Magma chasing hydrothermal fluids in distal skarns at Campiglia Marittima (Italy)

Magma chasing hydrothermal fluids in distal skarns at Campiglia Marittima (Italy)

Abstract

Most carbonate-hosted skarn deposits show a direct spatial relationship with magmatic intrusions and an intimate relationship with porphyry copper deposits. Their primary mineralogical and geochemical features indicate metasomatism by high-temperature magmatic fluids. Conversely, many distal Pb-Zn(-Ag) skarns pose an important challenge because there is no apparent association with intrusions, and external, meteoric and/or basinal fluids should become dominant with increasing distance from the igneous source.

Nevertheless recent investigation of distal skarns indicates that ore-forming fluids match the composition of proximal magmatic fluids in granitoid-related mineral deposits. Here we present exceptional evidence from a distal Pb-Zn(-Ag) skarn deposit at Campiglia Marittima, Italy, where a multipulse magmatic system, after having released metasomatic fluids, spewed out a mafic magma pulse that chased the newly formed skarns, invading large primary pockets. Underground mapping and petrologic data indicate that the intrusion of the mafic magma produced prograde back-reactions in the skarn silicates, precipitation of a new Cu-Fe ore assemblage and remobilization of the Pb-Zn(-Ag) ore. We document a serendipitous occurrence where the spatial gap between the magmatic emplacement level (multiple intrusions) and the overlying distal hydrothermal zone was bridged by a pulse of mafic magma escaping the main magmatic trap and producing reverse telescoping effects.

Introduction

Skarn deposits commonly results from the metasomatic alteration of a rock, usually

carbonate-rich, by infiltration of hydrothermal fluids

. Most skarns have an intimate spatial

relationship with magmatic intrusions (exo- and endoskarn) but in certain circumstances,

hydrothermal fluids may migrate considerable distances to produce distal skarns well outside

contact aureoles, without any obvious link with magmatic intrusions

. While proximal

skarns display geological, geochemical and isotopic features consistent with an origin by

predominant magmatic fluids with a late involvement of meteoric fluids

, distal skarn deposits

have been mainly ascribed to the circulation of external fluids (meteoric, basinal) variably

combined with magmatic fluids

. Recent studies indicate that, despite the distal nature of

carbonate replacement deposits, the ore fluids were dominantly magmatic, and that there was

little or no contribution from external fluids

.

Many skarn deposits are strictly associated with porphyry deposits in subduction-related magmatic arc settings, sharing many geochemical and alteration features

. Besides that metal producer (Pb, Zn, Ag, Cu, Fe, W, Sn), skarn deposits can be key to understand fluid dynamics at the periphery of magmatic-hydrothermal systems, unraveling pathways to hidden ore deposits (e.g., porphyry copper). In this scenario, research on distal skarns opens new perspectives for deep exploration of blind ore deposits and unconventional geothermal resources (supercritical fluids), because they should represent the outer limit of the hydrothermal system dominated by magmatic aqueous fluids.

Tuscany host several Fe, Pb-Zn-Ag and Pb-Zn-Ag-Cu-Fe skarn deposits (e.g. Elba Island, Campiglia Marittima, Massa Marittima)

that are spatially and temporally related to Miocene-Quaternary bimodal magmatism, dominated by crustal peraluminous granites with minor mantle-derived lamproites, shoshonites, K-andesites and latites (Fig. 1). They constitute the so-called Tuscan Magmatic Province (TMP), a diachronous sequence of magmatic centres triggered by post-collisional extension, as the Northern Apennine chain was progressively thinned and heated by asthenosphere upwelling during the eastward roll-back of the subducting Adriatic Plate

. All the magmatic centres in Tuscany are distributed along NE-trending lineaments and developed as a wave, moving northeastward across the region, suggesting that magmatism was focused by transfer zones that reactivated former transversal faults

. Most of these igneous centres were ephemeral, resulting in the emplacement of a single, small magmatic body. In contrast, multiple batches of magmas (felsic and mafic) sequentially emplaced in three discrete areas - Elba Island (8.5-6.9 Ma), Campiglia Marittima (5.7-4.3 Ma) and Larderello (3.8-1.3 Ma) – that are aligned along a main NE-trending structure (Fig. 1).

