A WEAPON FROM THE TURN OF THE EPOCHS A UNIQUE SPATHA FROM LAKE NIDAJNO IN PRUSSIA

: This paper discusses the results of new technological examinations of a spatha blade from a bog sacrificial place in Lake Nidajno, Czaszkowo (Zatzkowen), Mrągowo District, Prussia, PL. The site can be dated to the turn of the Late Roman Period and the Migration Period and its origin may have been related to interactions between Germanic, Balt and Black Sea and North-Eastern Mediterranean cul-tures, and perhaps to migrations of the Galindians to Southern Europe and back. The archaeometallurgical examinations demonstrated that the blade had been manufactured using a complex pattern-welding technology. As a result, a weapon which possessed both high combat values and unique aesthetic traits was produced. The blade itself may be of Roman provenance.


Introduction
The aim of this paper is to discuss the results of repeated technological examination of a unique spatha blade from a bog sacrificial place in Lake Nidajno, Czaszkowo (former Zatzkowen), Mrągowo District, Prussia, now within the borders of Poland (Fig. 1). This artefact and other sword finds from this site have already been analysed with regard to their technology of manufacture. 1 However, a new analysis provided new pieces of information which result in a better understanding of the manufacturing technology of this extraordinary weapon. Technological examinations were carried out by Dr Janusz Stępiński, to whom the author is indebted for his kind help.

Find context
The discussed sword blade 2 was discovered as part of a bog sacrifice from the turn of the Late Antiquity and the Migration Period. The deposit was composed of over 400 artefacts made from iron and other metals, including precious ones. These were mostly ritually destroyed (burnt and bent or broken) spearheads, sword blades, mail, as well as numerous ornaments and fittings made of bronze, silver and gold. As far as the cultural pertinence of these finds is concerned, some of them can be related to the local milieu, that is, the Bogaczewo Culture and the Olsztyn Group, while others display Germanic, Black Sea or Mediterranean traits. It is believed that this site is an example of a result of migrations and other types of interactions between Germanic, Balt and Black Sea and North-Eastern Mediterranean cultures, with special reference to a theory of migrations of the Galindians to Southern Europe and back. 3

Spatha Cz-P55 and its manufacturing technology
Among the finds from Czaszkowo there were five swords blades in all. Due to its extremely poor state of preservation (complete mineralisation), nothing can be said on the technology of manufacture of Sword Blade CZ/29/1 (cat. no. 1.1.1.2). Furthermore, metallic parts which survived in the cores of Sword Blades CZ/XVIII/10 (cat. no. 1.1.1.5) and CZ/33/10 (cat. no. 1.1.1.3) are too small to render any more accurate assessment possible. As regards Sword Blade CZ/28A/10 (cat. no. 1.1.1.1), its core was manufactured from at least two rods of low-carbon steel, which would suggest that it belongs to the technological Group B.II (blades made from rods). 4 The state of preservation of Sword Blade CZ-P55 (cat. no. 1.1.1.4) is far from satisfactory, too, and hardly any typochronological assessment is possible. Only the upper and a portion of the central part of the blade have survived (Fig. 2). On the other hand, it is preserved in a much better condition than the other swords from this site. Thus, much more can be said regarding its technology of manufacture.
A wedge-shaped sample (c. 18 × 2.5 mm) was taken for metallographic examination from the central part of the blade. The sample encompassed a half of the blade's cross-section. It was decided to apply this rather strongly invasive procedure, ast it allowed for a comprehensive of the blade's technology. Taking a small sample from the point of the blade would have obviously been less destructive, but the amount of information gained in result of such an approach would have been much smaller. The gap after sampling was filled with Axson resin and the surface of the fill was covered with Paraloid conservation layers. The 4 Żabiński et al. 2014, 139-143, Fig. 49;Żabiński 2016a, 88- aim of the metallographic examinations was to identify the material used for the manufacture of the blade and to analyse the manufacturing technology.  General view (Sides A and B), and a close-up on Side B. Traces of pattern-welding (stripes and herringbone) can be seen.

