Abstract

This study presents a comprehensive geochemical analysis of alkaline lavas from the Nkondjock area along the Cameroon Volcanic Line (CVL). Employing thin-section microscopy, ICP-AES (major elements), and ICP-MS (trace/REE elements), we characterize basanites (plagioclase, olivine, augite, opaque minerals and accessory spinel), hawaiites (plagioclase, olivine, augite and opaque minerals), mugearites (plagioclase, augite, aegirine, hornblende, opaque minerals and olivine), and benmoreites (plagioclase, augite, nepheline, sanidine, opaque minerals and accessory olivine). These lavas exhibit microlitic to porphyritic textures, indicating multi-stage crystallization: initial megacryst formation in magma chambers, phenocryst development in conduits, and rapid surface cooling. Geochemically, they are MgO-depleted (2.13 - 9.01 wt%) and Al₂O₃-enriched, with primitive mantle-normalized patterns showing LILE/LREE enrichment and OIB-like signatures. Crustal contamination is negligible, except in evolved samples (mugearite, benmoreite). Trace element ratios (e.g., La/Ta = 14.91 - 21.5; ΔNb > 0) confirm an asthenospheric garnet-lherzolite source, with low-degree partial melting (0.5% - 2%) at 70 - 80 km depth. Zr/Y (9.28 - 16.53) and Ti/Y (260 - 827) ratios classify them as intraplate basalts, consistent with reactivated Pan-African fault magmatism.

Keywords

Alkaline Lavas, Nkondjock, Cameroon Volcanic Line, OIB-Like Signatures, Asthenospheric Garnet-Lherzolite Source

Share and Cite:

1. Introduction

The Cameroon Volcanic Line (CVL), a major tectono-magmatic mega-structure trending N30˚E, extends over 1600 km in length and ~100 km in width (Tchoua, 1974; Njome & De Wit, 2014). Comprising both oceanic and continental segments, it represents a globally unique feature in Africa and worldwide (Déruelle et al., 1991, 2007). The oceanic sector (Atlantic) includes four volcanic islands, Pagalù, São Tomé, Príncipe, and Bioko, aligned within the Gulf of Guinea, along with two large seamounts between Bioko-Príncipe and Príncipe-São Tomé (Burke, 2001). The continental sector spans from Mount Cameroon to Lake Chad and hosts multiple volcanic edifices ranging in age from the Eocene to Recent (Fitton, 1987; Itiga et al., 2004). Mafic and felsic lavas occur across all CVL centers: Mounts Cameroon, Manengouba, Bambouto, Bamenda, and Mandara (Aka et al., 2004, 2018), except Mount Etindé, which consists exclusively of nephelinites, leucitites, and haüynophyres (Nkoumbou et al., 1995). While tectonically active since the Precambrian, magmatism only initiated at the end of the Cretaceous with the emplacement of ring complexes (Fosso et al., 2005).

The CVL’s origin remains debated within the global scientific community, particularly in Cameroon. Some authors attribute it to the reactivation of Precambrian faults driving rifting (Fitton, 1980; Moreau et al., 1994), while others invoke a mantle thermal anomaly (“hot spot”) (Meyers et al., 1998; Déruelle et al., 2007). Marzoli et al. (1999) contested Fitton’s unified hot-spot model (Fitton & Dunlop, 1985; Fitton, 1987) using ⁴⁰Ar/³⁹Ar dating. Despite extensive research, the Nkondjock locality, underlain by significant volcanic formations, remains unexplored along the CVL. This article presents novel geochemical data from Nkondjock (Lat. 4˚40' - 5˚01'N, Long. 10˚08' - 10˚24'E; Figure 1), situated between Mount Bana and the Noun Plain. Underlain by Pan-African granites/gneisses (Kamguia Kamani et al., 2021; Azefack Mbounou et al., 2023), the area exposes mafic-intermediate volcanic rocks but lacks felsic types (Tengomo et al., 2026). By analyzing these samples, we constrain magma sources and evolution processes, directly addressing the CVL’s disputed origin and refining regional tectonic-petrogenetic models.

