboles are probably comparable with the already described, but not chemically analyzed “arfvedsonite-like amphibole” of Amaya López et al. (2021). The amphiboles occur solitary as euhedral to subhedral crystals (Fig. 4A) but more commonly as anhedral rims around greenish to brownish clinopyroxenes (Fig. 4B, D) and contain inclusions of microcline, biotite, and titanite.

Biotite occurs as isolated crystals and as intergrowths with titanite and amphibole (Fig. 3A, F). It often contains microcline inclusions, opaque minerals, zircons and pyrochlore, the latter producing pleochroic halos. Biotites are also found as inclusions in microcline and nepheline. It has low concentrations of F and high concentrations of FeO (Supplementary Table S5). Biotite shows a Fe/Fe+Mg value of about 0.86. Sample SNSJG 4 reveals lower concentrations of Fe and Fe/Fe+Mg values (Supplementary Table S5), possibly due to secondary processes.

Britholite and apatite are members of the apatite supergroup M₅(TO₄)₃X [M: Ca, Pb, Ba, Sr, Mn, Na, Ce, La, Y; T: P, V, Si, S; X: F, OH, Cl]. It includes the britholite group with the general formula (REE,Ca)₅[(Si,P)O₄]₃(OH, F) (Pasero et al., 2010). Transformation of apatite into britholite is possible by the coupled substitution mechanism (e.g., Ronsbø, 1989; Zozulya et al., 2015, 2017):

Ca²⁺+P⁵⁺↔️REE³⁺+Si⁴⁺

In the analyzed syenites, apatite crystals occur in grain sizes of up to ~450 µm (Figs. 5A, 5B ,6). They are classified as fluorapatites with 0.6-0.7 apfu F   (Fig. 7A) and are characterized by high SrO₂ (0.19-0.51 wt%) and Y₂O₃ (0.03-0.51 wt%) concentrations (Table 1). Apatite grains occur independently (Fig. 5B), rimmed by fluorbritholite (Fig. 5A) or in complex intergrowth with britholite (Figs. 5E, F, 6). Britholite occurs as fluorbritholite-Ce with 1-1.2 apfu Ce and 0.5-0.7 apfu F, discarding a possible inaccuracy of F contents due to beam migration. Electron microscopy and EPMA analyses reveal a complex growth history for the larger apatites (Fig. 6; Table 1). BSE images display several small grains, with very low SiO₂ and high light REE concentrations, overgrown by a later generation of merged apatites with slightly higher SiO₂ and light REE concentrations (Fig. 5E). These grains then underwent transformation, forming a third apatite and britholite generation by the reaction:

REE-rich apatite → REE-poor apatite + REE-rich britholite (Figs. 5E, 5F, 7B)

which is here interpreted as an exsolution texture during changing physical or chemical conditions, confirming the miscibility gap between apatite and britholite described by Melluso et al. (2012), Zozulya et al. (2017), and Lorenz et al. (2019). Anenburg et al. (2018) showed that these exsolutions are a high-temperature process. The reaction or exsolution took place in channels transecting the apatite, which can be clearly seen in the element distribution map (Fig. 6). The process is accompanied by an increase in Sr, pointing to an influx of a Sr-bearing fluid (Fig. 6F). According to Chakhmouradian et al. (2017), the abundance of Mg and Fe in mantle-derived melts can increase the partitioning of Sr and the Ca-Sr substitution in apatite. The newly crystallized britholite forms ~2-10 µm sized grains (Fig. 5E, F). However, britholite occurs more frequently as isolated, <50 µm-sized single crystals or as inclusions in nepheline, albite or clinopyroxene. Compared to apatites, britholites have lower F contents, dismissing a possible inaccuracy of F contentes (as described above), higher ThO₂ concentrations, and similar Sr contents. When present as isolated grains, britholite often appears in contact with clinopyroxene, nepheline, apatite, or albite.

Fig. 6. BSE image and element distribution maps in an SJGS apatite crystal and delineation of britholite-rich domains: A. Exsolution of britholite (light colors), comparable to figure 5E. B. Britholite can be identified by its higher light REE concentrations, here exemplified by Ce. C-D. Apatite and britholite are easily distinguishable by their different P and Si concentrations.  E. Higher Th concentrations can help differentiate britholite from apatite. F. Sr map indicating the metasomatic ingress of Sr into the britholite-bearing zone. Sample LC 1 (all images). Horizontal side of the image: 400 µm.

Fig. 7. EPMA analyses of SJGS apatite and britholite crystals. A. REE/(REE+Ca) versus F/(F+OF) diagram. B. Ca+P versus Si+REE diagram. Dashed line shows the ideal substitution Si+REE ↔ Ca + P. All units in apfu (atoms per formula unit).

Pyrochlore occurs as a common inclusion in orthoclase, nepheline, sodalite, amphibole, biotite, and albite (Fig. 8). It typically forms euhedral grains with a size of ~50-100 µm, and only in rare cases crystal sizes of up to 250 µm are reached, which exhibit metamictic alteration (Fig. 8C). Few crystals show poikilitic structures with open euhedral cavities (Fig. 9),  forming so-called negative crystals, where faces of the cavities within the crystal are planes of the host.

Fig. 8. BSE images of SJGS pyrochlore crystals. A. Euhedral pyrochlore (Pcl) in K-feldspar (Kfs). Pl=plagioclase. Sample UNG 1762. B. Subhedral pyrochlore in amphibole (Amp). Bt=Biotite. Sample LC 1. C. Subhedral to euhedral and partially fractured pyrochlore in feldspar. Sample LC 1. D. Euhedral pyrochlore in biotite. Ne=Nepheline; Mgt=Magnetite. Sample LC 1. E. Detail from D, pyrochlore in nepheline shows inclusions of magnetite (light), albite (Ab, gray), nepheline (dark), and holes (dark). F. Euhedral pyrochlore in nepheline. Ap=Apatite. Sample UNG 1762.

Fig. 9. BSE (A, C) and SE (B, D) images of SJGS pyrochlore crystals. Zoom-ins from A and B are shown in C and D, respectively. Sample UNG 1762.