Research methods
The sample was mounted in resin and then it was ground and polished with the use of diamond paste of different grits. Its surface was etched with 4% nital in order to reveal the microstructure of the sample. Microstructure observations were carried out using a Leica DMLM optical microscope. A qualitative X-ray microanalysis of slag inclusions was done using a Hitachi S-3500N scanning microscope, equipped with an EDS-type spectrometer, with accelerating voltage of 20 kV. The content of carbon in the examined sample was approximately assessed on the basis of microscopic observations. Hardness tests were carried out using the Vickers method with a load of 10kG (98N).

Metallographic examinations
The macroscopic image of the sample's surface with spots of microscopic observations (1-4) and a schematic distribution of structural components and HV10 hardness tests can be seen in Fig. 3:a-b.
The examinations revealed that the blade was composed of several parts of different material. In the strongly corroded edge of the blade the microstructure changes from ferritic with tertiary cementite precipitates  to ferritic-pearlitic one with the carbon content of 0.2-0.3% C, which corresponds to soft steel .
The core of the blade was composed of three pattern-welded pieces of metal, marked as A, B, and C In the pattern-welded Pieces A and C the layers of iron and steel go straight from one to the other surface of the flats. However, the pattern-welded Piece B was twisted. Therefore, positions of individual layers in relation to the flats of the blade vary ( Fig. 6:a). Due to this diversified alignment of the layers in Pieces A, B, and C, different patterns can be seen on the surfaces of the flats. Pieces A and C produce a banded pattern, while Piece B produces a herringbone pattern. The microstructure of the iron layers is marked with a local presence of coarse grains, etching pits, ghost structures and phosphide precipitates (Figs. 5:a-b,e-f, 6:b-d, 7:d-f and 8:d-e). All these traits demonstrate an increased content of phosphorus in the iron layers of the pattern-welded blade. On the other hand, in the layers of soft steel one can see a partial degeneration of pearlite colonies . This can also be seen in the edge of the sword  and is related to prolonged heating of the blade and to the local presence of phosphorus in the metal.
As it can be seen on the surface of the sample, the surface of the blade is strongly damaged by corrosion. Large subsidences of metal are noticeable in almost every part of the blade (Figs. 3:c, 4:d, 6:a and 8:a).
Slag inclusions in individual parts of the blade are tiny and diversified with regard to their shape. Most of these are monophase. Multiphase slag inclusions can Fig. 4. Spatha CZ-P55, Czaszkowo, Mrągowo District, Prussia, microstructure of the edge and of the pattern-welded Piece A: a -surviving fragment of the edge neighbouring the pattern-welded Piece A, Zone 1b in Spot 1; b -ferritic-pearlitic microstructure in the edge of the sword, Zone 1b in Spot 1; c -multiphase slag inclusions and degenerated pearlite against the background of ferrite in the edge of the sword, Zone 1b in Spot 1; d -iron (ferritic) and steel (ferritic-pearlitic) layers in the pattern-welded Piece A in Spot 2. Photo J. Stępiński.
only be found in the edge of the sword, alongside with monophase ones (Fig. 4:c).
On the basis of the EDS X-ray microanalysis of the slag inclusions in the edge of the sword it can be said that globular parts in the multiphase slag inclusions contain Fe and O. Thus, these are wüstite precipitates. On the other hand, the background of these inclusions is silicate and is composed of such elements as Si, Fe, and O ( Fig. 9:a-c). Monophase slag inclusions in the edge of the sword are also of a silicate type. Apart from Si, Fe and O, they also contain Al, Ca and K (Fig. 9:d-f). The EDS X-ray microanalysis of slag inclusions in the layers of soft steel in the pattern-welded Pieces A, B, and C demonstrated that they contained such elements as Si, Fe, O, Al, Ca, and K, and sometimes also P or Mn. Therefore, these inclusions are also of silicate type (Figs. 10:a-c, 11:a-c and 12:a-c). The same chemical composition can also be found in the analysed slag inclusions in the iron layers in the pattern-welded Pieces A, B, and C. However, in this case phosphorus is present in almost every slag inclusion ).
Hardness tests on the surface of the sample yielded the following values: in the edge of the blade -78 HV10; in Piece A -140 HV10 in the iron layers and 114 HV10 in the steel layers; in Piece B -145 HV10 in the iron layers and 101 HV10 in the steel layers; in Piece C -150 HV10 in the iron layers and 99 HV10 in the steel layers.

