The measurement of several prehistoric golden objects has been proposed, not at all at random, at the beginning of the Arheomet project; these measurements are susceptible of offering information not only regarding the prehistoric societies, but as well precious comparisons for the following epochs. One of the problems raised by the elemental composition research for objects belonging to the historical heritage is the presumption that the metals have been reused, therefore re-fused and-probably-re-combined; potentially, the recurrence of the phenomena should be lower for the primitive societies, and the objects created by this society are, therefore, the most susceptible of using native metal. This is the reason why in the final section of the report (the analytical section), the measurements on the heritage objects are compared with measurements on native gold from the same geographical region. The research area itself, the way it has been defined by the archaeologists ever since the conceptual phase of the research, centralizes in the major auriferous basin from Romania's territory, on each side of the Apuseni Mountains. A number of 6 hoards (comprising 34 golden objects) and 3 pieces coming from isolated discoveries, have been selected for this phase of the project, all of the pieces dated in pre- and protohistory epochs (from the Bronze Age until the first Iron Age), part of the Romanian National History Museum's heritage. As it can be seen on the map from Figure 1, the selection of these hoards has not been done at random, considering their geographical distribution, as well as the dimension of the pieces and of the period when it is assumed that they have been manufactured (it must be mentioned that for some of these pieces it cannot be mentioned exactly the epoch, being perchance discoveries). In Figure 1 it can be observed a certain piece grouping from the distribution point of view, fact that may be correlated with the results of the analysis performed using XRF and micro-PIXE techniques, in the tentative of establishing the source areas for the raw material used at the piece manufacturing, as well as an eventual circulation of these pieces in time and space (considering the case of Central-Europe area, as well the Balkan region).

Hoards

Cacova (today Livezile, Alba county): the hoard contains 5 "loop rings" (Fig. 2) with different forms and dimensions. These pieces appear at the beginning of Bronze Age and may be found until the beginning of the first Iron Age (Hallstatt). Their more exact dating, as well as their cultural framing, is rather difficult due to the fact that they mostly appear isolated or in association with other pieces with a similar difficult dating (without clear archaeological contexts). We remind the fact that the rings from Cacova have been acquisitioned from a local individual from Aiud in 1929 (Popescu D 1956, p. 200, 203; Rusu M 1972; p. 46). The bibliography regarding the origin and diffusion of these objects is also very scattered. We may specify that this kind of pieces is rather frequent in Transylvania; for Europe we may find analogies in Hungary, Bohemia and Moravia.

Selected bibliography: Popescu D 1956; Rusu M 1972; Ghinea D 1998

• Fig. 2.1.
  Inv. 23485
  W=29.0g; D=23.4mm
  Description: Massive loop ring, with a smaller bar in the middle, which gets thicker towards the both ends. It has a circular form, with superposed ending parts. In the middle, the bar section is round, towards the ends becoming rather oval or semi-circular. The ends of the ring are rounded. It has a good preservation state, with several minor scratches, especially at the endings, on the outer surface.
  
• Fig.2.2
  Inv. 23486
  W=5.30g; L=158mm
  Description: Thick wire loop ring, manufactured by turning and hammering, and four times twisted spiral (three complete spirals, plus the endings). The section of the piece is round. The preservation state is good, in the middle (on the central loop) having an inner fold.
• Fig. 2.3.
  Inv. 23487
  W=4.76g; D=14mm
  Description: Double wire loop ring, manufactured by turning, spiral twisting, forming two complete loops, plus the endings. Preservation state is good.
• Fig.2.4.
  Inv. 23488
  W=4.53g, D=15mm
  Description: Massive loop ring, manufactured by hammering, out of bar with round section. It has a circular form, with superposed endings. The bar is thinner in the middle, getting thicker towards the endings. The preservation state is good, being slightly deformed in the middle.
• Fig.2.5.
  Inv. 23489
  W=4.40g; L=28mm
  Description: The piece is manufactured by hammering. Probably initially loop ring, at the present is broken in two pieces and twisted. It is thinner in the middle, getting thicker towards the endings, and then again thinner. The section of the bar is rather rhombic. It has a good preservation state.