The Campiglia magmatic system is made by a major peraluminous monzogranite pluton

(5.7 Ma), a subvolcanic complex (one mafic andesitic porphyry and two felsic rhyolitic

porphyry dike swarms; ca. 4.4 Ma) and a peraluminous rhyolite volcanic complex (4.3-4.4

Ma). The Monzogranite pluton intruded a thick reef limestone sequence (Triassic-Jurassic) at

ca. 5 km depth, producing widespread contact effect responsible for the formation of pure

slightly foliated marbles: the main host rock for skarn deposits in the area. In this paper, we

report on the Campiglia Pb-Zn-Ag-Cu-Fe skarn deposits (Temperino and Lanzi mines, Italy)

that are considered the first well-described skarns in scientific literature

. Since XIX century,

they have been adopted as a classical example of carbonate-hosted exoskarn with a clear

relationship between a mafic porphyry dyke and its mineralogically zoned, metasomatic envelope: porphyry magnetite ilvaite clinopyroxene marble

.

In spite of this consolidated interpretation, new underground mapping at Temperino and Lanzi mines (25 km of tunnels and shafts) and petrological investigation on samples and drill cores (ca. 19 km of exploration drill holes) allowed us to define a completely new geological scenario (information of the geological setting of Campiglia area is in the Supplementary Information). 3D reconstruction of skarn bodies (sub-vertical tabular sigmoidal bodies) indicates that neither axial mafic porphyry dike, nor systematic mineralogical zoning occurs.

Most of the deposits are constituted by simple distal Pb-Zn(-Ag) hedenbergite-ilvaite skarns, where sulphides overprint the skarn silicates (e.g. Lanzi mine). Only at Temperino mine, two skarn bodies have been later intruded by mafic magma, producing several small dykes and filling large primary pockets in the skarn. Using field and petrographic relationships among mafic porphyry and skarn silicates/sulfides, we show that ilvaite crystals/masses, originally lining the pockets, were heated by the magma and replaced by a magnetite + hedenbergite assemblage, meanwhile Mg-rich overgrowths developed on earlier hedenbergite. Fluid associated with mafic magma also produced a new Mg-rich actinolite-chlorite Cu-Fe ore overprinting the ilvaite-hedenbergite skarn, and remobilized the earlier Pb-Zn(-Ag) sulfides, producing a re-worked Zn-Pb-Cu(-Ag) ore type (see

). Stable and radiogenic isotope data (O, H and Pb) suggest a dominant magmatic origin of skarn forming fluids, and distinct magmatic sources (granite and mafic magmas) for base metals in the two main ore types. The complex geo-mineralogical and geochemical pattern of Campiglia skarns represents a serendipitous history where magma-fluid uncoupling, typical of distal skarn, was bridged by a pulse of mafic magma escaping the deeper magmatic system and producing an odd reverse telescoping effect. In recent years, application of advanced microanalytical techniques allowed to unravel the magmatic origin of fluids involved in the formation of distal skarns

; at Campiglia, for the first time, we were able to record, in the field, a direct link between intrusive magmatic systems and distal skarn deposits.

Primary skarn pockets: macroporosity for fluids and magmas

Many hydrothermal rocks (veins, lodes, replacement bodies, etc.), during the very late

stage of their crystallization, develop cavities (from µm

to several m

) where residual

aqueous fluids concentrate. Such open spaces usually act as closed systems allowing quiet

crystallization of large and idiomorphic crystals. Most of the exceptional mineral specimens

exposed in mineralogical museums, showing beautiful assemblages of silicate-oxide-sulfide crystals, where collected from such pockets. Pockets are the last “breath” of a hydrothermal system and therefore even the latest product of fluids possibly released by a magmatic intrusion.

Skarn deposits replacing pure limestones/marbles are tipically coarse-grained and display large residual pockets, up to several m

, lined by well-formed crystals of skarn silicates (hedenbergite, garnet, ilvaite; Fig. 2a) and gangue/ore minerals (quartz, carbonates, sulfides, etc)

. The pockets have highly variable shape: spheroidal, tabular, tubular, multi-convex, reflecting the textural pattern of the skarn bodies and the original fracture pattern exploited by metasomatic fluids. The most spectacular primary pockets are represented by the multi- inward-convex voids resulting from competitive growth of fibrous-radiating spheroids of hedenbergite (e.g. Dalnegorsk, Russia

; Elba Island

and Campiglia, Italy

). Pockets constitute the main primary macro-porosity of these skarn, but usually they are largely not interconnected and theoretically ineffective for later fluid flow. Nonetheless, late skarn fracturing usually produce a secondary porosity connecting the pockets and focusing lower temperature external fluids (meteoric, basinal) that are responsible for the retrograde and/or supergene alteration of skarns

.

