The Traversella skarn

Sentences without a label are documented facts; the others are marked "probable" or "uncertain".

About 30 million years ago a small body of magma comes to a stop beneath the Valchiusella, the Chiusella valley. Where it touches the dolomitic marbles its hot fluids replace them with a new rock, skarn, and at the edge of the skarn the magnetite is concentrated. Traversella is the only mineralised skarn in the Alps: this page explains how it formed and how geologists have read it, from 1901 to today.

The pluton

The Traversella pluton is an intrusive body ranging in composition from quartz diorite to monzodiorite, of Oligocene age: about 30 million years, 30±5 according to the dating by Krummenacher and Evernden of 1960. It belongs to the Periadriatic magmatism, which between about 43 and 29 million years ago produced the plutons lined up along the Insubric Line, such as the Adamello, the Bregaglia and the Cervo; these bodies show no effects of Alpine metamorphism or deformation. It was emplaced when the main phase of the uplift of the Alps was slowing down.

It is small: about a fifth of the Cervo pluton and less than a hundredth of the Adamello. According to Munari the magma, at about 800-900 degrees, stopped at a depth of about 7 kilometres, at 1,500-2,000 bar, and cooled over a period of between thousands and tens of thousands of years probable. For him the rock is a fine-grained quartz microdiorite precisely because the pluton, being small, cooled quickly probable. The diorite crops out from the left bank of the Bersella, in the area of the mines, as far as Brosso and Calea; it also forms the Torretta delle Cime (1,556 metres).

Sources: Costa et al. 2019, pp. 5-6; Dalla Costa 2011, p. 2; Munari 2019, pp. 39, 80-82, 97, 106; Colomba 1913, p. 275

The host rocks

The magma intrudes the Sesia-Lanzo Zone, an Austroalpine unit of mica schists and gneisses, in the Eclogitic Micaschist Complex: mica schists, gneisses, dolomitic marbles, calc-silicate hornfelses, calciphyres and lenses of eclogites. Munari presents these rocks as part of a unit born on the African margin of the Tethys probable; according to him the mica schists derive from mudstones of about 320 million years ago, dragged down by subduction to a depth of about 50 kilometres and then exhumed probable. Compared with the Cervo, which intruded the same zone, at Traversella there are in addition layers of dolomitic marble and the presence of the ore deposits.

Near the pluton the rocks change through contact metamorphism: the marbles recrystallise, the mica schists turn black because biotite replaces phengite, and the garnet turns from red to black. This can be seen on the surface at the Pian del Gallo and better along the Anglosarda tunnel. Munari calls the slightly transformed ones hornfelsic mica schists and the compact ones, with biotite in small black flakes, hornfelses: the alteration increases closer to the diorite probable.

Sources: Costa et al. 2019, pp. 6-7; Dalla Costa 2011, p. 3; Munari 2019, pp. 26, 34-35, 42, 99-100

How skarn forms

Skarn is a rock that forms by metasomatism, that is, by the exchange of substances, between the hot fluids given off by an intrusive body and the pre-existing rocks, with the formation of garnet, pyroxene and other minerals. At Traversella the system is complete: the magmatic body, the transformed host rocks, various types of skarn with the useful minerals, hydrothermal veins and breccias, visible both on the surface and underground. Traversella is the only mineralised skarn deposit in the Alps.

The first-generation skarn, called prograde, is a granular rock of diopside and garnet, often with forsterite, which completely replaces the original rock: it lies in contact with the diorite and formed at high temperature from the reaction between the magmatic fluids and the dolomitic marbles, with an input of silica. It crops out at the Pian del Gallo, at the Pian degli Svedesi and along the Anglosarda tunnel. Then progressively cooler fluids hydrate these minerals and form the retrograde skarn, with talc, tremolite, actinolite, carbonates and chlorite. The fluids seeped in mainly along the contacts between different rocks and less along the fractures.

As early as 1901 Novarese proposed a two-stage origin, first by gaseous emanations and then by thermal waters: the minerals would have replaced above all the masses of limestone that were attacked and carried away probable.