Căuaş (Satu Mare county): the hoard contains two massive bracelets and four fragments that come from integral pieces. These types of pieces represent the simplest form of this kind of bracelets (they are considered bracelets only when they are large enough to be worn on the arm). The golden ones are smaller, the Hallstatt bronze type being more numerous. They have been interpreted as standard change objects (possible pre-monetary pieces; see also the results of the XRF analyse). Their cultural framing (without being a certain one) is Hallstatt A-B (Rusu M 1972, p. 34, 45).

Selected bibliography: Ghinea D 1998; Popescu D 1956, p. 200, 203; Rusu M 1972, p. 46 (36)

• Fig.3.1.
  Inv. 23474
  W=136.46g; D=65.8mm
  Description: The piece is manufactured by hammering, out of a bar with rhombic (quadrangular) section, which gets thinner towards the endings; the endings are superposed. The preservation state is good.
  
• Fig. 3.2.
  Inv. 23475
  W=137.15g; D=46mm
  Description: The piece is manufactured by hammering, out of a bar with rhombic (quadrangular) section, which gets thinner towards the endings. The preservation state is good.
• Fig.3.3
  Inv. 23476
  W=34.66g; D=46mm
  Description: Fragment of a bracelet. The piece was probably recently sectioned (at the discovery moment).
• Fig.3.4
  Inv. 23477
  W=21.99g; L=51
  Description: Fragment of a bracelet. The piece was probably recently sectioned (at the discovery moment).
• Fig. 3.5
  Inv. 23478
  W=25.70g; L=38
  Description: Fragment of a bracelet. The piece was probably recently sectioned (at the discovery moment).
• Fig.3.6.
  Inv. 23479
  W=13.85g; L=17
  Description: Fragment of a bracelet. The piece was probably recently sectioned (at the discovery moment).

Pecica-Rovine (Arad County): the hoard contains a string with empty cones (48), with a hole in the middle so they can be put together as a necklace (probably to imitate a similar string manufactured of animal teeth; see Popescu D 1956, p. 206) and a small convex phalera (button?). The pieces have been discovered by chance in 1938, in a bowl belonging to Vatina culture (middle age Bronze), bowl that contained as well two Columbella shells (Dumitrescu V 1937-1940).

Selected bibliography: Dumitrescu V 1937-1940; Popescu D 1956

• Fig.4.1.
  Inv. 11354-11401
  Wtotal=12.63g; Lcon=13mm; Dcon=6mm
  Description: Small cones necklace (48), manufactured of gold plate, discovered by chance in 1938, in a two ears clay bowl, together with a small phalera and two Colimbella shells. The 48 cones are drilled in the middle in order to be assembled as a string. The preservation state is good.
• Fig.4.2.
  W=0.21g; Dmax=13.2; Dmin=12.8 Description: Disc of small dimensions, manufactured of gold plate, with a convex form, having on the edges two perforations in order to be applied on the clothes. The piece is decorated with small prominent points, made by au repoussé technique. The preservation state is good.

Săcuieni (Bihor county): The Săcuieni Hoard has been discovered by chance in 1927, in a vineyard. From the 8 phaleras discovered then, only three have been preserved, as well as a fragment from the fourth. Although the ornamentation is similar in style and conception, at every piece the ornamental décor is different.

