Skarns in Romania

The following is a field trip guide to the skarns of the Banat region in Romania. It is one of the classic skarn localities of the world. Many mineral names and skarn terms originate in this area. Thus, the skarns of Romania are of interest to all "skarnologists". The text is available here as an introduction to this fascinating area. A complete version (Stefan Nicolescu (1996) - Excursion Guide. Field trip E1, Banat and Transylvanian Gold District, Romania. M&M 3, 3rd International Conference on Mineralogy and Museums, Budapest, Hungary, June 9 - 13, 1996, 60 pp., ISBN 963 8221 25 9) with the original Romanian spellings and text characters is available from the author:

Yale University
Dept. of Geology & Geophysics
PO Box 208109
New Haven, CT 06520
Ph: 203 432-3169 - Office
203 432-6022 - Lab
Fax: 203 432-3134


University of Gothenburg, Earth Sciences Centre, Department of Geology, S - 413 81 Gothenburg, Sweden and University "Babes-Bolyai", Department of Mineralogy, RO - 3400 Cluj, Romania


The E1 field trip of the third International Minerals and Museums Conference will take the participants to some of the, until recently, forbidden areas of Europe, i.e., Banat and Transylvania in Romania. For almost everybody the name "Transylvania" rings a bloody bell (see Bram Stoker & Co.). However, for the geologist, and especially for the mineralogist, it means also plenty of mineral deposits (mainly gold) from which quite a few new mineral species were described.

The trip will visit the Ocna de Fier - Dognecea ("Vasko/Moravitza - Dognacska") deposit - probably the second contact-metasomatic deposit, after Baita Bihor ("Rezbanya") (Nicolescu & Marza, 1989), to be recognised as such - and the famous Metaliferi Mountains ("Siebenburgische Erzgebirge") gold deposits at Sacaramb ("Nagyag") and Rosia Montana ("Verespatak"). The type localities for ludwigite and veszelyite (Ocna de Fier), pseudobrookite (Uroi = "Aranyer Berg"), krautite, krennerite, muthmannite, nagyagite, petzite and stutzite (Sacaramb = "Nagyag") are included in the itinerary. Literature sources disregarded so far in Romanian mineral history indicate that Dognecea is type locality for wollastonite-1T (Clark, 1993; see further discussion in "Skarn mineralogy" section) and that Sacaramb is also type locality for alabandite (Beudant, 1832). At Ocna de Fier, Uroi and possibly Sacaramb type material may be collected.

We really hope that you will enjoy this trip, in spite of some bumpy roads (one of the inherent features of Romanian society after fifty years of dictatorship and economic mismanagement). During "the ordeal" you will have the chance to see some of the most renowned mineralogical areas of Europe, as well as to sample not just minerals and rocks, but also specific Transylvanian cuisine and drinks, and to gain insight into one of the oldest cultures of Europe.


I. Location

The Ocna de Fier - Dognecea mining area (Banat, SW Romania) is located in the western part of the South Carpathians, 12 km north-west of the Caras-Severin county capital city of Resita (Plate 1).

The area can be reached by car, either from Bocsa, or from Resita. These two localities are accessible:

1. by train: from Bucharest to Caransebes (route 100) - Resita (route 115) - Bocsa (route 122) or from Timisoara to Bocsa (route 122).

2. by car on E 70 to either Caransebes (from Bucharest) or Voiteg (from Timisoara) and then on the national road no. 58.

II. Place names (see Appendix 2 for List)

The skarn deposit at Ocna de Fier (= "iron mine" in Romanian) is better known in the geological literature under its old name - "Moravica/Moravitza/Moravicza" (also the name of the creek that flows through the village, spelt "Moravita" in Romanian). The official Hungarian name of the locality between 1882-1918, ("Vasko" = "iron stone" in Hungarian), was also widely used in the geological literature. Less known is its German equivalent - "Eisenstein". According to Suciu (1968), the original name of the locality was however Ocna de Fier (the initial miners were Romanian settlers from Wallachia).

The southern extension of the deposit, mined around the village of Dognecea, is known in the literature as "Dognacska/Dognaczka" (Hungarian) or "Dognatschka" (German spelling).

These localities acquired their present names after the collapse of the Austro-Hungarian Monarchy in 1918, when Transylvania and the Banat became part of Romania. A list of Romanian localities quoted in old geological literature, with their Hungarian and German equivalents is given in Appendix 2.

III. Mining history

The area to be visited is among the few in the world that supported almost continuous mining for about 4000 years.

Archaeological evidence shows that mining in the area goes back to the Bronze Age (1900 - 1700 BC) (Kissling, 1967). The main ore exploited by that time was native copper from the oxidation zone of the deposit. Later on, the important iron ores occurring in the area determined the gradual switch from copper to iron mining.

Mining was an important activity of the ancient inhabitants of present day Romanian territory, the Dacians. After the Roman conquest in 106 AD., mining expanded all over the Roman province of Dacia, Ocna de Fier area included. At Berzovis (present day Berzovia, ten kilometres north-west of Ocna de Fier) a Roman metallurgy school, Schola fabrorum, was established, showing the keen interest the Romans had in metal extraction. At Cracul cu Aur (= "Golden Hill" in Romanian), north of Ocna de Fier, old Roman gold mining galleries are still to be seen. They are similar to the much better preserved ones at Rosia Montana ("Verespatak") in the Apuseni Mountains.

Gold mining at Cracul cu Aur ("Wolfganger Gebirge" in German) was active until the 13th century. Mining in the area diminished from 1554 to 1718, during the Ottoman Turk occupation. After the Turkish-Austrian peace accord at Passarowitz (1718), the Banat region became part of the Austro-Hungarian Empire.

Mining at Ocna de Fier - Dognecea was boosted by the new administration. The present day village of Dognecea was founded around 1720, while Ocna de Fier is mentioned for the first time in 1760 - 1765 (Kissling, 1967; Suciu, 1968). In 1719 an iron smelter was built at Bocsa, 7 km north of Ocna de Fier, and in 1721 the first copper smelter started operating at Dognecea. Ruins of the old smelters, as well as old slag dumps, can be still seen at Dognecea.

At Ocna de Fier mining progressed in both open pits and underground, while the Dognecea section of the deposit was mined mainly underground. The main open pits, listed from north-east to south-west, were (Plate 1): Eleonora, Paulus, Franciscus, Ignatius, Terezia, Delius, Magnet, Sfintii Arhangheli, Simon-Iuda, Elias-Enoch, Iuliana.