Sources: Costa et al. 2019, pp. 2, 6-7; Dalla Costa 2011, pp. 3-4; Munari 2019, p. 102; Novarese 1901, p. 88

Where the magnetite is concentrated

The magnetite bodies are lined up roughly north-south, almost parallel to the contact with the pluton. They lie at the outer boundary of the prograde skarns, between these and the host rock, and their outer boundary is sharp. The coarse magnetite is associated above all with forsterite and diopside: for Costa and colleagues the magnetite bodies are therefore probably an iron-rich variety of the prograde skarn probable. The main mass of the skarn lies between a thick bed of dolomitic marble and the paragneisses that dip towards the pluton; along the Anglosarda the magnetite bodies can be seen right at the outer boundary of the skarns.

With the retrograde skarn come the sulphides, rare in the prograde: above all pyrrhotite, pyrite, chalcopyrite and arsenopyrite. Scheelite, the tungsten mineral, probably formed for the most part in this phase probable. Later still the circulation of fluids forms veins and breccias with quartz, carbonates, pyrite, gold and galena: many collectors' specimens come from these.

The miners recognised one variety of skarn and called it "Porta 'd Fer", the iron door, because it announced the ore a few metres away probable. Novarese already described the "garnet-bearing rock" of the deposit as limestone transformed into a mixture of garnet, epidote, pyroxene, chlorite and amphibole.

Sources: Costa et al. 2019, pp. 7, 9; Dalla Costa 2011, p. 3; Munari 2019, p. 103; Novarese 1901, p. 86

Brosso: the same magma, a different deposit

The Brosso deposits too, on the south-eastern edge of the pluton, depend on the intrusion, but they have very different characters: pyrite and haematite (the oligist of the old sources) in extensive limestone masses, interbedded with the mica schists. In 1901 Novarese found the great difference between Brosso and Traversella "obscure". The Brosso deposits are not in direct contact with the massif; Munari writes instead that at Brosso there is no lack of skarn magnetite in contact with the pluton uncertain. For him the hydrothermal fluids seeped into the more laminated mica schists near the Canavese Line and carried haematite, pyrite and chalcopyrite as far as Calea probable. Bonacossa could not establish whether pyrite or haematite had been deposited first and leaned towards pyrite probable.

Sources: Costa et al. 2019, p. 6; Novarese 1901, pp. 76, 79, 87, 90; Colomba 1915, p. 6; Munari 2019, pp. 106, 108

The ideas of the geologists of the time

The rock of the pluton was long thought to be a syenite. Rosenbusch and Novarese classified it as diorite, on the basis of Traverso's petrographic studies; in 1901 Novarese still calls it "syenite" to avoid misunderstandings, but reports that the dominant feldspar is a plagioclase and that chemical analysis confirms its dioritic nature. Novarese also shows that the massif is intrusive and younger than the mica schists, because it cuts across their layers.

Novarese held that the massif was no older than the Miocene, because it bears no traces of the Alpine folding; modern dating, to the Oligocene, supersedes this uncertain. According to Colomba, Müller saw at Traversella analogies with Persberg, in Sweden: first a skarn from the metamorphism of limestones, then the deposition of the metallic minerals probable. For the Riondello, Colomba, against Müller, proposes at least two superimposed beds of limestone uncertain.

Sources: Colomba 1913, p. 271; Novarese 1901, pp. 80-84, 92; Colomba 1915, pp. 5, 9-10

For the class

Skarn can be observed without going into a tunnel: at the Pian del Gallo the transformed rocks and the prograde skarns can be seen on the surface, while the best exposures of the diorite are in the quarries near Traversella and Vico Canavese. The specimen of "Porta 'd Fer" shown by Munari comes from the waste heap next to the Piano degli Svedesi probable; in a sample from the same heap the laboratories identified amphibole, andradite, calcite, clinochlore, magnetite, quartz and talc. Suggestion: compare the list with the minerals of the prograde and retrograde skarns on this page and say which phase each one belongs to.

Sources: Costa et al. 2019, p. 7; Munari 2019, p. 103; Baldassarre et al. 2025, pp. 9-10