Selected bibliography: Ghinea D 1998; Nestor I 1933; Popescu D 1956

• Fig.5.1
  Inv. c 1744
  W=27.95g; D=90.1mm
  Description: Phalera manufactured of thin gold plate, with four perforations on the edge. The décor is made by au repoussé technique; the piece presents a central umbo, surrounded by two pointed strips. This ornamentation is followed by pointed spiralled motive, while in the exterior part there is another pointed strip. The preservation state is rather good, although a fragment from the edge is missing, near one of the perforations, place where the piece is also broken.
• Fig.5.2
  Inv. c 1745
  W=26.70g; D=80.1mm
  Description: Circular phalera manufactured of thin gold plate, with three perforations on the edge, and with a fourth close to the centre of the piece (near one of the other three). The piece presents a central umbo and ornament made by au repoussé technique. The umbo is surrounded by a pointed string. On the edge of the piece is a second pointed string. A double spiral motive, made from pointed string (three lines and in between a row of points), is between the two strings. The preservation state is rather good, although a fragment from the edge is missing, and the piece is broken.
• Fig.5.3.
  Inv. c 1746
  W=23.72g; D=80.1mm
  Description: Circular, convex phalera manufactured of thin gold plate, with four perforations on the edge. The piece presents a central umbo (more prominent than to the other pieces, with a more prominent surrounding hollow), surrounded by a strip decorated with a triangular décor. This ornament is followed by two rows of pointed spirals (three lines and in between a row of points), and a pointed strip on the edge. The ornamentation technique is au repoussé. As regarding the conservation state, the piece presents several cracks, from one of the perforations as well.
• Fig.5.4.
  Inv. c 1747
  W=1.80g
  Description: Fragment of a phalera, from the edge of the piece, with three rows of pointed strip (a row of points between two lines), in the au repoussé technique. The piece seems to have been balder than the others. As regarding the preservation condition, we have only a fragment of the edge of the piece.

Şmig (Sibiu county): The Smig Hoard, discovered in 1880, in a bronze bowl, put in turn in a clay ware. It contained chained loop rings, golden spiral wire, golden pearls, a bracelet, 26 ornamental phaleras. The analyses have been performed on the eleven pieces which came in the custody of MNIR, and which will be further described.

Selective bibliography: Popescu D 1956; Repertoriul Sibiu (Sibiu Repertory) 2003; Rusu M 1972

• Fig. 6.1.
  Inv. 47580/1-3
  H=4mm; D=4.3mm
  Description: Perforated cylindrical bead (from a total of 160).
• Fig.6.2.
  Inv. 47581
  W=0.20g; D=3.5; H=2.5
  Description: Cylindrical bead, with a longitudinal perforation. Thus, it has been wrongly considered as being manufactured of common material. The preservation state is good.
• Fig.6.3
  Inv. 47575
  W=9.40; D=63mm; H=15
  Description: Phalera from a total of 26, present in the hoard, with three different sizes. Gold plate, au repoussé modelled and decorated, with the shape of a round convex disc. In the centre of the piece is represented an umbo, delimitated by a gutter, having on both sides a pseudo-granules circle. From these, three strips with three rows of prominented points, centrifugal disposed are going towards the edge of the piece. Between the three strips are grouped three prominented buttons, disposed at the same distance from the centre. The disc is bordered with an edge of prominented points, having also four perforations which served for the attachment of the piece. As regarding the preservation state, the piece is deformed, having several crakes, also missing small fragments on the edges, while on the back side it presents impurities.
• Fig.6.4
  Inv. 47576
  W==4.50g; D=37mm
  Description: Piece of a total of 26, present in the hoard in three different sizes. Gold plate modelled and decorated in the technique au repoussé, having the shape of a convex round disc. In the centre of the piece is represented an umbo, boarded by a gutter, and also by a pointed circle towards the edge. A similar circle is situated close to the edge of the piece, having as well two perforations. As regarding the preservation state, the piece is slightly deformed (especially on the edges), having several cracks on the edges, being scratched and stroke, while on the back it presents impurities and small perforations.
• Fig.6.5
  Inv. 47577
  W=32g; Dmax=68,6mm; Dmin= 64mm
  Description: Massive bracelet, with an oval shape, made of gold bar with plan-convex (or quadrat) section, thicker in the middle, getting thinner towards the edges, which are superposed on a length of 2 cm. The piece is manufactured by hammering. The preservation state is good.
• Fig.6.6
  Inv. 47578
  W=7.85g; L=35mm
  Description: Gold bar (loop ring) which initially had a rhombic section. Subsequently, the bar has been flattened, the lateral edges remaining sharp. The golden bar is narrower at the middle of the length, getting progressively thicker, and then thinner towards the edges. It has an oval shape, being bended at the half of the length. The edges are superposed where the bar is widen, and the extremities are curved towards the interior. The piece has been manufactured by hammering.
• Fig.6.7
  Inv. 47579
  W=4.65g; D=14mm
  Description: Loop ring, manufactured of a gold bar which at the middle of the length is thinner, getting progressively thicker towards the edges, while the bar diameter suddenly decreases towards the extremities. In the median area, the bar has a round profile, whereas in the broaden area the profile is plan-convex. The ring has a round shape, the edges superposed, with the extremities curved towards the interior. The piece is manufactured by hammering. At the half of the length, the piece is twisted. As regarding the preservation state, the piece presence strike traces.
• Fig.6.8
  Inv. 47582
  W=4.00g; D=25.5mm
  Description: Circular ear-ring, manufactured by hammering and turning, of golden bar with quadratic section. The bar is twisted until the edges get close; the edges remain smooth and superposed. The bar has a maximum profile at the half of the length and decreases progressively towards the edges, which become almost sharp. As regarding the preservation state, the piece is slightly deformed, with traces of metal testing.
• Fig.6.9
  Inv. 47583
  W=4.00g; D=25.5mm
  Description: Circular ear-ring, manufactured of golden bar with quadratic section, by hammering and turning. The bar is twisted until the edges get closer; the edges remain smooth, being superposed. The bar has a maximum profile at the half of the length and decreases progressively towards the edges, which are waned. The preservation state is good.