The analyzed pyrochlores are characterized by Nb+Ta > 2Ti and Nb > Ta (Table 2), being therefore pyrochlores following the classification of Hogarth (1977). The general formula of the pyrochlore supergroup is A₂B₂X₆Y, where the A site contains the large Na, Ca, Mn, Sr, Y, and REE cations in 8-fold coordination, and the B site contains the smaller 6-fold coordinated Ta, Nb, Ti, Fe³⁺, Mg, Al, and Si.

The classical pyrochlore nomenclature of Hogarth (1977) was modified by Atencio et al. (2010) and approved by the Commission on New Minerals Nomenclature and Classification (CNMNC) of the International Mineralogical Association (IMA). According to this new nomenclature, the high Na/Ca and Nb/Ta ratios, and the high fluorine content indicate that the crystals analyzed are fluornatropyrochlores. The EPMA analytical total is low (around 91-93 wt%) and the A site is below 2 apfu, which points to the presence of H₂O (Atencio et al., 2010). In addition, the crystals have light REEs of ~1 wt%. The composition of pyrochlore varies depending on the samples analyzed. For example, sample UNG 1762 has more Na, Sr, and F, and less U, Ti, and Si than those in sample LC 1. Element mapping shows that some pyrochlores are zoned, with an inner core and an outermost rim significantly richer in Na and F and poorer in U and Th (Fig. 10). The cavities between the core and inner regions (black colors) help explain the lower concentrations in the element distribution maps.

Fig. 10. BSE image (upper left) and element distribution maps of a pyrochlore crystal from sample UNG 1762 shown in figure 9. The strong zonation is due to an increase in Na, Ti and F concentrations. A false zonation is also given by the holey structure. Same color code for all element concentrations.

Calcite is usually found as isolated primary crystals in nepheline, microcline, and biotite or in contact with nepheline and albite (Fig. 11A, B). In addition, microcrystalline calcite aggregates are associated with cancrinite altering nepheline. However, in the latter case calcite shows high-temperature grain boundaries in contact with albite. Primary calcite is rich in light REEs with a total of up to 1,290 ppm (Table 3). La is up to 2,300 times and Ce up to 1,000 times above the chondrite-normalized concentrations. Additionally, primary calcite contains up to 3 wt% SrO, but is poor in MnO (≤0.84 wt%) and MgO (≤0.07 wt%), with low MgO/MnO ratios (0.06-0.08). The FeO concentrations are also low (≤0.35 wt%) (Table 3).

Fig. 11. Photomicrographs and BSE images of some SJGS samples showcasing carbonates. A. Euhedral primary calcite (Cc) at the albite-nepheline (Ab-Ne) interface. Bt=Biotite. Py=Pyrite. Sample SNSJG 6. B. Primary calcite between albite and recrystallized cancrinite (Ccn). Sample SNSJG 1. C. Rhodochrosite (Rds) with Mn-rich columbite (Cmb). Amp=Amphibole. Sample UNG 1762. D. Ancylite-Ce aggregate (Anc) in K-feldspar (Kfs). Sample UNG 1762. E. Strontianite (Str) as secondary product at the contact between albite and K-feldspar. Sample UNG 1762. F. Strontianite in fractured biotite. Sample UNG 1762.

Rhodochrosite occurs rarely. It forms up to 80 µm-long, anhedral to subhedral inclusions in amphibole. Itself contains µm-sized inclusions of columbite (Fig. 11C), which are probable exsolutions. The chemical composition of a SJGS rhodochrosite crystal is shown in table 3.

Ancylite-Ce appears as individual grains with sizes <200 µm and <400 µm as aggregates (Fig. 11D). It typically occurs as inclusions, but due to its (OH) content it is probably of secondary origin, so the interpretation of an earlier crystallized phase would possibly be a mistake result of a cut effect. The chemical composition of a SJGS ancylite-Ce crystal is shown in table 3.

Strontianite with up to 12 wt% CaO occurs as an alteration product in Fe-rich amphiboles and K-feldspar (Fig. 11E, F; Table 3).

Columbite occurs as euhedral inclusion or exsolution with a grain size of up to 10 µm in rhodochrosite (Fig. 11C). Columbite inclusions in the rhodochrosite represent chemically a Ta-poor (0.003 apfu Ta) manganocolumbite (or columbite-Mn) with a Mn+Fe value of 1.34 apfu (Table 4).

Euxenite occurs in sample LC 1 as a <50 µm-long euhedral inclusion in nepheline. It contains up to 4.2 wt% UO₂ and 2.3 wt% Ce₂O₃ (Table 4). The presence of euxenite as an inclusion in nepheline indicates an early crystallization in the magmatic evolution.

Wöhlerite occurs as very rare grains (Fig. 12A). Crystals are typically <250 µm-long, anhedral to subhedral, and form grain boundaries with clinopyroxene. The chemical composition of a SJGS wöhlerite crystal is shown in table 4.

Fig. 12. Photomicrographs and BSE images of some SJGS samples showing wöhlerite and iron oxides. A. Wöhlerite (Wöh). Sample   LC 2. B. Fluorite (Fl) in nepheline (Ne). Sample Muestra 3. C. Magnetite (Mgt, white domains) with exsolution lamellae of ilmenite (Ilm, gray elongated domains) as inclusion in nepheline. Sample SNSJG 5a. D. Magnetite and ilmenite in Kfs. Sample SNSJG 5a.

Titanite is frequently associated with opaque minerals (ilmenite-magnetite) and can be found as inclusions in clinopyroxene, amphibole, nepheline, biotite, and K-feldspar. It exhibits single polysynthetic twinning and is commonly euhedral, indicating early growth (Fig. 4A, C, D). It contains up to 2.2 wt% FeO (0.062 apfu) and traces of MnO, MgO, and Na₂O (Table 5).