During the 18th and the 19th centuries iron ore production increased steadily, up to 146,150 t in 1897. This increase, combined with the discovery of important coal deposits some 20 km south-east of the area, triggered the development of an important iron smelting centre at Resita.

After the 1st World War the production decreased, mainly due to exhaustion of the richest parts of the deposits. A short revival of the mining occurred in the sixties and seventies when better ore dressing methods were used, allowing mining of lower grade ore.

Today the magnetite, lead, zinc and copper ore reserves that once made the region famous are completely exhausted. Mining is restricted to recycling old dumps with 18 - 25 % FeO. However, an estimated 2 Mt of hematite bearing (mainly granditic) skarns remain in the area. Provided that an adequate ore dressing technology can be applied (e.g., electrostatic separation), the life of the deposit could be extended by a decade or so. This could be even longer if the important reserves of industrial minerals (garnet) and dimension stone (marble, granodiorite) present in the area were mined.

IV. Brief geological literature survey

The big ore reserves, the unusual (for mid-19th century geologists) genetic type, as well as the rich mineral parageneses present (about 140 mineral species and varieties described to date - Table 2), led to major scientific interest of the geologic community in the Ocna de Fier-Dognecea deposit. This is reflected by more than 250 scientific papers published on subjects related to the area.

One of the first papers dealing with the deposit was published in 1774 by Ignaz Edler von Born, following his 1770 visit to Banat. In his description, v. Born mentions the Simon-Iuda mine, which he considers the richest copper mine in Europe. More important is the discussion by Born of a paper published in Vienna by Christoph Traugott Delius, from which it is obvious that Delius knew two very important features of the deposit: 1) the ore is situated between two different rock types, limestone in the hanging- and hornfels in the footwall, and 2) the ore is younger than the surrounding rocks. This is the very first hint at what later became the contact-metasomatic ore genetic type.

A major cornerstone in the study of the deposit was laid in 1864 by the German Bernhard von Cotta. In his classical study, Cotta states specifically that Moravitza (Ocna de Fier), and other deposits he visited in Banat at Moldova Noua ("Ujmoldova"), Sasca Montana ("Szaszkabanya"), Ciclova ("Csiklova"), Oravita ("Oravitza"), are generated at the contact between younger intrusive rocks and Mesozoic limestones. Following the proposal of the Austrian geologist Karl F. Peters (1861), referring to the deposit at Baita Bihor ("Rezbanya") in the Apuseni Mountains, he names the contact formation "Contactbildung" (Nicolescu & Marza, 1989). He recognises that in all the contact type deposits from Banat, the intrusive, although highly variable petrographically, is consanguine. He proposes the name "banatite" for these rocks, mentioning that he does not intend to introduce a new rock type, but wants to underline the "banatite" consanguinity. In modern terms this is the equivalent of an intrusive suite. He also describes the typical metasomatic texture developed by magnetite in calcite - "Tiegererz", later called Liesegang texture, and today referred to mostly as "banded skarn".

In 1885-1886 the Swede Hjalmar Sjogren published two papers on the deposit, comparing the deposit at Ocna de Fier - Dognecea with the ones from the Varmland district, Central Sweden. After Niedzwiedzki (1873), who seems to have been the first to study a thin section from the area (a banatite sample from Dognecea), Sjogren was the first one to study the metamorphic rocks and the skarn under the polarising microscope. Thus, he is the first one to notice and discuss grandite zoning, twinning and birefringence in the area, as well as relationships between different skarn building phases - garnets, pyroxenes, amphiboles. For the first time he uses the term "skarn" in connection with the deposit. This is an old Swedish mining term, meaning "rubbish" and referring to silicate gangue minerals (pyroxene, garnet, amphibole) associated with Swedish ore deposits. It seems to have been used for the first time in the literature by Tornebohm (1875 a) in a paper on the Persberg iron deposit in Sweden. This paper predates the one on the Norberg deposit (Tornebohm, 1875 b), quoted by Burt (1982) as being the first using the term "skarn" in the geological literature (T. Bergman, personal communication). In disagreement with Cotta, Sjogren regards the deposit and the calcareous rocks (marble) as related to the "Archaean" schists in the area. The intrusive is regarded as post-dating the ore, and as being Upper Cretaceous in age (correct). Both the ore paragenesis and its association (in Sjogren's view) with the "Archaean" schists leads him to the conclusion that the deposit is similar with the ones at Persberg and Nordmark in Sweden.

In his classical, and still relevant study on contact metamorphism and metasomatism of the Christiania (Oslo) area (1912 for 1911), V. M. Goldschmidt mentions the deposits at "Vasko" and "Dognaczka", quoting two andradite and two manganoan hedenbergite analyses published previously.

In his treatise on metallogeny (1913) de Launay names the contact deposits north of the Danube as being of "Banat type".

In his well-known work on mineral deposits, Lindgren (1933) discusses "the classical deposits of the Banat province" (p.713). Following this work, the deposit at Ocna de Fier - Dognecea and the similar ones in Romania (Banat: Moldova Noua, Sasca Montana, Oravita-Ciclova, and Apuseni Mountains: Baita Bihor, Baisoara) leave the international stage and enter a long period of isolation.

In spite of the isolation, research by Romanian geologists continued, and many of the problems related to the geology and mineralogy of the deposit were solved.

The work that clarified the petrography and tectonics of the area was published in 1931 by Al. Codarcea. He confirms Cotta's hypothesis that all intrusive rocks in the area are consanguine and agrees with the generic term "banatite" for them. The dominant petrographic type for the area (and for the whole region in which banatites occur, from Serbia to the Apuseni Mts.) is determined as being granodiorite (in previous studies, the intrusive rocks were given a variety of names, from syenite, to quartz-trachy-andesite and quartz diorite). Another major contribution of Codarcea's work is the discrimination of the different metamorphic rocks occurring in the region: greenschist facies rocks (metapelites, meta-volcanics, meta-intrusives) and hornfelses. Finally, the geologic map accompanying the work (scale 1:25,000) is a major contribution, being the one in use ever since.

The last major studies on the deposit were published by Al. Kissling (1967) (Ocna de Fier) and N. Vlad-Serban (1974) (Dognecea). In these, the mineralogy and composition of major minerals in the deposit are discussed.