Tăuteu (Bihor county): the hoard comprises 6 pieces, of which 5 "Brăduţ type rings" (Popescu D 1956, p. 228), discovered in a bronze deposit.

Selective bibliography: Popescu D 1956

• Fig.7.1.
  Inv. 8992
  W=10.20g; D=55mm
  Description: Ring manufactured by hammering and incision. It comes into the indented rings category, Bradut type (after D. Popescu). The preservation state is good.
• Fig.7.2.
  Inv. 8993
  W=7.20g; D=34mm
  Description: Ring manufactured by hammering and incision. It comes into the indented rings category, Bradut type (after D. Popescu). The preservation state is good.
• Fig.7.3
  Inv. 8994
  W=5.90g; D=32.2mm;
  Description: Ring manufactured by hammering and incision. It comes into the indented rings category, Bradut type (after D. Popescu). The preservation state is good.
• Fig.7.4.
  Inv. 8995
  W=3.05; D=24.3mm;
  Description: Ring manufactured by hammering and incision. It comes into the indented rings category, Bradut type (after D. Popescu). The preservation state is good.
• Fig. 7.5.
  Inv. 8996
  W=2.40g; D=21.2mm;
  Description: Ring manufactured by hammering and incision. It comes into the indented ring category, Bradut type (after D. Popescu). The preservation state is good.
• Fig. 7.6
  Inv. 8997
  W=1.20g; L=10mm;
  Description: Edges of rings (2 pieces) manufactured by hammering and incision. It comes into the indented rings category, Bradut type (after D. Popescu). As regarding the preservation state, they have been fragmentary preserved.

Isolated pieces:
Apoldu de Sus (Sibiu county): two Brăduţ type loop rings. The pieces have been assigned to Wietemberg culture (middle bornze age-late bronze age).

Selective bibliography: EAIVR; Florescu G 1937-1940; Ghinea D 1998; Ghinea, Ghinea 2000; Horedt K 1956; Popescu D 1956; Rusu M 1972, Repertoriul Sibiu (Sibiu Repertory) 2003

• Fig.8.1
  Inv. 47589
  W=5.70g
  Description: Two loop rings, Bradut type, assembled together. Massive rings, with superposed edges, but not joint together; manufactured of bar with round section. They are thinner in the middle and get progressively thicker, only to get again thinner towards the edges, which are pretty sharp. Probably they have been manufactured by hammering. The preservation state is good, but they present several scratches.