Iron Oxide minerals in the syenites are commonly magnetite and ilmenite. They occur as individual grains or as ilmenite exsolutions in magnetite. The composition these iron oxides is very similar, independent of their occurrence as host grains, exsolutions or as individual grains (Table 6). The Fe²⁺/Fe³⁺ ratio of magnetite is constant around ~0.5. Ilmenite is Mn-rich with up to 15 wt% MnO (~0.3 apfu). Both minerals have mostly anhedral shapes although some exhibit cubic shapes (Fig. 12C, D). Ilmenite-magnetite pairs sometimes contain nepheline inclusions (Fig. 12c).

Pyrite is very rare and detectable as inclusions in nepheline (Fig.11A).

Fluorite appears rarely as up to 100 µm-large purplish inclusions in nepheline (Fig. 12B). It contains 0.56 wt% SrO, and 0.135 wt% FeO (Table 5). The REE concentrations are below the detection limit of the EPMA, except for Ce (250 ppm), Nd (350 ppm), and Gd (170 ppm).

Zircon appears in the studied samples very rarely and will not be discussed here (one grain in three thin sections of one rock sample). Larger zircons in the SJGS rock suite are described by Amaya López et al. (2021).

Kaolinite is a common secondary mineral, which in our samples is seen altering exclusively nepheline (Fig. 3E). The kaolinite forms rims around nepheline as well as complete substitutions.

5. Discussion and conclusions

Numerous accessory minerals, several dominated by so-called critical elements such as Nb and REEs, are present in the SJGS samples. These accessory minerals typically present along with mostly major mineral phases, such as nepheline, potassium feldspar, albite, biotite, titanite, Fe-rich clinopyroxenes, and Fe-rich amphiboles. In the study area, accessory minerals can be used as tracers to evaluate fluid and melt interactions during the SJGS magmatic evolution.

The textural and compositional characteristics of the SJGS minerals suggest different magmatic stages. We propose an early high-temperature crystallization stage dominated by the growth of nepheline, sodalite, potassium feldspar (orthoclase), Fe-rich clinopyroxene, albite, apatite, amphibole, and biotite. Inclusions of calcite and Nb-rich minerals (pyrochlore, columbite, and euxenite) indicate that they were formed during this early stage as well.

We suggest a large proportion of SJGS calcite is primary in origin. The high abundance of light REEs (~1,290 ppm) in calcite is similar to those observed in calcites from experiments on silica-rich carbonatites (sövites) of Oka (Canada), Kaiserstuhl (Germany), and Alnö (Sweden), which vary between 300 and 1,100 ppm (e.g., Hornig-Kjarsgaard, 1998). This abundance contrasts usually lower light REE concentrations in sedimentary and hydrothermal calcite. Sedimentary calcites, like the late Devonian reefal calcites (Australia), have light REE concentrations below 10 ppm (Nothdurfth et al., 2004). Likewise, hydrothermally formed calcites show lower light REE concentrations than those from calcites in carbonatites (Stipp et al., 2006).  Furthermore, the SJGS calcites have low MgO (≤0.07 wt%) and high SrO (~2.9 wt%) concentrations (Table 3), which is typical for calcites that crystallized from experimental silica-rich carbonatite melts (e.g., Hornig-Kjarsgaard, 1998). Additionally, the very low MgO/MnO ratios (0.062-0.083) are comparable to magmatic calcite, in contrast to hydrothermally formed calcites (Chakhmouradian et al.,2016). Calcite crystallization textures (Fig. 11A, B) suggest formation at high temperatures (e.g., Philpotts and Ague, 2021), and albite rims around nepheline (Fig. 3D) indicate high pressures during the crystallization of these mineral phases (i.e., fractionation of albite after nepheline), all in accordance with experimental studies (e.g., Philpotts and Ague, 2021). The high SrO concentrations in apatite (0.2-0.5 wt%) are typical for apatite crystallized from carbonatitic melts as well (Belousova et al., 2002), whereas apatite from mafic rocks has significant lower SrO concentrations (<0.13 wt%; Belousova et al.,2002). Finally, the high FeO and low TiO₂ contents in clinopyroxene (Supplementary Table S3) point to high CO₂ activities in the free volatile phase, i.e., CO₂-saturated conditions (e.g., Dolfi, 1996; Gozzi et al.,2014).

A decrease in temperature and an increase in P_H2O (P_H2O > P_CO2) is indicated by the transition of Fe-rich clinopyroxene into Fe-rich amphibole. Another process that possibly takes place simultaneously is the deformation-enhanced triclinization of orthoclase, forming microcline (some feldspars with shape-preferred orientation also show evidence of deformation during an early intrusion stage under an active stress field). Exsolution of albite from alkali feldspar during the formation of perthitic microcline as a second alkali feldspar generation also points to slow cooling from high temperatures. Britholite, a solid solution member of the apatite supergroup (higher Si and REE contents; see table 1), forms as overgrowths on apatite, or replace it as a product of the Ca+P ↔ REE+Si exchange reaction. Britholite has primarily been described from nepheline syenites (e.g., Ilimaussaq intrusive complex, Greenland; Winther, 1901), carbonatites (e.g., Angola; Torró et al., 2012; Anenburg et al., 2018), and metasomatized halos around alkaline intrusions (e.g., Kola Peninsula; Zozulya et al., 2015, 2017). According to the phase diagram of Anenburg et al. (2018), britholite can crystallize during cooling and redistribution of REEs from apatite, still at high (400-700 °C) temperatures. From Nolans Bore, a P-REE-Th-rich vein-style deposit in Australia, Anenburg et al. (2018) described exsolution of britholite from fluorapatite at a solvus under high CO₂ and Si activities. Synchronous, and probable reaction-enhanced, was the ingress of (possibly internally derived) Sr-rich fluids into the studied apatites.