2. Methods

Following field sampling, laboratory analyses comprised thin section preparation, microscopic description, and whole-rock geochemical analyses. Thin sections were prepared using standard protocols and examined under a polarizing microscope. Observations in both plane-polarized and cross-polarized light enabled the identification of mineral assemblages, textures, parageneses, and deformation microstructures based on classical optical criteria.

Whole-rock geochemical analyses targeted major elements, trace elements, and rare earth elements (REEs). Selected samples were powdered after removal of altered portions, controlled crushing, and milling. Major elements were analyzed by Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES), while trace elements and REEs were determined via Inductively Coupled Plasma Mass Spectrometry (ICP-MS) following lithium borate fusion. Major element concentrations are expressed as weight percent oxides; trace elements and REEs are reported in parts per million (ppm). Loss on ignition (LOI) was determined by calcination at 1000˚C to assess volatile content and the degree of sample alteration.

Figure 1. Location of the Nkondjock area (red star) along the Cameroon volcanic line (base map is from Njonfang et al., 2011). Locations of seamounts according to Burke (2001).

3. Results

3.1. Rock Classification

Volatile-free normalized major element compositions were used for classification. Preliminary analysis of Nkonjock volcanic rocks on a total alkalis vs. silica diagram (Le Bas et al., 1986) identifies mafic samples as alkaline, specifically comprising basanites, hawaiites, mugearites, and benmoreites (Figure 2(a)). This alkaline suite is further classified as potassic based on Figure 2(b).

3.2. Petrography

The study reveals a diversity of alkaline lavas: basanites, hawaiites, mugearites, and benmoreites, representing distinct stages of magmatic differentiation within the study area. All these lavas generally exhibit microlitic textures (ranging from aphyric to fluidal variants, Figure 3(a)) and porphyritic textures (Figure 3(b)).

Figure 2. Classification of the Nkondjock lavas. (a) Position of the lavas on the TAS diagram of Le Bas et al. (1986). The bold line represents the Irvine-Barragar boundary (Irvine & Baragar, 1971) separating the alkaline series from the sub-alkaline series. (b) Position of the lavas on the Na₂O versus K₂O diagram of Le Maitre (2002).

3.2.1. Basanites

Basanites outcrop primarily as prismatic flows, domes, and stream-bed flows. Prismatic flows (often hexagonal) exhibit diameters of 10 - 30 cm. These rocks are mostly aphyric at the outcrop scale, occasionally semi-crystalline with visible olivine and pyroxene crystals.

In thin section, basanites display aphyric to porphyritic microlitic textures, sometimes fluidal. They comprise plagioclase (30% - 35%), olivine (15% - 20%), augite (10% - 15%), opaque minerals (5% - 10%), and accessory spinel. The mesostasis is cryptocrystalline, consisting of plagioclase microlites, olivine microcrystals, opaques, and volcanic glass. Olivines are frequently altered to iddingsite, while augites show well-developed cleavage and oblique extinction. Spinels occur as octahedral phenocrysts, occasionally surrounded by reaction rims (Figure 3(h)).

Figure 3. Microphotography of Nkondjock lavas. (a) Microlitic texture. (b) Porphyritic texture. (c) Olivine iddingsitization. (d) Augite crystal rimmed by opaques. (e) Zoned plagioclase. (f) Alignment of plagioclase microlites. (g) Opaque minerals as inclusions in plagioclase, augite, and olivine. (h) Spinel surrounded by reaction rim.

3.2.2. Hawaiites

Hawaiites form vegetation-covered domes. Macroscopically, they contain plagioclase and pyroxene phenocrysts. Thin sections reveal dominantly porphyritic microlitic textures with fluidal structure marked by the preferential alignment of plagioclase microlites (Figure 3(f)). Mineralogy is dominated by plagioclase (45% - 50%), followed by olivine (5% - 10%), augite (5% - 10%), and opaque minerals (1% - 5%). The cryptocrystalline mesostasis is rich in plagioclase microlites, olivine/augite microcrystals, opaques, and volcanic glass. Olivines locally exhibit iddingsitization (Figure 3(c)); augites are euhedral, non-pleochroic, and display moderate oblique extinction.