Biia (Alba county): this kind of wares are rare in Transylvania, being usually connected with pieces of this kind from the end of Bronze Age-beginning of Hallstatt. The shape of the Biia ware is slightly similar with a leather "bag", being different of such similar wares due to the two bails, which end in double spiral. The ware is assigned to "north-Thracic circle" (Popescu D 1956, p. 234; Schroller H 1925, p. 114)

Selective bibliography: Popescu D 1956, p. 200, 233-234; Schroller H 1925, p. 114

• Fig.8.2
  Inv. 47584
  W=143.92g; Dsup=98mm; H=58mm;
  Description: Bowl manufactured of gold plate, by hammering, decorated in au repoussé technique. It has a semi-spherical body, flattened and the edge turned over the exterior. Two handles detach from the edge, arching towards the base; they are not jointed to the body of the bowl, ending in a double volute. A pointed zigzag décor is placed on the edge of the bowl. Two horizontal rows of prominences are placed at the maximum diameter of the bowl. Two prominenced circles and 5 groups of 3 concentric circles with umbo are placed on the bottom of the bowl. The preservation state is rather good, even though the bowl is slightly deformed.

Gubaucea (Dolj county):

• Fig.8.3
  Inv. 9002
  D=49mm; W=9.95g
  Description: Bracelet of gold bar, with round section, manufactured by turning. The continuous wire is twisted in order to be double, and presently is twisted making two almost complete volutes. The preservation state is good.

The objects (golden hoard pieces and coins) studied in the present research phase have been analysed through the X-Ray Fluorescence (XRF) method, using isotopic excitation sources: 238Pu (annular, 30 mCi) and 241 Am (annular, 10 mCi, with nickel opening). The plutonium source (238Pu) is better indicated for the Cu, Au, Fe and Ob detection, while the source of americium (241Am) is better indicated for Ag, Sn and Sb detection. A Si(Li) as well as a Ge hyper-pure detector have been used for detection. The later presents a high resolution (155 eV for K line of the Cu of 8.047 keV). Despite the excellent resolution, the egress of germanium K rays from the detector makes difficult the X-ray spectrum (Cojocaru, Pincovschi 2002, p. 35.).

For the analyze of the results, it has been used a distribution graphic (Fig. 9)

On top of the graphic are marked, with orange, the main sequences of gold presence, apparently contingent, being separated, on the left, a group with the pieces that have a low gold presence (72-81%; group I), and a second group, with gold in composition between 83-87% (group II). On top of these values (approximately 90% or more), we have four results with a bigger digits' dispersion, with different traces, which don't represent a "group", under the homogeneity aspect, but which, according to the information offered by our colleagues from IFIN, seem to have in common the metallurgical processing (in other words, a native gold refining), which we consider group III. It must be kept in mind though that there are gold samples from rivers that have similar values (Valea Oltului, Valea Arieşului), with values for gold of over 90%.

The first thing observed on the examination of the graphic, is that the pieces from the Căuaş hoard concentrate without exception in group II. All the pieces from Căuaş have copper traces, especially tin, in a bigger proportion than the medium, and the samples' homogeneity suggests a unique gold source. From the native gold samples, offered by the IFIN specialists, the most appropriate are the ones from Ruda-Brad (Au 84,3%; Ag 15,6%; Cu 0, 0186%), Muşariu-Brad (Au 83,5%; Ag 16,4%; Cu 0,015%) and Bradisor-Brad (Au 83,1%; Ag 16,82%; Cu 0,02%), very homogenous and, at the same time, similar to the measurements of the Cauaş hoard (with the mention that the later have a higher copper presence, approximately 0,04%). The presumption makes sense, historically and economically speaking, due to the fact that, of the Apuseni Mountains' auriferous basins, this is the only one oriented towards west, being on the Crişul Alb River, upstream of Baia de Criş (name with an intrinsic signification, in direct connection with the auriferous resources).

