Table 1.
The generally accepted chronological phasing rests on a combination of object typologies, object assemblages in hoards and burials, and 14C ranges.
Fig 1.
Analysed objects using lead isotope analysis (LIA) per period from the Neolithic to the established Bronze Age in southern Scandinavia.
The number of analyses presented in this study is related to the published analyses [1,3,4,6–8,14,50,51] and given in percentages of the overall number of artefacts assumed for each period based on [17,48]. For NBA IB and NBA II, the chief focus of this article, the distribution of the artefact categories analysed within the present study is shown in the separate diagram.
Fig 2.
Compositional patterns of the major clusters identified among 533 samples from objects of the Nordic Bronze Age.
Plotted are the medians (black), the interpercentiles (10 and 90%) and the total ranges (minimum and maximum). Clusters #7 to #11 contained only one or two samples and are considered as outliers or were combined with a larger cluster, which is compositionally most similar. Only logarithmic concentrations of As, Sb, Ni, Ag, and Bi were used for classification with cluster analysis. The ranges of tin concentrations are merely shown to demonstrate that certain chemical groups (C#3 and C#4) were not intentionally alloyed with tin.
Table 2.
Elemental compositions (average) of the ten major clusters discussed in this study.
Table 3.
Structure of the clustered data with numbers of artefacts/members, based on the average-link cluster analysis with logarithmic concentrations of arsenic (As), antimony (Sb), silver (Ag), nickel (Ni) and bismuth (Bi).
Table 4.
Measured values of NIST SRM 981 from different literature sources compared with those obtained in this study.
Fig 3.
206Pb/204Pb and 207Pb/204Pb isotope plot of the copper flat axes discussed, with comparative plot (A) indicating the difference of measurements between this study and published ones [14], obviously problematic in their 207Pb/204Pb ratios. The ore data are from: Italy AATV mining region [32,73–75], Serbian copper deposits [72,76], Bulgarian copper deposits [26,33].
Fig 4.
Distribution of the different copper types (clusters: For definition see text) in time.
The largest variety in compositional groups is observed for the LN II period. In contrast, in the developed Bronze Age, NBA IB-II, only two different compositional types of low-impurity copper were used, and only minor amounts of fahlore copper.
Fig 5.
Chemical and isotopic evaluation of LN II and NBA IA samples in this study.
The 206Pb/204Pb and 207Pb/204Pb diagram compares the LN II artefacts and the objects from the NBA IA Virring deposit, Jutland, with the discussed copper deposits. The trace elemental comparison of Sb-As and Sb–Ag allows for identification of British/Welsh metal and highlights the probability of use of Slovakian fahlore copper. For comparing elemental with classificatory data, the Anglo-Irish axes and hybrid forms are highlighted with the respective symbols shown in the Sb-As and Sb–Ag data-plot, placed on top of the cluster symbols. The ore data are from: Mitterberg ore district [35], Hron Valley, Slovakia [27,28], Inn Valley and Buchberg, Alpine region [29,30], Great Orme mining region, Wales [36–39,83], Alderley Edge mining region, Britain [39,83], central Wales mining region [39,84].
Fig 6.
Decorated shaft-hole axes from an unknown find location, NM 541 (top) and Juellinge, Maribo Amt (NM B10106).
Photo: Heide W. Nørgaard, by permission of the National Museum, Copenhagen.
Fig 7.
Valsømagle-type axes are elegantly sculptured. Geometric decoration is rare, and the axe from Engemarken (top) is one of few artefacts with such a decoration. Top: Decorated Valsømagle-type shaft-hole axe from Engemarken, Roskilde, København municipality (B16780). Bottom: The shaft-hole axe from Vesterå, Børglum, Hjørring municipality (NM 19176), is of south-east European style and can probably be dated to the beginning of the NBA II. Photo: Heide W. Nørgaard, by permission of the National Museum, Copenhagen.
Fig 8.
The subdivision of cluster C#5 is visible in the As/Ni diagram.
The study presented used the subdivision into eleven subclusters as a guide to define these artefact groups with a significant Ni = As balance.
Fig 9.
Diagrams of the 206Pb/204Pb and 207Pb/204Pb lead isotope ratios of the shaft-hole axes of NBA IB, including comparative reasons for the shaft-hole axes analysed by Ling and colleagues [6,8].
The three distinct groups are clearly recognisable (top). Isotope group 1 matches the isotope ratios of the Italian Alps [68], both in the conventional diagram (not shown), lying in between ratios of 2.09–2.112 208Pb/206Pb and 0.855–0.862 207Pb/206Pb, and in the plot presented. Isotope group 2 has a much larger variation, 2.03–2.083 208Pb/206Pb, 0.81–0.843 207Pb/206Pb and 18.58–19.69 206Pb/204Pb, within the range of the Slovakian and East Alpine ores from Mitterberg. Isotope group 3 shows a very dense concentration and does not exceed values of 18.555 206Pb/204Pb and 15.638 207Pb/204Pb, in line with British and Welsh ores. The ore data are from: Mitterberg ore district [35], Hron Valley, Slovakia [27,28], Inn Valley and Buchberg, Alpine region [29,30], Great Orme mining region, Wales [36–39,83], Alderley Edge mining region, Britain [39,83], central Wales mining region [39,84], Italy AATV mining region [32,73–75], Valais, Switzerland [31,85]. The analytical uncertainties are shown by the size of the symbols.