Skarn deposits at Campiglia are famous due to the finding of huge pockets (up to several m

, Fig 3) that provided world-class specimens of ilvaite crystals (up to 8 cm in length; Fig.

2) sometimes protruding from druses of quartz crystals (up to 50 cm in length)

. Distribution of primary pockets follows the geometry of skarn bodies (vertical sigmoidal bodies; see

) where they are concentrated in the internal portions, and frequently clusterized along vertical planes (ranging from 10 to 20 % of the total skarn volume).

These pockets have been found at both Temperino and Lanzi mines, but 3D exposures in the underground stopes of the former mine evidenced an exceptional and peculiar feature:

many large primary pockets at Temperino are completely filled by the andesitic mafic

porphyry (Fig. 3b). The mafic porphyry (with olivine, clinopyroxene, Mg-rich biotite,

plagioclase, chromite and resorbed xenocrysts of quartz, sanidine and Fe-rich biotite) occurs

only into the skarn bodies, as small dykes crosscutting the skarn structures/textures (0.3-2 m

thick and up to 30 m in length), and irregular masses filling pockets and embedding

idiomorphic crystals of ilvaite (Fig. 3 and 4). The marble-skarn contacts as well as fractures

in the marble host-rock were not preferentially exploited by the rising magma. Deep drilling

below the main skarn bodies intersected narrow structures with both skarn and andesitic mafic

porphyry that could represent feeders. The new field and petrographic observations indicate that the mafic porphyry does not constitute a single axial dyke in skarn bodies and it is not the direct causative magmatic rock for the metasomatic rock. Thus, Campiglia skarn deposits cannot be everymore considered as a classical example of proximal exoskarn deposit as reported in previous papers (e.g.,

).

Calculation made using geological maps at different mining levels indicates that, on average, mafic porphyry represents ca. 15 vol% of the Temperino skarn bodies. The porphyry/skarn volume ratio and the average size of porphyry masses increase with depth where, at the lower end of the skarn bodies, it can reach 50 vol%. Here, the mafic porphyry is mainly hosted by huge, primary pockets that originally provided a highly interconnected porosity (Fig. 3b). We propose that the basal pocket zone, filled by mafic magma, developed enough overpressure to break the skarn and propagate through the vertical pocket planes up to the upper end of the skarn bodies. It is worth to note that magma reached also the extreme ends of subhorizontal skarn mantos that locally propagate from the main bodies, never crossing the marble host.

Thus, pockets represented a significant primary macro porosity of the skarns, more efficient than either skarn-marble contacts or marble fractures, and able to efficiently drain magma from the same feeder zones previously exploited by the skarn forming fluids.

Prograde back-reactions and Cu-Fe overprinting in Zn-Pb(-Ag) distal skarns

Most of the skarn bodies at Campiglia (about 20 in an area of ca. 15 km

), including the Lanzi mine and portions of skarn bodies at Temperino mine, host Zn-Pb(-Ag) ores only and are not directly associated with magmatic rocks

. Skarn mineralogy (hedenbergite, johannsenite, and ilvaite) and textures can slightly change, but they are all characterized by pipe-like ore shoots made by fine-grained aggregates of galena and yellow-green to orange- brown, Fe-poor sphalerite. They display the typical features of distal Zn-Pb skarn deposits including a slightly variable composition with Pb/Zn ratios ranging between 0.4 and 0.8 (0.56 in average), and negligible amount of copper.

The homogeneity of the district is interrupted in correspondence of two pronounced

magnetic anomalies observed at Temperino mine, where the skarn bodies were invaded by the

mafic magma. Here, we document three main effects caused by the mafic magma on skarns: i)

formation, at the contact, of a narrow reaction band (up to 10 cm thick) of cryptocristalline

magnetite + hedenbergite after ilvaite; ii) development of Mg-rich overgrowths on earlier

hedenbergite; iii) formation of Cu-Fe ore with actinolite and Mg- and Cr-rich chlorite that overprints both the magnetite reaction bands and the Zn-Pb(-Ag) skarns.