On contrast, the Săcuieni pieces, present as well in Crişana, all part of the generic group I (distinctly therefore from the pieces from Căuaş), are of different factures, despite their formal resemblance, and may come from three different sources (one certainly distinct, with copper and tin maximum values, with the inventory c 1746). Historically speaking, the Săcuieni group didn't probably beneficiated of an auriferous source control, the phaleras being manufactured of gold collected from different sources, probably rather difficult. Due to the fact that on the lot of measurements of the Apuseni Mountains native gold does not exist such high copper concentrations (approximately 1%), we must assume that the gold doesn't come from the surrounding area, and, most probably, the pieces have not been manufactured locally, being present here due to the prestige changes, or with tribute value, from the areas situated far north-west, from a perimeter which, eventually, might be subsequently established. We have to mention that the historical critique of these Sacuieni phaleras supposes, on the basis of a purely formal critique (shape, decor), that they come from the northern Germany or even Ireland (of course, presumption less probable for the pieces, but possible for the raw sources, see Popescu D 1956).

The pieces from Tăuteu, although they are part of the IInd and IIIrd groups (thus fundamentally different), seems to have as well a source outside the Apuseni Mountains, from the same reason, of copper residues presence much higher than those measured on Transylvanian native gold.

A preliminary conclusion regarding the hoards from Crişana would be the fact that these communities, although in the neighborhood of the native gold sources from Apuseni Mountains, didn't have a permanent control over these resources, being constrained to procure their "royal" material, the gold (sign of "sovereignty") from other, distant sources. The historians will have to identify the competitor communities, eventually from the eastern slope of the Apuseni Mountains, of course, after a confrontation with the chronological data, on the basis of formal critique.

As regarding the south Transylvanian hoards, it must be mentioned from the beginning that their high number reveals a significant power centre which manifested its will a long period of time (lasting at least during the Bronze Age), succeeding to gather an impressive quantity of prestige products. On the other hand, the relative dispersion, in the inner area, discloses a conflict over the power, which has been successively distributed to several families (respectively geographical locations). Moreover, the position of this power centre makes possible the access to different sources, such as the gold from Apuseni, from the Olt rift, from the possible sub-Carpathian sources, and even Balkan gold, this doesn't excluding the distant sources.

A very interesting example is the Şmig hoard, which has pieces in all the three major groups. This leaves no doubt regarding the distinct sources, and historically speaking-regarding the distinct moments when he was constituted, probably during several generations. We observe that the graphic (Fig. 9) begins and ends with the Smig pieces, distributed especially in the IInd and IIIrd groups. The first enlists in the samples taken from Valea Morii (west from Câmpeni, therefore quite far); the second piece (47589) could come from Roşia Montană, eventually the third the same (47575); the group of three pieces from the middle of the graphic has high values for copper traces, hence, again, could come from gold of Occident origins (?); the bracelet 47557 could come, doubtless, from Runculeţ-Straja (where the copper traces are obvious, though small). In conclusion, the Smig hoard has been constituted in time, ending to be part of the hoard through the methods specific to the epoch (gift changes or tribute), coming from very different sources. The bowl from Biia, one of the early pieces of this kind, seems to be unique within the studied group, with a small probability of origin detecting, only through an eventually subsequent coincidence.

The Cacova rings probably have the same source, being most likely manufactured at the same time. From the existing comparative data it cannot be certified the Transylvanian origin; therefore it must be looked in other place.

The Gubaucea bracelet has, in turn, a typical composition for the gold from Apuseni Mountains, the most appropriate digits being at Runculeţ-Straja.

A final mention is necessary: the gold source from Runculeţ-Straja is positioned on the left side of the inferior stream of Mureş River, therefore in the proximity of the power area around today Sibiu.

The suspicions' recurence regarding the provenience from this source is thus according to the geographical data as well (see once more the map from Fig. 1)