As the cooling progressed, late magmatic hydrothermal-metasomatic processes dominated by fluids with variable P_CO2/P_H2O are represented by secondary cancrinite, Sr- and Mn-carbonates and ancylite-Ce. Ancylite is a typical hydrothermal alteration product (e.g., Larsen and Gault, 2002). Cancrinite occurs as a primary magmatic phase with calcite, nepheline, and sodalite, but more often as polycrystalline aggregates in nepheline. In the samples analyzed, the secondary growth of cancrinite probably occurred synchronously with the formation of the late magmatic carbonates strontianite (Sr) and rhodochrosite (Mn), which grew at the expense of feldspars and amphiboles. The influx of Sr into the system, transforming apatite and forming Sr-carbonates, possibly occurred coevally and had the same source.

The magmatic evolution of the SJGS rocks can be summarized as follows: a CO₂-rich environment during the early and late stages with variable contributions of (OH) and CO₂. The presence of magmatic calcite and apatite in the earliest stages of the SJGS nepheline-syenites can be addressed using different models:

1. The formation of two primary, independent melts derived from a highly oxidized carbonate-metasomatized lithospheric mantle (e.g., Wallace and Green, 1988; Ying et al., 2004; Braunger et al., 2018). In the case of the SJGS, this would imply a mingling stage of the two melts, resulting in larger carbonatite bodies, nodules or spheres. These textures are not observed; however, this process cannot be completely ruled out.
2. The more commonly discussed model suggests a single primary melt derived from a contaminated lithospheric mantle that fractionates and further decouples a carbonatitic residual melt by liquid immiscibility (Gittins, 1989; Beccaluva et al., 1992; Bell and Rukhlov, 2004; Downes et al., 2005; Halama et al., 2005; Brooker and Kjasgaard, 2011). Gittins and Mitchell (2023), however, discuss that there is still no reliable experimental work proving that a nephelinitic magma can develop to such an extent that separation of immiscible silicate and carbonate melts takes place. In our case study, the SJGS would be the carbonate-bearing silicate (syenite) residuum, with calcite, apatite, pyrochlore, and/or cancrinite. Due to its low density, the decoupled carbonatite melt reaches shallow intrusion levels more rapidly than silicate melts. Mineralogically speaking, this model fits with the syenites analyzed here.
3. “Simple” crystallization differentiation of a primary magma derived from contaminated mantle (Gittins and Mitchell, 2023). In this case, calcite is part of the silicate melt fractionation, and large carbonatite volumes decoupled from the evolving melt are lacking. Although the calcite textures observed favor this model, the apatite and pyrochlore compositions (U-rich) do not (Vasyukova and Williams-Jones, 2023).
4. Crystallization under the metasomatic influence of carbonatitic melts on solid rocks, forming secondary syenites (Vasyukova and Williams-Jones, 2022). The low abundance of calcite in the syenites and the absence of a convincing evidence for additional metasomatic processes render this model unlikely. Although the presence of U-rich pyrochlores could be due to interaction with carbonate melts (Vasyukova and Williams-Jones, 2022), no other evidence has been found in the SJGS that points to a metasomatic influence, such as biotitite or phoscorite formation as described by Vasyukova and Williams-Jones (2022).
5. Melting of carbonated crustal rocks. The typical light REE and Sr concentrations in calcites from melted crustal carbonates are significantly lower (>10 times lower light REE, >2 times lower Sr), than in the SJGS calcites, so this model is rendered as unlikely as well (see Gozzi et al., 2014).

In conclusion, the particular calcite and apatite compositions in the SJSG area suggest a previous coexistence with a carbonatitic melt. Cancrinite (Maravic and Morteani, 1980), britholite (Djeddi et al., 2021) and ancylite (Eby, 1975; Hornig-Kjarsgaard, 1998) reported from syenites coexisting with carbonatites elsewhere probably formed during the cooling or hydrothermal stages. We therefore favor model II, i.e., a single melt derived from a contaminated lithospheric mantle that upon advanced fractionation decoupled a carbonatitic residual melt by liquid immiscibility. Model I (two independent primary melts), although less likely, cannot be completely ruled out. The SJGS rocks possibly represent an older member of a nepheline-syenite-carbonatite association, such as those reported from the Southern Victoria Land, Antarctica (Worley et al., 1995) and the chemically equivalent phonolites associated with carbonatites in Kaiserstuhl (Germany) (Wimmenauer, 1962; Braunger et al., 2018). The unmixed carbonatite melts that coexisted with and decoupled from the parent melt possibly intruded shallow crustal levels and was subsequently eroded.

Acknowledgements
We would like to specially thank B. Fabian for drawing figure 1 and C. Möller, A. Musiol (Potsdam), and T. Hägar (Mainz) for supporting analytical procedures. M. Sieber and the reviews of M. Anenburg and M. Tshiningayamwe gave valuable and helpful comments to improve and condensate the manuscript. Special thanks D. Bertin for intensive editorial work. Thanks also to M. Timmerman (Potsdam) and E. Sobel (Potsdam), who helped correct the English grammar. H. Pingel (Potsdam) helped improve some figures.

References

Anenburg, M.; Walters, J.B. 2024. Metasomatic ijolite, glimmerite, silicocarbonatite, and antiskarn formation: carbonatite and silicate phase equilibria in the system Na₂O-CaO-K₂O-FeO-MgO-Al₂O₃-SiO₂-H₂O-O₂-CO₂. Contributions to Mineralogy and Petrology 179: 40. https://doi.org/10.1007/s00410-024-02109-0

Anenburg, M.; Burnham, A.D.; Mavrogenes, J.A. 2018. REE redistribution textures in altered fluorapatite: symplectites, veins, and phosphate-silicate-carbonate assemblages from the Nolans Bore P-REE-Th Deposit, Northern Territory, Australia. The Canadian Mineralogist 56 (3): 331-354. https://doi.org/10.3749/canmin.1700038

Anenburg, M.; Broom-Fendley, S.; Chen, W. 2021. Formation of Rare Earth deposits in carbonatites. Elements 17 (5): 327-332. https://doi.org/10.2138/gselements.17.5.327