3.2.3. Mugearites

Mugearites outcrop as extensive domes (partly concealed by vegetation), locally showing incipient prismation. Samples exhibit a semi-crystalline structure.

Microscopically, they feature porphyritic to aphyric microlitic textures with pronounced fluidal structure. Composition includes plagioclase (40% - 45%), augite (10% - 15%), aegirine (2% - 5%), basaltic hornblende (~3%), opaque minerals (1% - 5%), and rare olivine (~2%). Zoned plagioclases dominate (Figure 3(e)), forming local glomeroporphyritic clusters. Augites are euhedral with high oblique extinction, often associated with opaques and aegirines. Opaque minerals are often found as inclusions in plagioclase, augite, and olivine (Figure 3(g)). The mesostasis is rich in plagioclase microlites, mafic microcrystals, and volcanic glass.

3.2.4. Benmoreites

Benmoreites occur as domes and isolated blocks. Macroscopically, they contain visible pyroxene crystals. Thin sections show a porphyritic microlitic texture. The mineralogy is dominated by plagioclase (~55%), with augite (10% - 15%), nepheline (1% - 5%), sanidine (1% - 3%), opaque minerals (1% - 5%), and accessory olivine (1% - 2%). Plagioclases appear as megacrysts to microlites, often corroded by mesostasis. Augites are euhedral with distinct oblique extinction, commonly rimmed by opaques (Figure 3(d)). The coexistence of sanidine and nepheline attests to the evolved alkaline character of these lavas.

3.3. Whole-Rock Major- and Trace-Element Geochemistry

Major- and trace-element compositions of the studied samples are presented in Table 1 and Table 2. The Nkonjock alkaline suite exhibits SiO₂ contents ranging from 39.6 to 56 wt% and MgO from 9.01 to 2.13 wt%. Benmoreite NJ45 has the highest SiO₂ and lowest MgO, whereas basanite NG60 shows the lowest SiO₂ and highest MgO (Figure 4). All basanites contain ≤45.8 wt% SiO₂, while mugearites range between 51 - 52.5 wt%.

Table 1. Whole-rock major element analysis and CIPW norms of Nkondjock lavas.

Table 2. Whole-rock trace element and REE analysis of Nkondjock lavas.

Figure 4. Major element variations for Nkondjock lavas against their Mg#.

Compared to lavas from the Cameroon Volcanic Line (CVL; Kagou et al., 2010; Tchuimegnie Ngongang et al., 2014; Mulimbi Kagarabi et al., 2025), Nkonjock lavas are depleted in MgO, enriched in Al₂O₃, and comparable in SiO₂, Na₂O, and K₂O. Normative mineralogy reveals nepheline in some samples and ubiquitous olivine, with hypersthene present only in benmoreite.

Primitive mantle-normalized (Sun & McDonough, 1989) trace element spider diagrams and REE patterns indicate enrichment in LILEs and LREEs relative to HFSEs and HREEs (Figure 5). This is evidenced by high (La/Yb) ratios: basanites (16.96 - 21.14), hawaiite (15.69), mugearites (16.52 - 18.47), and benmoreite (20.20). Patterns are parallel to Ocean Island Basalts (OIBs; Figure 5). Most samples exhibit slight positive Ba, Nb, La, and Zr anomalies, significant K and Ti anomalies (Figure 6(b)), and no Eu anomaly.

Figure 5. (a) Chondrite and (b) Primitive Mantle normalized spider diagrams of the Nkondjock lavas; normalization values are from Sun and McDonough (1989).

Figure 6. Studied lavas in the Zr/Y vs. Nb/Y diagram, relative to the mantle compositional components (circles) and fields of rocks from different tectonic settings after Condie, 2005.

