Archaeologia Islandica - 01.01.2007, Blaðsíða 70

Arne Espelund
Figure 10. The author visiting the site Sandartunga in 2001. Photo by Jón Halfdanarson.
extraction of iron. One is apt to think that
the corresponding metal would have a
high content of carbon. By means of a
magnet also a small piece of metal was
retrieved and analysed (figure 11; table
9). This again is an excellent metal: low
in detrimental phosphorus and sulphur, as
well as slag (represented by Mn and Si).
Dating
Margrét Hermanns-Auðardóttir collected
samples of charcoal from the slag heaps
during our round trip and had them 14C-
dated at the Laboratory for radiological
dating in Trondheim. In a report dated Nov.
25th 1997 the following results were given
(table 10). It should be noted that pieces
of charcoal were taken from the top of the
slag heaps, probably representing the last
period of ironmaking. As expected, only
birch was used for iron production. In con-
trast to pine (often used in Scandinavia)
birch usually gives reliable values. The
results from Fnjoskadalur and Ormstadir
are comparable, representing ironmaking
in the period 1020-1275 while Geirstadir
is a bit older and Refsmyri much younger,
near the end of ironmaking in Iceland
(Fridriksson & al 1992). The results and
the technique are in full agreement with
contemporary iron production in Norway,
when roasting and furnace construction
are left out. A continuity during the rel-
evant period is evident.
Production
Relative to a certain amount of slag, say
10 kg, we are interested in the corre-
68