Technology
The examined sword blade is composed of rods. It was probably forged from five rods of pre-manufactured Fig. 5. Spatha CZ-P55, Czaszkowo, Mrągowo District, Prussia, microstructure of the pattern-welded Piece A: a -iron layer, Zone 2a in Spot 2; b -microstructure of the iron layer: coarse-grained ferrite, slag inclusions and phosphide precipitates, Zone 2a in Spot 2; c, d -ferritic-pearlitic microstructure and slag inclusions in the steel layer, Zone 2b in Spot 2; e -another example of a steel layer, Zone 2c in Spot 2; f -iron layer adjacent to the steel layer in Fig. 3e, ghost structure and some carburisation can be seen, Zone 2c in Spot 2. Photo J. Stępiński. pieces of metal obtained in bloomery process in which ores with and without phosphorus were used. The edges of the sword were in all probability made from soft steel which was perhaps evenly carburised up to 0.2-0.3% C. However, it should be remembered that the sword blade went through the funeral pyre which may have caused partial decarburisation of the metal. Furthermore, the examined edge is strongly damaged by corrosion and it survived in a vestigial state only. Three pattern-welded rods were used for the manufacture of the blade's core. Each of these was composed of nine alternate layers of phosphorus-rich iron and layers of soft steel. The central piece was additionally twisted, which resulted in a different pattern (not going from flat to flat) in the cross--section of the blade. The presence of material with different contents of carbon and phosphorus in the pattern-welded pieces produced a well-visible pattern in the cross-section of the blade. A considerable hardness of the iron layers in pattern-welded pieces of the metal is related to the presence of phosphorus, which is also implied by the results of slag inclusion analyses. Phosphorus chiefly dissolves in ferrite and hardens it considerably.

Conclusions
The sword blade from Czaszkowo we have discussed deserves interest in every respect. Its sophisticated manufacturing technology, that is, pattern-welding, demonstrates that the blade was made by a highly competent specialist and was of extraordinary utilitarian and aesthetic quality. Due to the state of preservation of the artefact (prolonged heating in the funeral pyre and very advanced corrosion of the edge parts), it is not possible any more to state whether the blade underwent any other procedures, such as thermal treatment (e.g. quenching). It cannot be excluded that it was manufactured in the territory of the Roman Empire; regrettably, this cannot be proved unequivocally. Although several blades which were made in a similar manner can be pointed out, no exact technological counterpart is known to the author. Due to the state of preservation of the artefact, no typological classification is possible and its proposed chronology must remain within a general framework of the Late Roman and the Migration Period.
As stated above, it cannot be clearly determined whether the discussed sword was in use in the Balt, the Germanic or other cultural milieu. It seems, however, that the very fact of the discovery of this blade and other swords in Lake Nidajno suggests that swords were not as rare in the Balt zone during the Roman Period as it has been previously assumed. Fig. 11. Spatha CZ-P55, Czaszkowo, Mrągowo District, Prussia, slag inclusions in the pattern-welded Piece B with results of the EDS point analysis in spots shown in SEM microphotos: a -slag inclusions in the steel layer analysed in Points A and B; b -EDS spectrum from Point A, peaks from Si, Fe, O, Ca, P, Al, and K can be seen; c -EDS spectrum from Point B, peaks from Si, Fe, O, Al, Ca, and P can be seen; d -slag inclusions in the iron layer analysed in Points C and D; e -EDS spectrum from Point C, peaks from P, Fe, Si, Ca, O, Mn, Al, and K can be seen; f -EDS spectrum from Point D, peaks from Si, Fe, O, P, Ca, Al, and K can be seen. Photo J. Stępiński.