The first reaction in order of time is the replacement of ilvaite by a microgranular intergrowth of magnetite + hedenbergite (Fig. 4). At contact with mafic porphyry the substitution is almost complete while, at increasing distance, the new assemblage forms an irregular network embedding ilvaite relics. In many cases, the pockets were already lined by druses of large idiomorphic crystals of ilvaite, and the magma flowed between crystals filling any interstices. Centimetric ilvaite crystals were detached by the magma flow and replaced by magnetite + hedenbergite; now they are found as floating crystals in the porphyritic rock. The observed reaction is coherent with an heating process triggered by the mafic magma, in accordance with experimental studies on ilvaite

, showing that the reaction:

6 ilvaite + ½ O2 → 6 hedenbergite + 4 magnetite + 3 H

O

develops at T > 450 °C in a wide range of f(O2). The stoichiometry of the reaction requires the formation of 30.5 vol% magnetite and 69.5 vol% hedenbergite. Image analysis of SEM- BSE images of the natural assemblage from Campiglia provided similar data, 28 ± 3 vol%

and 72 ± 3 vol% respectively.

In parallel with ilvaite destabilization, hedenbergite experienced the formation of Mg-rich overgrowths on crystals from small skarn cavities. Published microprobe analysis of clinopyroxenes from Campiglia skarns indicate a large compositional variability from iron- rich to strongly manganesiferous terms

. This variability represents local, minor situations, while large volumes of skarn at Temperino and Lanzi mines contain hedenbergite with a quite constant Fe-rich composition. The Mg-rich overgrowths observed near the contact with the mafic porphyry define a completely different chemical trend, unusual for distal Zn-Pb(-Ag) skarns, and pointing towards the diopsidic component (up to 50% Di) (Fig. 5).

Finally, the Cu-Fe ore invariably overprinted the skarn at the contact with the mafic

porphyry masses and dykelets. Partial to total replacement and pervasive veining of both

skarn and magnetite reaction bands occurred around the mafic porphyry bodies. High- to

medium-grade Cu-Fe ore (up to 10 wt% Cu) made by chalcopyrite, pyrrhotite, pyrite and

magnetite extends in the nearby skarn volumes for several meters/tens of meters. Ilvaite was

mainly replaced by a Cu-Fe sulfide assemblage (Fig. 2), while hedenbergite was transformed

in actinolite + Mg-rich chlorite, with disseminations and veins of Cu-Fe sulfides. In some

cases, the Cu-Fe ore forming fluids reached small empty pockets producing unusual

pseudomorphosis of Cu-Fe sulphides after large idiomorphic crystals of ilvaite (Fig. 2). The

preferential replacement of ilvaite by Cu-Fe sulphides is documented also by beautiful comb structures where ilvaite was progressively replaced by sulphides while hedenbergite was hydrathed to the actinolite-chlorite assemblage (Fig. 3C, 6). Electron Microprobe analyses indicate that actinolite contain up to 23 wt% MgO and 2.5 wt% Al

O

; chlorite shows variable composition but all the analysed aggregates are classifiable as clinochlore with a significant content in Cr (up to 0.75 wt% Cr

O

). It is worth to note that also the andesitic mafic porphyry underwent a pervasive hydrothermal alteration with the formation of a clinochlore, actinolite, epidote, calcite assemblage. Magnesium, chromium and aluminium were not present in the primary skarn assemblage (Fe-Ca-Mn silicates) and they had to be introduced by the Cu-Fe ore forming fluids associated with the intrusion of the mafic porphyry, that is enriched in magnesium and chromium (ca. 6 wt% MgO, 300 ppm Cr) and contain easily alterable olivine. Such unusual textures and mineral parageneses can be considered as proofs for the causative relationships between the mafic porphyry and the Cu-Fe ores. The emplacement of mafic magma (and associated hydrothermal fluids) into the distal Zn-Pb skarns, beside activating prograde back-reactions, triggered the overprinting of high- temperature Fe-Cu sulfide ore on lower-temperature Zn-Pb sulfide assemblage. The addition of “exotic” components (Cu, Mg, Cr) was accompanied by the local re-working of earlier Zn- Pb ores with formation of a polymetallic Zn-Pb-Cu(-Ag) assemblage, as documented by field observations, skarn-ore textural observations, drill core chemical data and bulk-ore-grade ratios (

).