Fig 10.
Diagrams of Ni/As, Sb/As and As/Ag concentrations of isotope group 2 axes.
The similarity to the Nebra hoard (data are taken from [40]) is distinct, and the Sb/Ag diagram in particular allows the exclusion of the Slovakian Ore Mountains as a possible source for the Fårdrup-type axes. Valsømagle-type axes (#102, #268), as well as the axe from Löt, Sweden (KE33–465–10), show values that rule out such a clear statement. The comparative data on shaft-hole axes from Sweden, like the double analysis from B16780, derive from the Laboratory of Isotope Geology, Stockholm [6,8]. The ore data are from: Mitterberg ore district [35], Hron Valley, Slovakia [27,28], Inn Valley and Buchberg, Alpine region [29,30]. The analytical uncertainties are comparable with the size of the symbols.
Fig 11.
Diagrams of Sb/As and Sb/Ag concentrations of isotope group 3 axes identified through lead isotope data, compared to values from the Mitterberg mining region [35] and Great Orme, Wales [36].
Highlighted is shaft-hole axe #108, which demonstrates the difficulties in allocating the copper of group 3 to one specific ore deposit and the axe from Brørup (#186) and another axe (#106) that most likely derive from Welsh metal. The analytical uncertainties are comparable with the size of the symbols.
Fig 12.
The data evaluation illustrates that Italian AATV copper may have been first used in in southern Scandinavia from NBA IB (1600–1500 BC).
Its importance increased drastically towards the mature NBA II c. 1500–1300 BC, and from 1450 BC Italian AATV copper seems to dominate the metal supply. This study shows that probably 40% of the artefacts were made of this copper, while for 23% evidence for mixing metals with other sources is possible.
Fig 13.
Diagram of Sb/Ag of the NBA IB artefacts consisting of high-impurity copper compared to the low-impurity copper C#6.
The Sb/Ag concentration suggests an origin for the fahlore copper in the Slovakian Ore Mountains, while the low-impurity copper is consistent with the tendencies discovered for the shaft-hole axes and point to East Alpine copper. The ore data are from: Mitterberg ore district [35], Hron Valley, Slovakia [27,28], Inn Valley/Buchberg, Alpine region [29,30], Great Orme mining region, Wales [36–39].
Fig 14.
Lead isotope ratios of artefacts consisting of low-impurity copper in NBA IB.
The 206Pb/204Pb/207Pb/204Pb diagram (bottom) shows that the flanged axes and spearheads are in line with the identified groups of shaft-hole axes. The Valsømagle-type spearhead #262 and the Bagterp-type spearheads #288–290 and #306 (Bagterp and Over-Vindinge) are highlighted. The ore data are from: Mitterberg ore district [35], Hron Valley, Slovakia [27,28], Inn Valley and Buchberg, Alpine region [29,30], Great Orme mining region, Wales [36–39], Alderley Edge mining region, Britain [39,83], central Wales mining region [39,84], Italy AATV mining region [32,73–75]. The analytical uncertainties are comparable with the size of the symbols.
Fig 15.
206Pb/204Pb/207Pb/204Pb diagram of NBA II artefacts analysed within this study, compared to contemporaneous artefacts.
(A) compares the ornaments and swords with the most probable Central and South European ore deposits. Consistent with contemporary artefacts with secured AATV provenance (data taken from [125]), a large group of artefacts is comparable with South Alpine Italian deposits. Artefacts of the Ølby burial, sword #273, neck collar #274 and belt plate #275 are highlighted. (B) compares NBA II artefacts with artefacts of proved Slovakian provenance (data taken from [8,50,51]). The Ølby neck collar #274 and the spearhead from Ullerslev #272 are probably made of Slovakian low-impurity copper. (C) shows that the majority of the artefacts analysed elsewhere (data from [6–8]) fall within the range of the Italian AATV deposits consistent with the data presented here. A significant group (marked) can, at the present state of research, only be defined, but not provenanced. The sheet-tubes (VM1680) from Vognserup Mose [8] and the octagonal-hilted sword MA-071206 [51] are highlighted, plotting an area where several Welsh deposits are present. The ore data are from: Hron Valley, Slovakia [27,28], Inn Valley and Buchberg, Alpine region [29,30], Italy AATV mining region [32,73–75]. The analytical uncertainties are comparable with the size of the symbols.