Amaya López, C.; Weber, M.; Ibáñez-Mejía, M.; Cuadros, F.A.; Restrepo, J.J.; Botelho. N.; Maya, M.; Pérez, O.; Ramírez, C. 2021. San José del Guaviare Syenite, Colombia: repeated Ediacaran intrusions in the northwestern Amazonian Craton. Boletín Geológico 48 (1): 49-79. http://dx.doi.org/10.32685/0120-1425/bol.geol.48.1.2021.503

Arango. M.I.; Nivia. A.; Zapata, G.; Giraldo, M.I.; Bermúdez, J.G.; Albarracín, H. 2011. Geología y geoquímica de la Plancha 350 San José del Guaviare. Memoria explicativa. Escala 1:100.000. Servicio Geológico Colombiano: 166 p. Medellín. https://recordcenter.sgc.gov.co/B13/23008010024451/documento/pdf/2105244511101000.pdf

Atanasova, P.; Marks, M.A.W.; Heinig, T.; Krause, J.; Gutzmer, J.; Markl, G. 2017. Distinguishing magmatic and metamorphic processes in peralkaline rocks of the Norra Kärr complex (southern Sweden) using textural and compositional variations of clinopyroxene and eudialyte-group minerals. Journal of Petrology 58 (2): 361-384. https://doi.org/10.1093/petrology/egx019

Atencio, D.; Andrade, M.B.; Christy, A.G.; Gieré, R.; Kartashov, P.M. 2010. The pyrochlore supergroup of minerals: nomenclature. The Canadian Mineralogist 48 (3): 673-698. https://doi.org/10.3749/canmin.48.3.673

Beard, C.D.; Goodenough, K.M.; Borst, A.M.; Wall, F.; Siegfried, P.R.; Deady, E.A.; Pohl, C.; Hutschinson, W.; Finch, A.A.; Walter, B.F.; Elliot, H.A.L.; Brauch, K. 2023. Alkaline-silicate REE-HFSE systems. Economic Geology 118 (1): 177-208. https://doi.org/10.5382/econgeo.4956

Beccaluva, L.; Barbieri, M.; Born, H.; Brotzu, P.; Coltorti, M.; Conte, A.; Garbarino, C.; Gomes, C.B.; Macciotta, G.; Morbidelli, L.; Ruberti, E.; Siena, F.; Traversa, G. 1992. Fractional crystallization and liquid immiscibility processes in the alkaline-carbonatite complex of Juquiá (São Paulo, Brazil). Journal of Petrology 33 (6): 1371-1404. https://doi.org/10.1093/petrology/33.6.1371

Bell, K.; Rukhlov, A.S. 2004. Carbonatites from the Kola alkaline province: origin, evolution and source characteristics. In Phoscorites and carbonatites from mantle to mine: the key example of the Kola alkaline province (Zaitsev, A.N.; Wall, F.; editors). The Mineralogical Society Series 10: 433-468. London. https://doi.org/10.1180/MSS.10.13

Bell, K.; Simonetti, A. 2010. Source of parental melts to carbonatites - critical isotopic constraints. Mineralogy and Petrology 98: 77-89. https://doi.org/10.1007/s00710-009-0059-0

Belousova, E.A.; Griffin, W.L.; O’Reilly, S.Y.; Fisher, N.I. 2002. Apatite as an indicator mineral for mineral exploration: trace-element compositions and their relationship to host rock type. Journal of Geochemical Exploration 76 (1): 45-69. https://doi.org/10.1016/S0375-6742(02)00204-2

Bonin, B.; Giret, A. 1984. The plutonic alkaline series: the problem of their origin and differentiation, the role of their mineralogical assemblages. Physics of the Earth and Planetary Interiors 35 (1-3): 212-221. https://doi.org/10.1016/0031-9201(84)90044-X

Braunger, S.; Marks, M.A.W.; Walter, B.F.; Neubauer, R.; Reich, R.; Wenzel, T.; Parsapoor, A.; Markl, G. 2018. The petrology of the Kaiserstuhl Volcanic Complex, SW Germany: the importance of metasomatized and oxidized lithospheric mantle for carbonatite generation. Journal of Petrology 59 (9): 1731-1762. https://doi.org/10.1093/petrology/egy078

Brooker, R.; Kjarsgaard, B. 2011. Silicate-carbonate liquid immiscibility and phase relations in the system SiO₂-Na₂O-Al₂O₃-CaO-CO₂ at 0.1-2.5GPa with applications to carbonatite genesis. Journal of Petrology 52 (7-8): 1281-1305. https://doi.org/10.1093/petrology/egq081

Cawood, P.A.; Pisarevsky, S.A. 2017. Laurentia-Baltica-Amazonia relations during Rodinia assembly. Precambrian Research 292: 386-397. https://doi.org/10.1016/j.precamres.2017.01.031

Cawthorn, R.G.; Collerson, K.D. 1974. The recalculation of pyroxene end-member parameters and the estimation of ferrous and ferric iron content from electron microprobe analyses. American Mineralogist 59 (11-12): 1203-1208.

Chakhmouradian, A.R.; Williams, C.T. 2004. Mineralogy of high-field-strength elements (Ti, Nb, Zr, Ta, Hf) in phoscoritic and carbonatitic rocks of the Kola Peninsula, Russia. In Phoscorites and carbonatites from mantle to mine: the key example of the Kola alkaline province (Zaitsev, A.N.; Wall, F.; editors). The Mineralogical Society Series 10: 293-340. London. https://doi.org/10.1180/MSS.10.9

Chakhmouradian, A.R.; Wall, F. 2012. Rare earth elements: minerals, mines, magnets (and more). Elements 8 (5): 333-340. https://doi.org/10.2113/gselements.8.5.333

Chakhmouradian, A.R.; Reguir, E.P.; Couëslan, C.; Yang, P. 2016. Calcite and dolomite in intrusive carbonatites. II. Trace-element variations. Mineralogy and Petrology 110: 361-377. https://doi.org/10.1007/s00710-015-0392-4

Chakhmouradian, A.R.; Reguir, E.P.; Zaitsev, A.N.; Couëslan, C.; Xu, C.; Kynický, J.; Mumin, A.H.; Yang, P. 2017. Apatite in carbonatitic rocks: compositional variation, zoning, element partitioning and petrogenetic significance. Lithos 274-275: 188-213. https://doi.org/10.1016/j.lithos.2016.12.037