In spite of the many studies published so far, basic questions related to the genesis of the deposit: thermobaric regime and source of fluids/metals, are yet to be answered.

V. Geology of the deposit (Plate 1)


The basement of the region comprises greenschist facies Precambrian metapelites and Palaeozoic meta-igneous rocks.

A narrow, NNE - SSW trending (N25E) Upper Jurassic-Lower Cretaceous limestone syncline lies discordantly over the metapelites. It extends beyond the limits of the deposit, from Ezeris in the north 30 km to Carnecea in the south. It has a maximum width of 1 km at Dognecea.

The whole region is part of a fold and thrust belt, related to the Alpine orogeny. The metamorphic rocks and limestones were intruded ~ 65 Ma ago (K/T boundary) by a suite of calc-alkaline igneous rocks, mainly of granodioritic composition ("banatite" - Cotta, 1864). The contact between the banatite and the intruded rocks generated a strong thermal metamorphic and metasomatic halo. Depending on the nature of the affected rocks (metapelites or limestones), hornfelses, skarns and marbles were formed. Geothermobarometry on hornfelses indicates a peak temperature of 600* C at a depth of ~ 8 km (2.8 kb) at the time of intrusion (Nicolescu & Cornell, 1996).


The main S-shaped igneous body in the area (Plate 1) is linked in the north with the large banatite laccolith at Bocsa. Smaller banatite bodies crop out further south, down to the Danube, at Oravita, Sasca Montana and Moldova Noua (Fig. 1), along the western limit of the Resita - Moldova Noua limestone synclinorium. Each of these banatite - limestone contacts hosts a skarn deposit. In Romanian geological literature the magmatic event that generated the banatites is referred to as the Laramide orogeny, by analogy with the North American orogen.

(Fig. 1)

The igneous rocks associated with the skarn deposit at Ocna de Fier - Dognecea, are normal (metaluminous), *mol. % (Na2O + K2O) < mol. % Al2O3 < mol. % (CaO + Na2O + K2O)*, I-type (Table 1), calc-alkaline, volcanic-arc granodiorites (Fig. 2) (Nicolescu, unpubl. data).

(Table 1)

(Fig. 2)



Skarn occurs quasi-continuously for ~ 7 km, from the inflection point of the Moravita valley (Eleonora and Paulus open pits) north of Ocna de Fier, to the Aurora mining area (3 km south of Dognecea) in the south. The only discontinuities are along two deep transversal valleys, Vintilii and Lacului Mic, where erosion removed both skarn and accompanying marble. Skarn envelops the marble on both flanks of the limestone syncline, sometimes replacing it completely. The average skarn thickness on both sides of the syncline is 30 - 40 m, with maxima of 300 m in the Iuliana area. Vertically, skarn extends ~ 100 m, down to the bottom of the syncline. A rough estimate is that at present erosion level, the deposit contained over 200 Mt. of skarn.

Skarn types

Most of the skarn at Ocna de Fier - Dognecea is located at the contact between metamorphic rock and limestone (Fig. 3). Calcic exoskarn is dominant, with subordinate endoskarn. Magnesian exoskarn, predating calcic skarn, is also found. The bulk of the skarn is of infiltration type, with local development of diffusion skarn. Banded skarn, developed on decimetric scale throughout the area, comprises magnetite-marble (most frequent), magnetite-garnet, magnetite-pyroxene, garnet-pyroxene.

Skarn zonation and ore grade

From north to south there is an obvious metallogenetic zonation, with calcic Fe skarn (overprinting subordinate magnesian Fe skarn) at Ocna de Fier in the north, mixed calcic Fe - calcic Zn-Pb (Cu) skarn in the central section and calcic Zn-Pb (Cu) skarn at Dognecea in the south (classification of Einaudi et al., 1981; Einaudi & Burt, 1982; Meinert, 1992).

Zonation is reflected by both prograde silicates and ore minerals. Prograde skarn silicates are distributed as follows:

- north: grandite (mainly andradite) ± diopside

- centre: grandite + diopside ± hedenbergite

- south: (manganoan) hedenbergite ± grandite

while the corresponding ore mineral distribution is:

- north: magnetite, hematite ± Zn-Pb (Cu) sulphides

- centre: hematite, magnetite + Zn-Pb (Cu) sulphides

- south: Zn-Pb (Cu) sulphides ± hematite

The magnetite ore is considered to be related to the skarn proper, while the hematite and the sulphides are of later hydrothermal origin.

The initial grade of the iron ore was around 60 % FeO. After extensive mining, the grade has dropped to 18 - 25 % FeO. Sulphide ore grade was at the level of several percents for each Pb, Zn and Cu, with interesting amounts of Au and Ag. Sulphide ore mining ceased in the early eighties. Minor scale magnetite mining (recycling of old mine dumps) is the only mining activity continuing today.

(Fig. 3)

Skarn mineralogy

Around 140 mineral species and varieties have been mentioned so far from the contact zone of the deposit. Table 2 gives a (comprehensive?) list of mineral species and varieties formed by contact metamorphism/metasomatism and subsequent hydrothermal processes. The list, based mainly on old mine records and on the work of Clark (1993), Codarcea (1931), Cotta (1864), Ghergari et al. (1986), Ilinca et al. (1993), Kissling (1967), Nicolescu (1982), Vlad (1974) and other references cited in the table, does not include rock forming minerals of igneous, metamorphic and sedimentary rocks occurring in the area. Underlined species in italics are likely to be collected in the field.

(Table 2)

Ludwigite was described "from the southern part" of the deposit (Iuliana area?) by Tschermak (1874), based on the analysis of Ludwig (hence the mineral name). It is an intermediate member of the ludwigite - vonsenite series, with up to 30 mol. % vonsenite (Kissling, 1967). Veszelyite, discovered by Schrauf the same year as ludwigite, was never found again at Ocna de Fier. According to data published by Clark (1993), in 1818 J. Leman renamed the "Tafelspath" of Stutz (1793, cited also by Karsten in 1800) to wollastonite (wollastonite-1T), on samples from Dognecea. However, although present at Dognecea, wollastonite has a subordinate character in the skarn here. It seems very likely that Stutz, who apparently did not visit the deposit, described "Tafelspath" in a hand-specimen brought to him, probably from another skarn deposit in Banat, i.e., Ciclova, where wollastonite is widespread and well developed. Thus, Ciclova and not Dognecea might be the true type locality for wollastonite (G. Papp, personal communication; see also Estner, 1797, and Esmark, 1798). Beudant (1832, p. 218 - 219), too, quotes "Csiklova" as the sole wollastonite occurrence in Banat. However, at this stage seems that more data are needed to "move" the type locality of wollastonite from Dognecea to Ciclova. It is sure though that wollastonite has to be added to the list of minerals first described from present day Romanian territory. This is a novelty for Romanian mineral history, overlooked so far in all reviews on mineral species described from Romania.