Isotopic evidences (O, H, Pb) for multiple magmatic sources

Origin of fluids responsible for distal skarns is still debated, although some recent fluid

inclusion studies indicate a dominant magmatic origin

. A common characteristic of

almost all skarns is that the replacement of the

O-enriched carbonate host rocks by skarn

silicates produce rocks with δ

O values in near equilibrium with the causative igneous

intrusions. As noted by several authors, these characteristics are strong evidence that early-

stage skarn fluids are dominated by magmatic fluids. Campiglia skarns provide an exceptional

case study where clinopyroxene and ilvaite are associated, and rhythmically crystallized

during the main stage of skarn formation

. For this reason we decided to perform oxygen and

hydrogen isotope analyses of hedenbergite and ilvaite in order to verify the potential

magmatic origin of fluids (methods in the electronic repository). Moreover, determination of

Pb isotope composition in sulphides from the early Zn-Pb(-Ag) ores and the later Cu-Fe ores

allowed to determine if the earlier distal mineralization had either a similar or a distinct source of metals with respect to the late “exotic” Cu-Fe(-Mg-Cr)-rich assemblage.

Oxygen and hydrogen isotope composition of ilvaite and hedenbergite sampled at different levels of the Lanzi and Temperino mine display a restricted overall variation. The δ

O and δD values of ilvaite vary in the range -1.7 / 1.9 ‰ and -149.3 / -130.0‰ respectively, while hedenbergite shows sistematically higher δ

O values: 1.88 / 5.15‰. Minerals from Temperino deposit are isotopically heavier than in Lanzi one. Composition of fluids calculated from both ilvaite (δ

O and δD) and hedenbergite (δ

O), using available fractionation factors

and temperature estimates (400°C;

and reference therein), plot near the typical magmatic reservoirs (I-type and S-type;

) and far away from the meteoric water line (Fig. 7). Fluid composition calculated from δ

O and δD values of representative TMP magmatic products display a bimodal distribution that clearly differentiate mantle-derived (δ

O ≈ 5.5/11; δD ≈ -78/-52) and crustal-derived (δ

O ≈ 12/13; δD ≈ -74/-39) contributions. Composition of the calculated skarn-forming fluid is consistent with an origin by dominant magmatic fluids, but their relatively lower δ

O and higher δD is indicative of some mixing with a low δ

O water reservoirs (meteoric, marine, basinal). The uncertainty about hydrogen fractionation factors in ilvaite do not allow to precisely model the sources and the extent of mixing. However, the overall composition of skarn-forming fluids was significantly different from that of local meteoric waters and modified meteoric waters that interacted with the Tuscan basement in the nearby Larderello geothermal field (Fig. 7). To estimate temperature, we considered hedenbergite-ilvaite isotopic equilibria by using the mineral-H

O fractionation coefficients of

. Our calculations provide a temperature range of ca. 370-450°C with an average of 410°C. Such temperature estimate is consistent with the upper temperature limits of ilvaite stability (ca. 450°C) suggested by

and with previous temperature estimates made by

and

.

Lead isotope compositions were determined on separate minerals (galena and chalcopyrite) from the three main ore types occurring in the Campiglia skarn deposits: (i) Zn-Pb(-Ag) ore;

(ii) Cu-Fe ore and (iii) Zn-Pb-Cu(-Ag) ore. The overall Pb isotope composition of Zn-Pb(-Ag)

and Zn-Pb-Cu(-Ag) ore types (galena) define a discrete field overlapping the compositional

field of TMP granites and in particular, matching composition of the monzogranite pluton in

Campiglia (Fig. 7). On the other hand, chalcopyrite from the Cu-Fe and Zn-Pb-Cu(-Ag) ore

types define a trend pointing towards the compositional field of TMP HK dacites-andesites/

latites and the latite-andesite mafic porphyry occurring in the Temperino deposit. Such a behaviour is particularly evident in the

Pb/

Pb vs.