Chiwona, A.G.; Cortés, J.A.; Gaulton, R.G.; Manning, D.A.C. 2020. Petrology and geochemistry of selected nepheline syenites from Malawi and their potential as alternative potash sources. Journal of African Earth Sciences 164: 103769. https://doi.org/10.1016/j.jafrearsci.2020.103769

Condie, K.C. 2011. The supercontinent cycle. In Earth as an evolving planetary system (Second Edition) (Condie, K.C.; editor): 317-355. Academic Press. https://doi.org/10.1016/B978-0-12-385227-4.00003-1

Cordani, U.G.; Teixeira, W. 2007. Proterozoic accretionary belts in the Amazonian Craton. In 4-D framework of continental crust (Hatcher, R.D.; Carlson, M.P.; McBride, J.H.; Catalan, J.R.M.; editors). Geological Society of America Memoirs 200: 297-320. https://doi.org/10.1130/2007.1200(14)

Djeddi, A.; Parat, F.; Bodinier, J.-L.; Ouzegane, K.; Dautria, J.-M. 2021.The syenite-carbonatite complex of Ihouhaouene (Western Hoggar, Algeria): interplay between alkaline magma differentiation and hybridization of cumulus crystal mushes. Frontiers in Earth Science 8: 605116. https://doi.org/10.3389/feart.2020.605116

Dolfi, D. 1996. Preliminary results of the effect of oxygen fugacity on magmatic clinopyroxene and coexisting liquid compositions. Periodico di Mineralogia 65: 5-14.

Downes, H.; Balaganskaya, E.; Beard, A.; Liferovich, R.; Demaiffe, D. 2005. Petrogenetic processes in the ultramafic, alkaline and carbonatitic magmatism in the Kola Alkaline Province: a review. Lithos 85: 48-75.

Droop, G.T.R. 1987. A general equation for estimating Fe³⁺ concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria. Mineralogical Magazine 51 (361): 431-435. https://doi.org/10.1180/minmag.1987.051.361.10

Eby, G.N. 1975. Abundance and distribution of the rare earth elements and yttrium in the rocks and minerals of the Oka carbonatite complex, Quebec. Geochimica et Cosmochimica Acta 39 (5): 597-620. https://doi.org/10.1016/0016-7037(75)90005-8

Engler, A. 2009. The geology of South America. In Encyclopedia of life support systems: Geology, Volume IV (De Vivo, B.; Grasemann, B.; Stüwe, K.; editors). Eolss Publishers Co. Ltd.: 457 p. Oxford.

Finch, A.A.; McCreath, J.A.; Reekie, C.D.J.; Hutchison, W.; Ismaila, A.; Armour-Brown, A.; Andersen, T.; Simonsen, S.L. 2019. From Mantle to Motzfeldt: a genetic model for syenite-hosted Ta, Nb-mineralisation. Ore Geology Reviews 107: 402-416. https://doi.org/10.1016/j.oregeorev.2019.02.032

Franco-Victoria, J.A.; Muñoz, J.A.; Piraquive, A.; Bonilla, A.; Amaya, Z.; Cramer, T.; Campos, H. 2018. Geochronology of the nepheline syenite of el Jordán, Guaviare Colombia, evidence of Neoproterozoic-Cambrian intraplate magmatism and its implications during Pan-African tectonics in western Gondwana. 20^th EGU General Assembly, Proceedings: p. 10861. Vienna.

Galvis. J.; Huguett. A.; Ruge, P. 1979. Geología de la Amazonía Colombiana: Informe N°1792. Boletín Geológico 22 (3): 3-86.

Gittins, J. 1989. The origin and evolution of carbonatite Magmas. In Carbonatites: genesis and evolution (Bell, K.; editor). Unwin Hyman: 580-600. London.

Gittins, J.; Mitchell, R.H. 2023. The genesis of calcite and dolomite carbonatite-forming magma by liquid immiscibility: a critical appraisal. Geological Magazine 160 (8) : 1463-1480. https://doi.org/10.1017/S001675682300050X

Gómez, J.; Montes, N.E. 2020. Geological map of Colombia 2020. Scale 1:1,000,000. Servicio Geológico Colombiano: 2 sheets. Bogotá. https://www2.sgc.gov.co/MGC/Paginas/gmc_1M2020.aspx

Gozzi, F.; Gaetta, M.; Freda, C.; Molla, S.; Di Rocco, T.; Marra, F.; Dallai, L.; Pack, A. 2014. Primary magmatic calcite reveals origin from crustal carbonate. Lithos 190-191: 191-203. https://doi.org/10.1016/j.lithos.2013.12.008

Halama, R.; Vennemann, T.; Siebel, W.; Markl, G. 2005. The Grønnedal-Ika carbonatite-syenite complex, south Greenland: carbonatite formation by liquid immiscibility. Journal of Petrology 46 (1): 191-217. https://doi.org/10.1093/petrology/egh069

Hogarth, D.D. 1977. Classification and nomenclature of the pyrochlore group. American Mineralogist 62 (5-6): 403-410.

Hornig-Kjarsgaard, I.M. 1998. Rare earth elements in sövitic carbonatites and their mineral phases. Journal of Petrology 39 (11-12): 2105-2121. https://doi.org/10.1093/petroj/39.11-12.2105

Ibáñez-Mejía, M.; Cordani, U.G. 2020. Zircon U-Pb geochronology and Hf-Nd-O isotope geochemistry of the Paleo- to Mesoproterozoic basement in the westernmost Guiana Shield. In The Geology of Colombia, Volume 1 Proterozoic - Paleozoic (Gómez, J.; Mateus-Zabala, D.; editors). Servicio Geológico Colombiano, Publicaciones Geológicas Especiales 35: 65-90. https://doi.org/10.32685/pub.esp.35.2019.04

Ibáñez-Mejía, M.; Ruiz, J.; Valencia, V.A.; Cardona, A.; Gehrels, G.E.; Mora, A.R. 2011. The Putumayo orogen of Amazonia and its implications for Rodinia reconstructions: new U-Pb geochronological insights into the Proterozoic tectonic evolution of northwestern South America. Precambrian Research 191 (1-2): 58-77. https://doi.org/10.1016/j.precamres.2011.09.005

Jones, A.P.; Genge, M.; Carmody, L. 2013. Carbonate melts and carbonatites. Reviews in Mineralogy and Geochemistry 75 (1): 289-322. https://doi.org/10.2138/rmg.2013.75.10

Keller, J. 1984a. Der jungtertiäre vulkanismus südwestdeutschlands: Exkursionen im Kaiserstuhl und Hegau. Fortschritte der Mineralogie 62: 2-35.