Two other mineral species described from the area, were discredited later. "Dognacskaite" was described by Krenner (1884) from Dognecea as Cu2Bi4S7. It was shown by Koch (1948) to be a mixture of bismuthinite, chalcocite and chalcopyrite. "Warthaite" (Pb4Bi2S7?) was described by Krenner (1925) from Ocna de Fier, but Thompson (1949) showed it to be a mixture of cosalite and galena (synonymous with "goongarrite").

To end the discussion of non-valid mineral species from Ocna de Fier - Dognecea, the situation of "rezbanyite" must be mentioned. A mineral with this name was initially described by R. Hermann (1858) from another banatite-related skarn deposit, at Baita Bihor ("Rezbanya") (Apuseni Mountains). After showing that Hermann's "rezbanyite" is in fact cosalite, A. Frenzel (1883) reused the name for a mineral he described also from "Rezbanya". After Frenzel's holotype material was lost, Koch (1930) redefined the mineral as Cu2Pb3Bi10S19 in samples from Ocna de Fier. In 1992 Zak et al. investigated Koch's holotype or cotype material housed in the Mineralogical Collection at the University of Szeged (Hungary) and showed that the material (homogeneous by optical microscopy) consists of a mixture of minerals in the bismuthinite - aikinite group (hammarite and krupkaite) and subordinate Cu-bearing cosalite. Thus, "rezbanyite" was formally discredited too (Zak & Mumme, 1994).

To conclude the mineralogy section, the following minerals are worth mentioning: ludwigite, veszelyite and wollastonite-1T as topotype minerals, large grandite crystals (rhomb dodecahedra up to 10 cm, frequently of 1 - 3 cm), magnetite crystals (1 cm rhomb dodecahedra), large hematite rosettes - "iron roses" (up to 8 cm diameter), phlogopite plates (up to several square decimetres, frequently 1 - 2 cm2), large pyroxene (up to 10 cm) and amphibole (exceptionally up to 30 cm, but frequent of 10 cm) aggregates, ludwigite aggregates (2 - 4 cm sized), various pyrite crystal forms (~ 200 described to date, mainly from Dognecea), calcite crystals (~ 30 forms; up to 25 cm large rhombohedra), different bismuth sulphosalts, Japanese and "coaxial" (Gruescu, 1975) quartz twins.

Mineral pseudomorphs

The recurrent character of both skarn and hydrothermal events resulted in the occurrence of extensive and interesting mineral replacements and pseudomorphs:

(a) Recurrent crystallisation of minerals formed during roughly the same event, e.g., several diopside and grandite generations with respect to magnetite (Kissling, 1967; Nicolescu, 1982)

(b) Pseudomorphs: rhodonite after grandite (Radulescu & Dimitrescu, 1966) magnetite after grandite (Udubasa et al., 1992), magnetite after hematite ("mushketovite") (Kissling, 1967; Marza et al., 1990), hematite after magnetite ("martite") (Kissling, 1967; Marza et al., 1990), dolomite after magnetite (Marza & Bedelean, 1978).

(c) A very unusual occurrence, mentioned so far twice from Ocna de Fier - Dognecea, is the spectacular and very rare appearance of fossils in skarn. The first mention, a skarnified coral, was made by K. F. Peters, quoted by Sjogren (1886). Unfortunately, Sjogren does not give the initial description of Peters (made on a museum label that Sjogren had access to). The specimen, a radial, coral-like aggregate on a magnetite sample, was in the custody of J. Szabo at the Mineralogical Collection of the University in Budapest (Hungary) and a similar sample, from the same locality, is said by Sjogren to be in the collection of the "Ober-Realschulle" in Budapest. Sjogren concluded that the radial aggregates, made of carbonate, quartz, epidote, a manganese-bearing mineral (manganoan epidote or rhodochrosite) and a pyroxene or wollastonite-like mineral, are of inorganic origin. His opinion was shared by Krenner, who wrote him: "Ich glaube, dass der Anblick dieser Stuecke Sie ueberzeugen wird, dass hier von einer Koralle nicht die Rede sein kann", i.e., "I think that the look of this fragment will convince you that, without doubt, this is not a coral". It should be mentioned that it was very important to Sjogren that the sample not be of organic origin. In his view, the deposit was Archaean, so he did not expect any organic remnants in it. However, almost 100 years later, the question of fossils in the skarn at Ocna de Fier was raised again by Marza (1979), who described a talc-epidote substituted lamaellibranchiate (Pachiodonte) valve from the garnet-diopside-epidote skarn. Another authentic Romanian fossil occurrence in skarn is the one on Lesului valley in the Vladeasa Mountains (Apuseni Mountains). The protoliths are similar to those at Ocna de Fier - banatitic intrusives and Mesozoic (Senonian) limestones. As a consequence of skarnification, the abundant gastropod (Actaeonella) shells in the limestone are substituted by wollastonite, while their inner part is substituted by an aggregate of fine diopside and large (0.5 cm) green andradite crystals (Nicolescu & Marza, unpubl. data).

Detrital iron ore

At the Strosu and Amelia Hills (~ 1 km north of Ocna de Fier, along the limestone syncline), important concentrations of detrital magnetite and hematite ores were mined. These concentrations in Pliocene detrital rocks (sands, silts, clays), were made up of very large, over 1 m3, rolled magnetite and hematite blocks, with little or no skarn silicates. It is generally accepted that the ore blocks were eroded and transported from the primary skarn deposit, although it is difficult to imagine an agent strong enough to carry and roll magnetite blocks sometimes larger than 1 m3.


NOTE: Most of the areas to be visited are long abandoned open-pits, with steep slopes, many loose blocks and uncovered, deep shafts. It is therefore highly recommended strictly to follow the route indicated by your field guide.