Pb/

Pb diagram because mantle- derived magmas in Tuscany display a distinct thorogenic Pb isotope composition with respect to the crustal-derived magmas. This behaviour strengthen the proposed interpretation for a later emplacement of mafic porphyry in skarn bodies and the overprinting of the distal Zn- Pb(-Ag) ore by a Cu-Fe ore assemblage. Ore forming fluids seem to have collected metals from distinct magmatic sources: granite magmas/rocks for the Zn-Pb(-Ag) ore and HK calcalkaline magmas/rocks for the Cu-Fe ore.

Conclusions

Our investigation indicates that the Campiglia polymetallic skarn deposits cannot be considered anymore a classic example of zoned exoskarns, but rather they represent a prime example of distal Zn-Pb(-Ag) hedenbergite-ilvaite skarn where the spatial gap between the magmatic emplacement level (multiple intrusions) and the overlying distal hydrothermal zone was bridged by a pulse of mafic magma (and associated hydrothermal fluids) that escaped the main magmatic trap, adding “exotic” components like Cu, Mg and Cr. Mineralogical, geochemical and isotopic data indicate that the polymetallic skarns were produced by a time- integrated process involving distinct skarn forming- and ore forming-fluids (dominantly magmatic), and magmas. Magma intrusion into the skarn bodies was driven by the exceptionally high primary porosity (i.e. pockets) of the metasomatic bodies at Campiglia.

Similar Cu overprinting on Zn-Pb(-Ag) skarn deposits occur in other localities (e.g., the giant deposits Kamioka, Japan and Madan, Bulgaria;

) rising open questions on the genesis and evolution of these metasomatic systems. These odd and reverse geochemical patterns could be reconsidered in the light of the unique observations made in Campiglia skarn deposits. Reverse telescoping could be a common process in many districts but, while in Campiglia the causative magma reached the telescoped skarn, in other settings fluids only hit the distal skarns leaving behind the magma.

Magma chasing hydrothermal fluids in a distal skarn system is probably a serendipitous

occurrence, but the Campiglia case study provides evidences for developing a new powerful

conceptual model for deciphering the origin of such complex magmatic-hydrothermal

systems. Reverse telescoping in distal Zn-Pb(-Ag) systems could be also a marker in

exploration for hidden, higher-T Cu deposits at depth.

References

1. Meinert, L., Dipple, G., and Nicolescu, S. (2005) World skarn deposits: Economic Geology 100

Anniversary Volume, 299–336.

2. Einaudi, M.T., and Burt, D.M. (1982) Introduction, terminology, classification, and composition of skarn deposits. Economic Geology, 77, 745-754.

3. Einaudi, M.T. (1982) Description of skarns associated with porphyry copper plutons, southwestern North America. in Titley, S.R., ed., “Advances in the geology of porphyry copper deposits, southwestern North America”. Tucson, Univ. Arizona Press, 139-184.

4. Megaw, P., Ruiz, J., and Titley, S. (1988) High- temperature, carbonate-hosted Ag- Pb-Zn(Cu) deposits of northern Mexico. Economic Geology and the Bulletin of the Society of Economic Geologists, 83, 1856–1885.

5. Palinskaš, S., Palinskaš, L.A., Renac, C., Spangenberg, J.E., Lüders, V., Molnar, F.

and Maliqi, G. (2013) Metallogenic Model of the Trepca Pb-Zn-Ag Skarn Deposit, Kosovo: Evidence from Fluid Inclusions, Rare Earth Elements, and Stable Isotope Data. Economic Geology, 108, 135-162.

6. Einaudi M.T., Meinert L.D. and Newberry R.J. (1981). Skarn deposits. Econ. Geol., 75th anniversary volume, 317-391.

7. Haynes, F., and Kesler, S. (1988) Compositions and sources of mineralizing fluids for chimney and manto limestone-replacement ores in Mexico. Economic Geology and the Bulletin of the Society of Economic Geologists, 83, 1985–1992.

8. Megaw, P. (1998) Carbonate-hosted Pb-Zn-Ag- Cu-Au replacement deposits: An exploration perspective, in Lentz, D.R., ed., Mineralized intrusion-related skarn systems: Mineralogical Association of Canada Short Course Notes, 26, 337–357.