Keller, J. 1984b. Geochemie und magmenentwicklung im Kaiserstuhl. Fortschritte der Mineralogie 62: 116-118.

Larrota Rincón, D.; Concha, A.E. 2018. Petrografía y geoquímica de la sienita nefelínica de San José del Guaviare en cercanías de El Capricho. Geología Colombiana 41: 27-42.

Larsen, A.O.; Gault, R.A. 2002. Calcio-ancylite-(Ce) from syenite pegmatite at Tvedalen, Oslo Region, Norway. Neues Jahrbuch für Mineralogie - Monatshefte 9: 411-423. https://doi.org/10.1127/0028-3649/2002/2002-0411

Leake, B.E.; Woolley, A.R.; Arps, C.E.S.; Birch, W.D.; Gilbert, M.C.; Grice, J.D.; Hawthorne, F.C.; Kato, A.; Kisch, H.J.; Krivovichev, V.G.; Linthout, K.; Laird, J.; Mandarino, J.; Maresch, W.V.; Nickel, E.H.; Rock, N.M.S.; Schumacher, J.C.; Smith, D.C.; Stephenson, N.C.N.; Ungaretti, L.; Whittaker, E.J.W.; Youzhi, G. 1997. Nomenclature of amphiboles; report of the subcommittee on amphiboles of the International Mineralogical Association Commission on New Minerals and Mineral Names. Mineralogical Magazine 61: 295-310. https://doi.org/10.1180/minmag.1997.061.405.13

Leake, B.E.; Woolley, A.R.; Birch, W.D.; Burke, E.A.J.; Ferraris, G.; Grice, J.D.; Hawthorne, F.C.; Kisch, H.J.; Krivovichev, V.G.; Schumacher, J.C.; Stephenson, N.C.N.; Whittaker, E.J.W. 2004. Nomenclature of amphiboles: additions and revisions to the International Mineralogical Association’s amphibole nomenclature. Mineralogical Magazine 68 (1): 209-215. https://doi.org/10.1180/0026461046810182

Lorenz, M.; Altenberger, U.; Trumbull, R.B.; Lira, R.; López de Luchi, M.; Günter, C., Eidner, S. 2019. Chemical and textural relations of britholite- and apatite-group minerals from hydrothermal REE mineralization at the Rodeo de los Molles deposit, Central Argentina. American Mineralogist 104 (12): 1840-1850. https://doi.org/10.2138/am-2019-6969

Maravic, H.v.; Morteani, G. 1980. Petrology and geochemistry of the carbonatite and syenite complex of Lueshe (N.E. Zaire). Lithos 13 (2): 159-170. https://doi.org/10.1016/0024-4937(80)90017-1

Marks, M.A.W.; Markl, G., 2017. A global review on agpaitic rocks. Earth-Science Reviews 173: 229-258. https://doi.org/10.1016/j.earscirev.2017.06.002

Maya, M.; Amaya López, C.; Restrepo, J.J.; Duque, J.; Palacio, A.F.; Gutiérrez, P.A.; Pérez, O.; Ríos, C.; Arias, E.N.; Bedoya, J.D. 2018. Memoria Explicativa de la Plancha 372-El Retorno. Servicio Geológico Colombiano: 330 p. Bogotá.

Melluso, L.; de’ Gennaro, R.; Fedele, L.; Franciosi, L.; Morra, V. 2012. Evidence of crystallization in residual, Cl-F-rich, agpaitic, trachyphonolitic magmas and primitive Mg-rich basalt-trachyphonolite interaction in the lava domes of the Phlegrean Fields (Italy). Geological Magazine 149 (3): 532-550. https://doi.org/10.1017/S0016756811000902

Nothdurft, L.K.; Webb, G.F.; Kamber, B.Z. 2004. Rare earth element geochemistry of Late Devonian reefal carbonates, Canning Basin, Western Australia: confirmation of a seawater REE proxy in ancient limestones. Geochimica et Cosmochimica Acta 68 (2): 263-283. https://doi.org/10.1016/S0016-7037(03)00422-8

Pasero, M.; Kampf, A.R.; Ferraris, C.; Pekov, I.V.; Rakovan, J.; White, T.J. 2010. Nomenclature of the apatite supergroup minerals. European Journal of Mineralogy 22 (2): 163-179. https://doi.org/10.1127/0935-1221/2010/0022-2022

Philpotts, A.R.; Ague, J.J. 2021. Principles of Igneous and Metamorphic Petrology, 3^rd Edition. Cambridge University Press: 778 p. Cambridge. https://doi.org/10.1017/9781108631419

Priem, H.N.A.; Kroonenberg, S.B.; Boelrijk, N.A.I.M.; Hebeda, E.H. 1982. Rb-Sr and K-Ar evidence for the presence of a 1.6 Ga basement underlying the 1.2 Ga Garzón-Santa Marta granulite belt in the Colombian Andes. Precambrian Research 42 (3-4): 315-324. https://doi.org/10.1016/0301-9268(89)90016-8