1. Paulus open-pit, Ocna de Fier

The pit, located on the left bank of Moravita, 1250 m north-east of the intrusive body (Plate 1), can be accessed from the main road entering Ocna de Fier from north. The headgear of the abandoned Paulus mine lies immediately next to the pit entrance. A good geological reference point is the 300 m wide marble outcrop, on the right bank of Moravita. A short walk (150 m) south of the road leads to the pit.

Paulus is the northernmost pit of the deposit. It is crescent shaped (250 x 125 m), with the longer dimension running NW - SE. Calcitic and dolomitic marble are dominant rock types. Material from different parts of the deposit was dumped here, thus loose fragments at the pit bottom are atypical for this area.

Of interest are the western wall of the pit and the marble ridge that separates its western and eastern sectors. Most of the western wall is made up of dolomite with an irregular magnetite vein net. Metre sized lensoid magnetite bodies occur towards the central part of the wall, at the joint with the marble ridge. Approximately in the middle of the ridge, a 4 m wide white calcite vein can be seen. The calcite rhombohedra are very large, up to 25 cm. Dolomite and calcite compositions are almost ideal (Fig. 4a).

Rare earth element (REE) analyses of magnetite reveal slight light rare earth element (LREE) enrichment and unusual negative Ce anomalies (Fig. 5c). The magnetite REE profile is thus similar to that of carbonate rocks and of modern sea-water (Fig. 5a). This suggests a limestone derived source for magnetite. The REE profile of the large calcite rhombohedra (Fig. 5b), with heavy rare earth elements (HREE) enrichment and positive Eu anomaly, clearly indicates the hydrothermal origin of the calcite.

(Fig. 4)

2. Terezia open-pit, Ocna de Fier

The Terezia open pit is located 250 m NE of the main banatite body crossing the syncline along Vintilii brook (Plate 1). It can be accessed from the dirt road going upstream Moravita, 750 m from the church in Ocna de Fier. Access to the pit is along the marble ridge that separates its northern and southern sectors (J.4 area, Plate 2). The northern pit, 70 m deep ("Terezia Mare", i.e., Big Terezia), has circular shape, with a diameter of 120 m. The southern pit ("Terezia Mica", i.e., Little Terezia) is elliptical (200 x 100 m) and is 95 m deep (Little Terezia is actually bigger!).

The dominant rock type in both pits is granditic skarn (Fig. 4b, Plate 2). Its impressive development gives an idea about the scale of the reactions that generated the deposit. In Terezia Mare characteristic metasomatic textures: grandite concretions in marble, "orbicular" textures (ring ore) and banded skarns (Liesegang texture), can be seen. Decimetre scale garnet concretions in marble are visible on the south-western wall. Close to the bottom of the south-eastern wall (H.5 area, Plate 2) Kissling (1967) describes "orbicular" textures (ring ore): diopside cores surrounded by magnetite rings, all in massive granditic mass. Banded skarns (grandite-magnetite, marble-magnetite) occur on the southern wall. Rhomb-dodecahedral garnet crystals, sometimes several centimetres in diameter, develop at garnet-marble contacts. In a huge block (several cubic metres volume), detached from the western wall, the zonation: grandite skarn, diopside skarn, marble is well represented. The diopside from Terezia Mare has a composition of about Di90 (Fig. 4c). Cross-cutting the marble, later hematite - quartz veins are developed. South of this block, the pit wall is made almost entirely of epidote retro-skarn. In this area Kissling (1967) mentions the northern-most occurrence of ludwigite at Ocna de Fier (K.6 area, Plate 2). Along the southern and eastern walls, grandite skarn is extensive. Sporadic magnetite lenses occur here. In the north-eastern wall of Terezia Mare pit, several banatite veins occur - gabbro, lamprophyre and granodiorite-porphyry. Besides the above mentioned minerals and rocks, minor phlogopite-rich magnesian skarn occurs towards the bottom of Terezia Mica pit (M.4, P.4 areas, Plate 2). Green phlogopite crystals, up to 2 cm diameter, occur in a finer and somewhat lighter phlogopite-chlorite "matrix".

Minerals to be found in the Terezia pits (less frequent minerals in parentheses) are: garnet, diopside, epidote, tremolite, (phlogopite), calcite, quartz, magnetite, hematite, (ludwigite), chalcedony. According to old mine plans and literature, the mineral selection was initially much bigger. At higher levels (now removed) native bismuth, bismuthinite, bismutite, stibnite and greenockite were also found.

(Fig. 5)

3. Iuliana open-pit, Ocna de Fier

The pit can be best reached from the dirt road connecting Ocna de Fier and Dognecea. Three and a half kilometres south of the divide between the Moravita and Dognecea hydrographic basins (Dealovat - Plate 1), a road turns off to north-west, along the Aron valley and leads three and a half kilometres to Iuliana. One kilometre upstream along this very poorly maintained road, a road turn-off to the west leads to the entrance of the pit. Access to the pit is via its second terrace (P.2 area, Plate 3). This is the largest pit in the area: 280 x 150 x 90 m. It is located on the western edge of the limestone syncline (Plate 1), in the mixed calcic Fe - calcic Zn-Pb (Cu) skarn zone. The northern and western walls are in grandite (± tremolite) skarn, while the southern wall is marble dominated (Plate 3).

Garnets from this area do not differ from those in the rest of the deposit (Fig. 4b, Table 3). They developed beautiful idiomorphic rhomb-dodecahedral crystals, sometimes 2 cm across. Smaller sized crystals are very frequent. They range in colour from dark-brown to yellow-brown, depending on andradite content (green andradite crystals were found too). In thin section they show oscillatory zonation (isotropic zones alternate with birefringent ones) and sector twinning (Fig. 6). Isotropic zones are pure andradite while birefringent ones are intermediate andradite-grossular. Birefringence increases (up to 0.01) with grossular content, but occurs only when andradite end-member is below 92 mol. % (Table 3) (Nicolescu, 1995). Birefringent zones have sometimes biaxial optical character. Birefringence is not uncommon among skarn garnets (Deer et al., 1993), being ascribed to stress induced by departure from the cubic symmetry.