9. Samson I.M., Williams-Jones A.E., Ault K.M., Gagnon J.E. and Fryer B.J. (2008) Source of fluids forming distal Zn-Pb-Ag skarns: Evidence from laser ablation–

inductively coupled plasma–mass spectrometry analysis of fluid inclusions from El Mochito, Honduras. Geology, 36, 947-950.

10. Williams-Jones A.E., Samson I.M., Ault K.M., Gagnon J.E. and Fryer B.J. (2010) The Genesis of Distal Zinc Skarns: Evidence from the Mochito Deposit, Honduras.

Economic Geology, 105, 1411-1440.

11. Tanelli G. (1977) I giacimenti a skarn della Toscana. Rendiconti della Società

Italiana di Mineralogia e Petrologia, 33, 875-903.

12. Serri, G., Innocenti, F. and Manetti, P. (1993) Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the Neogene-Quaternary magmatism of central Italy. Tectonophysics, 223, 117–147.

13. Peccerillo A. (2003) Plio-Quaternary magmatism in Italy. Episodes, 26, 222-226.

14. Dini A., Innocenti F., Rocchi S., Westerman D.S. (2008) Magma emplacement in a transfer zone: the Miocene mafic Orano dyke swarm of Elba Island (Tuscany). In:

Thomson K., Petford N., Structure and Emplacement of High-Level Magmatic Systems. Geological Society, London, Special Publications, 302, 131-148.

15. Burt D.M. (1982) Skarn deposits – historical bibliography through 1970. Economic Geology, 77, 755-763.

16. Rodolico F. (1931) Ricerche sulle rocce eruttive recenti della Toscana - Le rocce del Campigliese. Atti Società Toscana Scienze Naturali, 41, 197-251.

17. Bartholomé P. and Evrard P. (1970) On the Genesis of the zoned Skarn Complex at Temperino, Tuscany. International Union of Geological Science, Serie A2, 53-57.

18. Burt D.M. (1977) Mineralogy and petrology of skarn deposits. Rendiconti della Società Italiana di Mineralogia e Petrologia, 33, 859-873.

19. Corsini, F., Cortecci, G., Leone, G. and Tanelli, G. (1980) Sulfur isotope study of the skarn-(Cu-Pb-Zn) sulfide deposit of Valle del Temperino, Campiglia Marittima, Tuscany, Italy. Economic Geology, 75(1), 83-96.

20. Dill,   H.   G.   (2010)   The  "chessboard"  classi9ication  scheme  of  mineral   deposits:  

Mineralogy and geology from aluminum to zirconium. Earth-Science Reviews 100(1-4), 1-420.

21. Vezzoni, S., Dini, A., Rocchi, S. (in prep) Reverse telescoping in a distal skarn (Campiglia Marittima, Italy). Chapter 6 of this thesis.

22. Ottens, B. and G. Neumeier (2012) The Huanggang Mine, Inner Mongolia, China.

Mineralogical Record 43, 529-563.

23. Gauthier, G. & Albandakis, N. (1991): Minerals of the Seriphos Skarn, Greece.

Mineralogical Record 22(4) 303-308.

24. Moroshkin V.V. and Frishman N. (2001) Dalnegorsk: Notes on Mineralogy.

Mineralogical Almanac, volume 4.

25. Pezzotta (2005) The Italian Island of Elba – A mineralogical jewel in the Tuscan

Archipelago. extraLapis English, Lapis International LLC, 8, pp. 96.

26. Dini A., Guideri S. and Orlandi P. (2013) Miniere e minerali del Campigliese – Il mondo sotterraneo del Parco di San Silvestro. Gruppo Mineralogico Lombardo, Milano, pp. 98.

27. Gustafson,   W.   I.   (1974)  The   Stability  of   Andradite,   Hedenbergite,   and   Related   Minerals  in  the  System  Ca-­‐-­‐Fe-­‐-­‐Si-­‐-­‐O-­‐-­‐H.  Journal  of  Petrology  15(3),  455-­‐496.

28. Liou, J. G. (1974) Stability Relations of Andradite-Quartz in the System Ca-Fe-Si- O-H. American Mineralogist 59, 1016-1025.

