The Napélépéra pluton contains basic xenoliths and is cut by dolerite dykes (Figure 2(a)). Overall, the structure of the pluton is slightly foliated (Figure 2(b)) to massive. The texture is gritty porphyroid in a mineralogical matrix of the order of a millimetre. The main minerals are quartz (20% - 25%), plagioclase (15% - 20%), biotite (30% - 35%) and amphibole (10% - 15%) (Figure 2(c) & Figure 2(d)). Alkali feldspars account for a small proportion (1% - 5%). In general, plagioclase alters to sericite, quartz occurs either in xenomorphic form with rolling extinction characteristic of primary quartz or in recrystallised or cracked form under the pressure-temperature effect of deformation. Biotite is abundant with a slight mineral orientation and amphibole alters to chlorite.

The Napélépéra pluton shows high levels of Al₂O₃ while the sum of alkalis Na₂O + K₂O (4.42% - 5.08%) is low except in the mineralised zone where it is ≈ 6% - 7%. The TAS diagram shows a dioritic composition (Figure 3(a)) for the granodiorite. It is located in the tholeiitic and calc-alkaline series (Figure 3(b)) of the SiO-K₂O diagram except for highly altered samples. It is metaluminous (Figure 3(c)) according to the Fe + Mg + Ti vs. Al-(K + Na + 2Ca) diagram. The

REE pattern (Figure 3(d)) shows a parallelism of spectra that could suggest a cogenetic link between the samples. Microscopic precision reveals that, apart from the variable nature of the alteration, the samples come from the same rock.

The alterations deduced from the multi-scale observations define zones, each with a predominant type of alteration. They form paragenetic associations based mainly on chlorite, carbonate sericite and quartz.

Chloritisation: Chlorite replaces the ferromagnesian phases of granodiorite. It is developed on biotite and amphibole minerals. It occurs as a free mineral around grains of quartz or pyrite (Figure 4(a)) when the alteration is intense and as veinlets associated with sericite (Figure 4(b)).

Chlorite is found in varying proportions throughout the rock, with a predominance in the distal zone, where in places it is the only alteration mineral. The paragenesis chlorite + sericite ± silica is formed in these zones.

Sericitisation: Sericite is the second most common alteration mineral found

in rock. Macroscopically, it appears as fine flakes with a silky sheen. Microscopically, it forms a matrix of small needles overprinting the shapes of the feldspar relics from which it is derived (Figure 4(c)). It is also present in the form of veinlets, associated with chlorite or calcite (Figure 4(d)). Sericite is intense in the proximal mineralised zone where it is associated with carbonate. Where sericite predominates, it is mainly associated with carbonate and secondarily with silica and chlorite. In these zones, the paragenesis sericite + carbonate ± silica ± chlorite is formed.

Silicification: This is associated with quartz or quartz-carbonate veins, invasive silicification breccias (Figure 4(e)), or in pervasive form as very fine or recrystallised quartz grains (Figure 4(f)). Silicification is observed in all the cores described.

Carbonation: Carbonates appear in the form of 1) free grains, 2) veinlets and 3) fine micrometric matrix carbonate grains, 4) intense circulation accompanied by haematisation (Figure 4(g)). Carbonation is frequently only observed in the proximal, mineralised zone. In the mineralised zone, the rock is almost completely replaced by carbonates (Figure 4(h)). Carbonation is often associated with oxidation. The association of carbonate + silica ± sericite is characteristic of these zones.

From the diagram of alterations Na₂O + CaO versus Fe₂O₃ + MgO [18] , the distribution of samples confirms chloritisation and carbonation as the main alterations (Figure 5). Samples N20 and N46 are healthy to slightly chloritised

samples. Sample N49 in the proximal zone is sericitised. Samples N21 and N47 from the mineralised zone have strong carbonate alteration with a high pyrite concentration for N47. The intensity of carbonation increases from the distal zone (N20, N46) to the proximal zone (N51) to the heart of the mineralisation (N21, N47).

Although several types of alteration have been described in the Napélépéra gold-bearing granodiorite, carbonation is the alteration associated with mineralisation. The paragenesis gold + pyrite ± carbonate ± silica ± sericite characterises the gold mineralisation at Napélépéra (Figure 6). This carbonation is the result of fluid input in the shear corridor of the mineralised zone. This alteration is characteristic of intrusive-related gold mineralisation described by [19] and [20] . These alterations have the same characteristics as those described in NW Ghana in the Julie gold deposit [21] [22] and in the Chirano district [23] .

The geochemical balance of the major elements and metals in the mineralised zones (ZM), compared with that in the proximal (ZP) and distal (ZD) nonmineralised zones, shows that the mineralising fluids have induced leaching of chemical elements characterised by ratios of less than 1 and input when these ratios are greater than 1 (Figure 7). The SiO₂ shows a mass balance that is justified by the forms in which the silica is observed, either in the form of invasive silicification that brecciated the granodiorite or in the form of recrystallisation.

The concentrations of Al₂O₃ are similar in samples taken from all levels. Feldspars are the main minerals containing this oxide and, as they weather to form sericite, the Al₂O₃ content did not vary significantly during the weathering process, which was marked by relatively little leaching.

Fe₂O₃ remains balanced between the proximal and distal facies due to the relatively low intensity of hydrothermal alteration. Its mass balance compared with the mineralised zone shows an enrichment due to the oxidation of pyrite and the formation of haematite minerals.

CaO is depleted in the proximal zone due to the alteration of the plagioclase to sericite. The contribution of mineralising fluid has favoured the carbonation process characteristic of mineralisation and is shown by a contribution in the mass balance compared between ZM and ZP.

MgO is depleted in both the proximal and mineralised zones. This depletion is due to the leaching of Mg²⁺ ions during the alteration of ferromagnesian minerals (biotite and amphibole).

Na₂O: sodium oxide is the opposite of the other oxides analysed in all the samples. It is characterised by a slightly constant content in the healthy rock, with depletion in the proximal zones due to alteration of the orthoclase and microcline. An enrichment in the mineralised zone is characteristic of the contribution of mineralising fluids.

K₂O: the ZP/ZD ratio shows enrichment in potassium oxides due to the sericitisation of orthoclase and microcline and later leached by mineralising fluids through concentration.

Cr₂O₃ and TiO₂ remain virtually immobile during the hydrothermal alteration process.

MnO and SrO are enriched by carbonation.

The mass balance of comparative metals in the proximal and distal zones shows the absence of metals associated with sericitisation and chloritisation. Metal enrichment is the result of the circulation of mineralising fluid. As, Hg, Ag and Bi were enriched 8 to 650 times, depending on the metal, from the distal zone to the mineralised zone (Figure 7). The high content of Na₂O and SrO in the Napélépéra mineralisation could be explained by an influx of fluid rich in these metals. It forms a peak in the mineralised zone, as do metals and nonmetals such as arsenic (As), tellurium (Te), silver (Ag), mercury (Hg), bismuth (Bi) and selenium (Se). This phenomenon would justify these elements having

the same source of enrichment.