(Fig. 6)

The granditic skarn is extensively replaced by amphibole (tremolite-actinolite) skarn. Amphiboles form large radial aggregates, with needle shaped crystals up to several tens of centimetres (frequently 8 - 10 cm). Radial aggregates of epidote (cm-sized) also occur. Different degrees of substitution of primary granditic skarn by secondary amphibole skarn, cross-cut by subsequent hematite-quartz veins, are well reflected in the skarn blocks on terrace two (column O, Plate 3). In vugs and along fissures in the rock, mostly pyrite (± galena, sphalerite and chalcopyrite) occur. This indicates the transition from the oxidic facies of the mineralization in the north (Ocna de Fier) to the sulphidic one in the south (Dognecea).

The main ore mineral in the quarry is hematite (specular iron), that forms "iron roses", several centimetres in diameter. Often the hematite is accompanied by quartz crystals. Fluctuations in the redox regime at ore forming time led to alternate magnetite-hematite ("mushketovite") and hematite-magnetite ("martite") substitutions.

Iuliana pit is the southernmost occurrence of ludwigite at Ocna de Fier. Ludwigite forms decimetre thick bands, associated with magnetite and serpentine (I.12, I.13 areas, Plate 3). The pitch-black coloured ludwigite bands are made of fibrous aggregates, either radiating or sub-parallel (prior to Tschermack's paper it was usually called "black asbestos"). A detailed crystallographic and chemical study of ludwigite from Ocna de Fier (on specimens from another occurrence, Magnet open pit, one kilometre south of Terezia) was done by Kissling (1967). Opinions are divided about the exact place within the deposit from where ludwigite was originally described. Probably because it is more abundant in the Magnet open pit, most opinions tend to place its type locality there. In his paper Tschermack (1874) states only that the samples came "from the southern part" of the deposit. It is therefore more likely that the type locality is at Iuliana. The pit at Magnet is unfortunately not accessible anymore.

Rare earth element data were collected on several mineral species from Iuliana (Nicolescu & Cornell, 1996). REE profiles of both individual grandite zones and bulk granditic skarn have similar features. First of all, the garnet REE content is surprisingly low. Then, the chondrite normalised profiles (Fig. 5d) show LREE enrichment (normal for grandite) and negative Ce and Eu anomalies. These are features similar to those of carbonate rocks and magnetite in the deposit, and also to modern sea-water (Figs. 5a, c). REE profiles of both intrusive and metamorphic rocks in the area have typical granodiorite and metapelite patterns respectively (Fig. 5e, f). These data strongly suggest that, at least partly, the skarn mineralising fluids and the associated metals were limestone-derived and not of magmatic or metamorphic origin. Cressey (1987) drew similar conclusions on the genesis of a small skarn occurrence from the Isle of Arran, Scotland. The REE pattern of tremolite from the Iuliana pit (Fig. 5d) has positive Eu anomaly and even lower contents than garnet, suggesting its formation at a different (later) stage, from other fluids than the ones from which the grandite precipitated.

Mineral species possible to collect in the Iuliana open pit are: grandite, tremolite-actinolite, epidote, calcite, magnetite, ludwigite, serpentine, hematite, pyrite, galena, sphalerite, chalcopyrite, marcasite, quartz.

(Table 3)

4. Ferdinand mine dump, Dognecea

Access to the dump is 1700 metres upstream on Lacului Mic valley, along a poorly maintained dirt road.

The main mineralogical differences between Ocna de Fier and Dognecea fields are the dominance of pyroxene (mainly hedenbergite - manganoan hedenbergite, Fig. 4c) over garnet and the sulphidic character of the mineralization at Dognecea as opposed to the garnet dominated oxidic mineralization at Ocna de Fier. Thus, Dognecea is a good example of calcic Zn-Pb skarn deposits (Einaudi et al., 1981; Einaudi & Burt, 1982; Meinert, 1992).

As opposed to Ocna de Fier, the Dognecea sector of the deposit was mined mostly underground. The Ferdinand mine extended for 7 km along the whole length of the deposit, from Ocna de Fier to Dognecea. Thus, the dump at this southern end of the mine hosts a collection of all rock types and many minerals occurring in the deposit.

Besides the already mentioned mineral species, hedenbergite and ilvaite and their manganoan varieties are characteristic for Dognecea. Hedenbergite is abundant, usually associated with sulphides. It occurs in large dark-brown or brown-black radial aggregates, with prismatic crystals frequently 8 - 10 cm long. Ilvaite is a much rarer species and good ilvaite samples are scarce.

On the way back to Ocna de Fier, time permitting, a cordierite occurrence can be visited. This is located one kilometre north of the Moravita - Dognecea water divide (or one and a half kilometre from the church in Ocna de Fier). Here the road crosses the contact between the main intrusive body and the metapelites. Cordierite rich hornfelses are most easily found in boulders along the road - the area is heavily forested and outcrops are scarce. The boulders are made of bluish-black cordierite, sometimes with banded texture.

5. Gruescu mineral collection, Ocna de Fier

A very good selection of many beautiful mineral specimens from the area is in the collection of Constantin Gruescu (retired mine official) in Ocna de Fier.

The collection is located at 113 Vale St., RO - 1736 Ocna de Fier, Romania. Mr. Gruescu (born at Dognecea in 1924) acquired the first samples in 1941. These first items were inherited from his grand- and great-grandfathers, who witnessed the best years in the millennia long history of the deposit. The collection as such was started in 1951, and gradually grew to its present dimension. Without any outside financial support (in spite of its outstanding value, the colection produced "just" fame), it was moved from the house proper, were it started taking over the inhabitants, to a wing built on purpose, provided with apropriate security systems. Today the collection comprises between 4 - 5000 samples, out of which 1400 are on permament display. The main part of the collection (1100 items) is made up of mineral samples representative for the Ocna de Fier - Dognecea deposit. Beside this, a small collection of samples representative for the northern Transylvanian mining area of Maramures and some minerals obtained by exchange, mainly from abroad (Germany, France, Italy, etc.), can be seen. Many samples (estimated at 2000) were donated during the years to several Romanian high schools, universities and museums.

Among the "big guns" of the collection are: garnet crystals up to 8 cm diameter, 10 cm long diopside crystals, a ~ 40 cm2 phlogopite plate, one centimetre large magnetite rhomb dodecahedra, magnetite pseudomorphs after silicate minerals (garnets, pyroxenes, etc.), hematite "roses", dolomite pseudomorphs after octahedral magnetite crystals, Japanese quartz twins, unusual quartz aggregates (so called "coaxial" twins), two centimetre large radial malachite aggregates and many more. Overall, the collection represents an excellent summary of the mineralogy of the deposit at Ocna de Fier - Dognecea.