29. Capitani, G. C. and Mellini, M. (2000) The johannsenite-hedenbergite complete solid solution: clinopyroxenes from the Campiglia Marittima skarn. European Journal of Mineralogy 12(6), 1215-1227.

30. Samson, I. M., Williams-Jones, A. E., Ault, K. M., Gagnon, J. E. and Fryer, B. J.

(2008) Source of fluids forming distal Zn-Pb-Ag skarns: Evidence from laser ablation- inductively coupled plasma-mass spectrometry analysis of fluid inclusions from El Mochito, Honduras. Geology 36(12): 947-950.

31. Yaqian, Q. and Jibao, G. (1993) Study of hydrogen isotope equilibrium and kinetic fractionation in the ilvaite-water system. Geochimica et Cosmochimica Acta 57(13):

3073-3082.

32. Zheng, Y.-F. (1993) Calculation of oxygen isotope fractionation in hydroxyl-bearing silicates. Earth and Planetary Science Letters 120(3–4): 247-263.

33. Bowman, J. R. (1998) Stable-isotope systematics of skarns. Mineralized intrusion- related skarn systems. D. R. Lentz. Québec City, Mineralogical Association of Canada.

26: 99-145.

34. Zheng, Y.-F. (1993). Calculation of oxygen isotope fractionation in anhydrous silicate minerals. Geochimica et Cosmochimica Acta 57(5): 1079-1091.

35. Mariko, T., Kawada, M., Miura, M. and Ono, S. (1996) Ore Formation Processes of the Mozumi Skarn-type Pb-Zn-Ag Deposit in the Kamioka Mine, Gifu Prefecture, Central Japan - A Mineral Chemistry and Fluid Inclusion Study. Resource Geology 46(6): 337-354.

36. Bonev, I. K. (1977) Primary fluid inclusions in galena crystals. I. Morphology and

origin. Mineralium Deposita 12(1): 64-76.

Captions

Fig 1 – Structural sketch map of the northern Tyrrhenian area and distribution of the main magmatic centres of the Tuscan Magmatic Province. Blue dashed area are geological and geophysical lineaments with transversal orientation to the Apennine chain (after Dini et al., 2008 b). Stars indicate the main skarn deposits.

Fig 2 – A) Euhedral ilvaite crystals in pocket; B) Euhedral ilvaite crystal partially replaced by Fe-Cu sulphides; C) Similar to B but sulphides completely replaced ilvaite.

Fig 3 – A) Skarn pocket partially filled by quartz. The fibrous-radiating hedenbergite are responsible for the typical multi-inward-convex surface of the pockets; B) Another very large pocket (up to 10 m

) with similar textural features previously described but completely filled by mafic porphyry.

Fig 4 – Magnetite + hedenbergite zone replaced ilvaite at the direct contact with mafic porphyry. Also, Fe-Cu sulphides replaced and cut ilvaite layer. B and C) BSE-SEM image of the magnetite + hedenbergite zone.

Fig 5 – Diopside-hedenbergite-johannsenite plot for clinopyroxenes (from EPMA analysis). The clinopyroxenes show a main cluster near the hedenbergite apex with a minor variation toward johannsenite-rich composition. Only the overgrowths show a distinct variation with high-Mg content. SEM image shows the typical textures of the Mg-rich overgrowths.

Fig 6 - A) Comb structures of ilvaite crystals (Rio Marina, Elba). Partial (B) to total (C) replacement of ilvaite by Fe-Cu sulphides (Temperino mine, Campiglia Marittima).

Figura 7 – A) Composition of fluids calculated from both ilvaite (δ

O and δD) and

hedenbergite (δ

O) and from δ

O and δD values of representative TMP magmatic products

(for more details see text). B)

Pb/

Pb vs

Pb/

Pb on galena and chalcopyrite from the

three main ore types.

Fig.  1

Fig.  2

Fig.  3

Fig.  4

Fig.  5

Fig.  6

Fig.  7

Appendix A

App. A: Representative analyses of chlorite and actinolite replacing hedenbergite

Appendix B

App. B: δD and δ

O analyses of ilvaite and δ

O analyses of hedenbergite

Appendix C

App. C: δD and δ

O analyses of ilvaite and δ

O analyses of hedenbergite