4. Discussion

4.1. Petrography

The studied rocks generally exhibit microlitic textures (ranging from aphyric to fluidal variants) and porphyritic textures, with some samples containing plagioclase megacrysts. This suggests three distinct crystallization phases: formation of plagioclase megacrysts and opaque minerals in the magma chamber, development of phenocrysts/microphenocrysts of plagioclase, augite, olivine, and aegirine in the volcanic conduit, and rapid surface cooling leading to mesostasis formation. Corrosion embayments or growth gaps in plagioclase and olivine phenocrysts likely resulted from late-stage crystallization in residual melt during magma ascent. Opaque mineral rims around augite crystals (Figure 3(d)) indicate thermodynamic disequilibrium, attributed to rapid cooling or magmatic fluctuations. Plagioclase crystals preserve extensive magmatic histories; chemical zoning and mineral inclusions (Figure 3(e)) reflect evolving host-magma chemistry during growth, linked to shifting crystallization conditions. Normal zoning (anorthitic cores → albitic rims; Lindsey, 1966) implies constant-pressure cooling. Plagioclase alignment (Figure 3(f)) may indicate surface lava flow direction. Inclusions of opaque minerals and olivine in augite and ubiquitous opaque inclusions in plagioclase, olivine, and augite (Figure 3(g)) support fractional crystallization. Spinel phenocrysts (Figure 3(h)) suggest a mantle source. The textures and mineralogy reveal a multi-stage cooling history, emphasizing fractional crystallization and disequilibrium conditions. The absence of quartz/hypersthene (except in benmoreite) and spinel-olivine associations points to mantle-derived melts with limited shallow-crust interaction.

4.2. Petrogenesis

Provenance analysis for Nkondjock lavas considers ΔNb values: Mantle-derived lavas typically have ΔNb > 0. Here, ΔNb = 0.04 - 0.36 (Table 2), except for mugearite NO62 and benmoreite NJ45 (ΔNb = −0.07 and −0.22). In Nb/Y vs. Zr/Y plots (Weaver, 1991; Condie, 2005; Figure 6), most samples cluster in the mantle array (above ΔNb = 0), while NO62 and NJ45 plot below, indicating divergent behavior attributed to mantle source heterogeneities.

Vanadium (V) decreases systematically from basanites to benmoreites, confirming fractional crystallization as the primary differentiation mechanism. Steady Na₂O enrichment (Figure 4(g)) rules out significant plagioclase fractionation; instead, olivine, clinopyroxene, and Fe-Ti oxides dominated. Similar trends occur in Bafang, Fotouni, and Noun basalts. Uniform Nb/Th, Nb/Y, Zr/Nb, and Zr/Y ratios, consistent across the Cameroon Volcanic Line (CVL; e.g., Fitton, 1987; Kamgang et al., 2008), support a mantle origin.

4.3. Mantle Source Characteristics and Partial Melting

La/Ta ratios distinguish asthenospheric (≤22) from lithospheric (22 - 30) or contaminated (>30) sources. Nkondjock lavas yield La/Ta = 14.91 - 21.5, favoring an asthenospheric origin. Ce/Yb vs. La/Ta plots (Figure 7) further indicate a garnet-bearing asthenospheric mantle. Trace elements show high La/Yb (21.87 - 29.47) and Zr/Y (9.85 - 16.53) with near-constant Y (30.8 - 35.2 ppm) and Yb (2.42 - 3.27 ppm), typifying garnet-dominated sources. Garnet stability (>3 GPa; ~70 - 80 km depth; Robinson & Wood, 1998) implies deep melting. Ce/Y vs. Zr/Nb (Figure 8) and La/Yb vs. Zr/Nb (Figure 9) diagrams align samples with primitive garnet peridotite (GP), suggesting 0.5% - 2% partial melting of garnet lherzolite (2% - 4% garnet), consistent with Bafang, Fotouni, Mt. Cameroon, and Manengouba lavas.

Figure 7. Ce/Yb vs. La/Ta for the source of the Nkondjock lavas after Dorais, Marvinney, and Markert (2017).

Figure 8. Ce/Y vs. Zr/Nb for the source of the Nkondjock lavas after de Hardarson and Fitton (1991). The bold curves represent non-modal fractional melting for four mantle compositions: GD: depleted garnet lherzolite; GP: primitive garnet lherzolite; SD: depleted spinel lherzolite; SP: primitive spinel lherzolite. The numbers on the lines correspond to the melting percentages.