The author wishes to express his gratitude to the following individuals:

- Dr. Gabor Papp (Hungarian Natural History Museum, Budapest) whose extremely useful and pertinent comments and suggestions considerably improved the accuracy of mineral history data provided in this guide;

- Dr. David H. Cornell (University of Gothenburg, Sweden) who changed the "Romenglish" of the initial manuscript into an understandable language; changes in the text were made after his kind help, thus any spelling/grammar mistakes are the full resposibility of the author and the editor;

- last but not least, Ionut Caprita - mine geologist at Ocna de Fier and "nea Costica" Gruescu - devoted mineral collector from Ocna de Fier, who during the many years of geological "love affair" of the author with the deposit provided help, advise, samples, shelter and friendship.


Beudant, F.S. (1832) - Traité élémentaire de minéralogie. Tome II.

Chez VerdiŹre, Libraire-ƒditeur, Paris, 797 pp.Born, I.E. (1774) - Briefe uber Mineralogische Gegenstande, auf seiner Reise durch das Temeswarer Banat, Siebenburgen, Ober- und Nieder-Hungarn an den Herausgegeber derselben, Johann Jakob Ferber, geschrieben, Frankfurt u. Leipzig, 228 p.

Burt, D.M. (1982) - Skarn Deposits-Historical Bibliography through 1970, Econ. Geol., 77/4, p. 755 - 763.

Cioflica G., Vlad-Serban, N. (1973) - Contributions a la connaissance des types structuraux de pyrometasomatites laramiens de Roumanie, Rev. roum. Geol. Geophys. Geogr., ser. Geol., 17, p. 3 - 13.

Clark, A.M. (1993) - Hey's Mineral Index. Mineral Species, Varieties and Synonyms, Chapman & Hall, London, etc., 852 p.

Codarcea, Al. (1931) - Etude geologique et petrographique de la region Ocna de Fier - Bocsa Montana, Banat, Roumanie, An. Inst. Geol. Rom., 15, p. 261 - 424.

Cotta, B.v. (1864) - Erzlagerstatten im Banat und in Serbien, W. Braumuller, Wien, 108 p.

Cressey, G. (1987) - Skarn Formation Between Metachalk and Agglomerate in the Central Ring Complex, Isle of Arran, Scotland, Mineral. Mag., 51, p. 231 - 246.

Deer, W.A., Howie, R.A., Zussman, J. (1993) - An Introduction to the Rock-Forming Minerals, 2nd ed., Longman Scientific, 696 p.

Einaudi, M.T., Meinert, L.D., Newberry, R.J. (1981) - Skarn Deposits, Econ. Geol., 75th Anniversary Vol., p. 317 - 391.

Einaudi, M.T., Burt, D.M. (1982) - Introduction - Terminology, Classification, and Composition of Skarn Deposits, Econ. Geol., 77/4, p. 745 - 754.

Esmark, J. (1798) - Kurze Beschreibung einer mineralogischen Reise durch Ungarn, Siebenburgen und das Bannat, Crazische Buchh., Freiberg, 191 p.

Estner, F. J. A. (1797) - Versuch einer Mineralogie fur Anfanger und Liebhaber, II Bd., 2 Abt., Schmidt, Wien.

Evensen, N.M., Hamilton, P.J.,O'Nions, R.K. (1978) - Rare-earth Abundances in Chondritic Meteorites, Geochim. Cosmochim. Acta, 42, p. 1199 - 1212.

Frenzel, A. (1883) - Rezbanyit, eine neue Mineralgattung, Tschermak's. Min. Petr. Mitt. 5, p. 175.

Ghergari, L., Nicolescu, S., Cociuba, M. (1986) - Manjiroite and Todorokite from Ocna de Fier, Romania, Studia Univ. Babes-Bolyai, Geol.-Geogr., 31/2, p. 3 - 11.

Goldschmidt, V.M. (1912) - Die Kontaktmetamorphose im Kristianiagebiet, Skrifter Videnskapsselskapet i Kristiania, I, Mat.-Naturvidensk. Klasse, 483 p.

Gruescu, C. (1975) - Co-axial quartz twin crystals with radiate concrescence in the deposit Ocna-de-Fier, Banat, Rumania, Acta Miner. Petr. Szeged, 22/1, p. 143 - 146.

Hermann, R. (1858) - Untersuchungen einiger Wismutherze, sowie uber Oxysulphuret von Wismuth. 2. Ueber Rezbanyit, J. prakt. Chem., 75, p. 450 - 452.

Ilinca, G., Marincea, S., Russo-Sandulescu, D., Iancu, V., Seghedi, I. (1993) - Mineral Occurrences in Southwestern Banat, Romania, Rom. J. Mineral., 76, suppl. 2, 38 p.

Karsten, D.L.G. (1800) - Min. Tab., 1st ed., p. 26, 71.

Kissling, Al. (1967) - Studii mineralogice si petrografice in zona de exoskarn de la Ocna de Fier (Banat), Ed. Academiei RSR, Bucuresti, 172 p.

Koch, S. (1930) - Einige Wismuthmineralien aus dem Banater Kontaktgebiet, Centralblatt f. Miner. Geol. Pal., A, p. 49 - 56.

Koch, S. (1948) - Bizmutasvanyok a Karpatmedencebol, Acta Miner. Petr. Szeged, 2, p. 1 - 23.

Krenner, J.A. (1884) - Mineralogische Mitteilungen (Bismutin von Moravicza), Zeit. Krist., t.8, p. 537.

Krenner, J.A. (1925) - Warthait, egy uj asvany Magyarorszagbol, Math. term.-tud. Ert., 42, p. 4 - 5.

de Launay, E. (1913) - Traite de Metallogenie, vols. 1 - 3, Paris.

Leman, J. (1818) - Nouv. Dict. Hist. Nat., 20, p. 31.

Lindgren, W. (1933) - Mineral Deposits, 4th ed., McGraw-Hill, New York, 930 p.

Marza, I. (1979) - Prezenta unor pachiodonte fosile in skarnul granatifer-diopsidic-epidotic din depozitele pirometasomatice de la Ocna de Fier, Romania, Rev. Roum. Geol. Geophys. Geogr., ser. Geol., 23, p. 117 - 120.