Rodriguez-Corcho, A.F.; Rojas-Agramonte, Y.; Barrera-González, J.A.; Marroquín-Gómez, M.P.; Bonilla-Correa, S.; Izquierdo-Camacho, D.; Delgado-Balaguera, S.M.; Cartwright-Buitrago, D.; Muñoz-Granados, M.D.; Carantón-Mateus, W.G.; Corrales-García, A.; Laverde-Martinez, A.F.; Cuervo-Gómez, A.; Rodríguez-Ruiz, M.A.; Marín-Jaramillo, J.P.; Salazar-Cuellar, N.; Daroca, M.E.; Carvajal, A.S.; Perea-Pescador, A.M.; Solano-Acosta, J.D.; Díaz, J.; Guillen, A.; Bayona, G.; Cardona-Molina, A.; Eglington, B.; Montes, C. 2021. The Colombian geochronological database (CGD). International Geology Review 64 (12): 1635-1669. https://doi.org/10.1080/00206814.2021.1954556

Rønsbo, J.G. 1989. Coupled substitutions involving REEs and Na and Si in apatites in alkaline rocks from the Ilimaussaq intrusion, South Greenland, and the petrological implications. American Mineralogist 74 (7-8): 896-901.

Stein, E. 2000. On the emplacement of granitoids: Case studies comparing fabrics and emplacement mechanism of the White-Inyo Mountains, California, USA, and the Bergsträsser Odenwald, Germany. E. Schweizerbart’sche Verlagsbuchhandlung: 344 p. Stuttgart.

Stipp, S.L.S.; Christensen, J.T.; Lakshtanov, L.Z.; Baker, J.A.; Waight, T.E. 2006. Rare Earth element (REE) incorporation in natural calcite: Upper limits for actinide uptake in a secondary phase. Radiochimica Acta 94 (9-11): 523-528. https://doi.org/10.1524/ract.2006.94.9-11.523

Stoppa, F.; Schiazza, M.; Rosatelli, G.; Castorina, F.; Sharygin, V.V.; Ambrosio, F.A.; Vicentini, N. 2019. Italian carbonatite system: From mantle to ore-deposit. Ore Geology Reviews 114. https://doi.org/10.1016/j.oregeorev.2019.103041.

Teixeira, W.; Reis, N.J.; Bettencourt, J.S.; Klein, E.L.; Oliveira, D.C. 2019. Intraplate Proterozoic magmatism in the Amazonian craton reviewed: Geochronology, crustal tectonics and global barcode matches. In Dyke swarms of the world: A modern perspective (Srivastava, R.K.; Ernst, R.E.; Peng, P.; editors). Springer Geology: 111-154. Singapore. https://doi.org/10.1007/978-981-13-1666-1_4

Torró, L.; Villanova, C.; Castillo, M.; Campeny, M.; Gonçalves, A.O.; Melgarejo, J.C. 2012. Niobium and rare earth minerals from the Virulundo carbonatite, Namibe, Angola. Mineralogical Magazine 76 (2): 393-409. https://doi.org/10.1180/minmag.2012.076.2.08

Vasyukova, O.V.; Williams-Jones, A.E. 2022. Carbonatite metasomatism, the key to unlocking the carbonatite-phoscorite-ultramafic rock paradox. Chemical Geology 602: 120888. https://doi.org/10.1016/j.chemgeo.2022.120888

Vasyukova, O.V.; Williams-Jones, A.E. 2023. A new model for the origin of pyrochlore: Evidence from the St Honoré Carbonatite, Canada. Chemical Geology 632: 121549. https://doi.org/10.1016/j.chemgeo.2023.121549

Wallace, M.E.; Green, D.H. 1988. An experimental determination of primary carbonatite magma composition. Nature 335: 343-346. https://doi.org/10.1038/335343a0

Wimmenauer, W. 1962. Beiträge zur Petrographie des Kaiserstuhls. Teil IV: Die gesteine der phonolithischen familie. Teil V: Die subvulkanischen Breccien. Neues Jahrbuch für Mineralogie Abhandlungen 98: 367-415.

Winther, C. 1901. Britholith, ein neues Mineral. Zeitschript für Kristallografie und Mineralogie 34: 685-687.

Woolley, A.R. 2001. Alkaline rocks and carbonatites of the world. Part 3: Africa. The Geological Society of London: 384 p. Bath.

Worley, A.; Cooper, A.F.; Hall, C.E. 1995. Petrogenesis of carbonate-bearing nepheline syenites and carbonatites from Southern Victoria Land, Antarctica: origin of carbon and the effects of calcite-graphite equilibrium. Lithos 35 (3-4): 183-199. https://doi.org/10.1016/0024-4937(94)00050-C

Yang, K.; Fan, H.; Pirajno, F.; Li, X. 2019. The Bayan Obo (China) giant REE accumulation conundrum elucidated by intense magmatic differentiation of carbonatite. Geology 47 (12): 1198-1202. https://doi.org/10.1130/G46674.1

Yaxley, G.M.; Anenburg, M.; Tappe, S.; Decree, S.; Guzmics, T. 2022. Carbonatites: classification, sources, evolution, and emplacement. Annual Review of Earth and Planetary Sciences 50: 261-293. https://doi.org/10.1146/annurev-earth-032320-104243

Ying, J.; Zhou, X.; Zhang, H. 2004. Geochemical and isotopic investigation of the Laiwu-Zibo carbonatites from western Shandong Province, China, and implications for their petrogenesis and enriched mantle source. Lithos 75 (3-4): 413-426. https://doi.org/10.1016/j.lithos.2004.04.037

Zozulya, D.R.; Lyalina, L.M.; Savchenko, Y.E. 2015. Britholite ores of the Sakharjok Zr-Y-REE deposit, Kola Peninsula: Geochemistry, mineralogy, and formation stages. Geochemistry International 53: 892-902. https://doi.org/10.1134/S0016702915080108

Zozulya, D.R.; Lyalina, L.M.; Savchenko, Y.E. 2017. Britholite-group minerals as sensitive indicators of changing fluid composition during pegmatite formation: evidence from the Keivy alkaline province, Kola peninsula, NW Russia. Mineralogy and Petrology 111: 511-522. https://doi.org/10.1007/s00710-017-0493-3