Figure 9. Modeled melting (Zr/Nb vs. La/Yb) result for the CVL mafic rocks with MgO > 4. Melts that produced most basanites and alkali basalts were generated by <3% partial melting of a dominantly garnet (<6%) bearing mantle lherzolite.

4.4. Crustal Contamination

Key contamination indicators (e.g., basement xenoliths, normative quartz/hypersthene) are absent, except in benmoreite. Elevated Ba (positive multi-element anomaly) hints at assimilation, but La/Nb (0.70 - 1.15) and La/Ta (13.37 - 21.5) contradict the criteria for crustal input (La/Nb > 1.5, La/Ta > 22; Hart et al., 1989). Negative Nb/Zr anomalies typical of contamination (Cai et al., 2010) are also lacking. Thus, crustal effects were negligible. This is confirmed by the Rb/Y vs. Nb/Y diagram (Figure 10) after Cox and Hawkesworth (1985). In this diagram, the studied lavas fall into the uncontaminated lava field.

Figure 10. Nkondjock lavas in the Rb/Y vs. Nb/Y diagram after Cox and Hawkesworth (1985).

4.5. Geotectonic Context

Immobile elements (Ti, Y, Zr) constrain tectonic settings. High Ti/Y (260.23 - 827.36) and Zr/Y (9.28 - 16.53) ratios, coupled with positive Nb anomalies, characterize intraplate magmatism (Pearce & Cann, 1973). Zr/Ba ratios (0.41 - 0.70; Ormerod et al., 1988, 1991) exceed the subduction-related threshold (<0.2), confirming an intraplate affinity. Discrimination diagram (Zr/Y vs. Zr, Figure 11) consistently plots samples within within-plate basalt (WPB) fields. The high enrichment in incompatible elements (Rb, Ba, Nb, Th, and Zr), along with the REE spectra enriched in LREE, without marked Eu anomalies, indicates low-grade partial melting of an EM1 or EM2 enriched mantle, consistent with an OIB signature (Sun & McDonough, 1989). This context is reinforced by predominantly positive ΔNb values, thus ruling out a subduction environment (Pearce & Cann, 1973). Taken together, these data therefore suggest intraplate alkali volcanism.

Figure 11. Nkondjock lavas in the tectonic evolution discriminant diagram Zr/Y vs. Zr after Pearce and Norry (1979). WPB = Intraplate Basalts, VAB = Volcanic Arc Basalts, MORB = Mid-Ocean Ridge Basalts.

This aligns with CVL models invoking intraplate alkaline magmatism along reactivated Pan-African faults (Njonfang et al., 2013).

5. Conclusion

The Nkondjock lavas originate from low-degree melting (0.5% - 2%) of garnet-bearing asthenospheric mantle at 70 - 80 km depth, fractionating via olivine, clinopyroxene, and Fe-Ti oxides during ascent. Their intraplate affinity; evidenced by high Zr/Y (9.28 - 16.53), positive Nb anomalies, and classification within within-plate basalt fields; provides robust support for the fault reactivation model over the mantle plume hypothesis for CVL genesis. This geochemical signature aligns unambiguously with magmatism driven by reactivated Pan-African structures, as indicated by:

• Trace element ratios (e.g., ΔNb > 0, La/Ta = 14.91 - 21.5) confirming asthenospheric sourcing without plume-like thermal anomalies.

• Tectonic discrimination (Zr/Y vs. Zr) excluding mantle plume affinities and instead correlating with rift-related fault systems. Petrographic textures further record disequilibrium crystallization during multi-stage ascent, while minimal crustal assimilation (La/Nb < 1.5) underscores mantle-dominated processes. Collectively, these findings position the CVL as a tectonically controlled rift province, with Nkondjock offering critical insights into deep melting dynamics at reactivated lithospheric discontinuities.

Conflicts of Interest

The authors declare no conflicts of interest regarding the publication of this paper.

References