Marza, I., Bedelean, I. (1978) - Contributii la cunoasterea pseudomorfozelor din zacamantul pirometasomatic de la Ocna de Fier (Banat), Studia Univ. Babes-Bolyai, Geol.-Geogr., 23/1, p. 11 - 18.

Marza, I., Ghergari, L., Nicolescu, S. (1990) - Mushketovitization in Romanian ore deposits, Studia Univ. Babes-Bolyai, Geol., 35/1, p. 57 - 67.

Meinert, L. (1992) - Skarns and Skarn Deposits, Geoscience Canada, 19/4, p. 145 - 162.

Muresan, I., Tirlea-Hotiu, I., Cimpean, A. (1990) - Minerale din Romania descrise prima data in literatura mondiala (consideratii generale), Studia Univ. Babes-Bolyai, Geol., 35/1, p. 3 - 33.

Nambu, M. Tanida, K. (1968) - Manjiroite, a New Manganese Dioxide Mineral, from Kohare Mine, Iwate Prefecture, Japan, J. Jap. Assoc. Miner., Petrol., Econ. Geol., 58, p. 39 - 54 [Japanese with English summary], abs. in Amer. Mineral., 53, p. 2103.

Nicolescu, S. (1982) - Studiul geologic si petrografic al zonei Ocna de Fier - Dognecea, cu privire speciala..., Unpublished BSc thesis, University of Cluj, 112 p.

Nicolescu, S. (1995) - Major and Rare Earth Element Distribution in Birefringent Grandite from Ocna de Fier, Banat, Romania, (abs.), Rom. J. Miner., 77/suppl. 1, p. 33.

Nicolescu, S., Marza, I. (1989) - On the History of Ore Deposit Studies in Romania. I. Bernhard Cotta, Studia Univ. Babes-Bolyai, Geol.-Geogr., 34/1, p. 93 - 99.

Nicolescu, S., Cornell, D.H. (1996) - Skarn Genesis at Ocna de Fier, Romania (abs.), Abstracts vol., The 22nd Nordic Geological Winter meeting, Turku (Abo), Finland, p. 145.

Niedzwiedzki, I. (1873) - Zur Kenntniss der Banater Eruptivgesteine. Gestein von Dognacska, Tschermack's Min. Mitt., p. 169 - 275.

Ohmoto, H. (1986) - Stable Isotope Geochemistry of Ore Deposits in: Walley, J.W., Taylor, H.P., O'Neil, J.R. (Eds.), Stable Isotopes in High Temperature Geological Processes, Reviews in Mineralogy, 16, p. 491 - 559.

Peters, K.F. (1861) - Geologische und Mineralogische Studien aus dem sudostlichen Ungarn, insbesondere aus der Umgegend von Rezbanya, Sitzungsb. d. k. Akad. d. Wiss., Math.-Naturwiss. Kl., Theil I, 43/I, 385 - 463, Theil II, 44/I, p. 81-187.

Radulescu, D.P., Dimitrescu, R. (1966) - Mineralogia topografica a Romaniei, Ed. Academiei RSR, Bucuresti, 376 p.

Raymond, L.A. (1995) - Petrology: The Study of Igneous, Sedimentary and Metamorphic Rocks, Wm. C. Brown Publ., 742 p.

Rollinson, H. (1993) - Using Geochemical Data: Evaluation, Presentation, Interpretation, Longman Scientific, 352 p.

Schrauf, A. (1874) - Untersuchung eines neuen Mineral's, genannt Veszelyit, Anz. k. Akad. Wiss., Wien, 11, p. 135 - 137 Schrauf, A. (1880) - Uber Phosphorkupfererze Veszelyit, Zeit. Krist., 4, p. 31 - 33.

Sjogren, Hj. (1885) - Om jernmalmerna vid Moravicza och Dognacska i Banatet, Geol. Foren. Stockholm Forhandl., 93, VII/9, p. 514 - 537.

Sjogren, Hj. (1886) - Beitrage zur Kenntniss der Erzlagerstatten von Moravica und Dognacska im Banat und Vergleichung derselben mit den schwedishen Eisenerzlagerstatten, Jb. d. k.-k. geol. Reichsanst.. 36/4, p. 605 - 668.

Strunz, H. (1982) - Mineralogische Tabellen. Eine Klassifizierung der Mineralien auf kristallchemischer Grundlage, 8th ed., Akad. Verlagsg, Geest & Portig K.-G., Leipzig, 621 pp.

Stutz, A. (1793) - Neue Einrichtung der k.-k. Naturalien-Sammlung zu Wien, p. 61, 144.

Suciu, C. (1968) - Dictionar istoric al localitatilor din Transilvania, Ed Academiei RSR, Bucuresti, vol. I - 431 p., vol. II - 447 p.

Thompson, R.M. (1949) - Goongarrite and warthaite discredited, Amer. Mineral., 34, p. 459 - 460.

Tornebohm, A.E. (1875 a) - Geognostisk beskrifning ofver Persbergets grufvefalt, Sveriges geologiska undersokning, Ser. C/14, 21 p.

Tornebohm, A.E. (1875 b) - Om lagerfoljden inom Norbergs malmfalt, Geol. Foren. Stockholm Forhandl., 2, p. 329 - 335.

Tschermak, G. (1874) - Ludwigit, ein neues Mineral aus dem Banate, Tschermak's Min. Mitt., p. 59 - 66.

Udubasa, G., Ilinca, G., Marincea, S., Sabau, G., Radan, S. (1992) - Minerals in Romania: the State of the Art 1991, Rom. J. Mineral., 75, p. 1 - 51.

Vendl, Marie (1927) - Kristallographische Untersuchungen an ungarischen Calcite, Zeit. Krist., 65, Heft1/2, p. 636 - 679.

Vlad-Serban, N. (1974) - Mineralogeneza skarnelor de la Dognecea, Ed. Academiei RSR, Bucuresti, 119 p.

Zak, L., Megarskaya, L., Mumme, W.G. (1992) - Rezbanyite from Ocna de Fier (Vasko): A Mixture of Bismuthinite Derivatives and Cosalite (abs.), Amer. Minerral., 77, p. 1308 - 1309.

Zak, L., Mumme, W.G. (1994) - Rezbanyite discredited, Neues Jahrb. Min., Mh., 7, p. 314 - 316.

Zemann, J. (1992) - What is monsmedite? (abs.), Rom. J. Mineral., 75/suppl. 1, p. 49.

Last Updated: October 19, 2007

Return to:

Skarn home page