Hi Ned,

I am an amateur in this field and am following the discussion since this is
an enormously learning experience for me. THank you for your comments on
the subject of steel from Sri Lanka. The work by Gill Juleff and others
are infact on the basis that it was crucible steel (Wootz??) that was
produced at the Samanalawewa sites from Sri Lanka.

I am giving below a review of Gill Juleffs book Early Iron and Steel in Sri
Lanka by David J Killick, Associate Professor, Department of Anthropology,
University of Arizona, Tucson, AZ, USA. This review is obtained from the
following site.

http://users.ox.ac.uk/~salter/arch-metals/met-review.htm

Hope this can stimulate further discussion.

Anan.

1367
David Killick on Gillian Juleff's "Early Iron and Steel in Sri Lanka"
(Mainz: Verlag Philipp von Zabern, 1998 [ISBN 3-8053-2512-6].
This is a weighty large format tome (422 pp), beautifully produced with
many monochrome and full-colour photographs.

Although I have been following this project closely, I am still blown away
by the quality of this volume and would urge anyone at all interested in
archaeometallurgy to read it closely. I think that it is particularly
notable for its exemplary integration of field survey, excavation,
documentary and oral history, experimental archaeology and archaeometry.

The volume describes the discovery, during survey of a valley to be flooded
by construction of a large dam, of two features of interest. The first was
the site of the crucible steel production famously described by
Coomaraswamy in 1904. Juleff found that the descendants of those
steelworkers still possessed some blooms, crucibles and ingots of crucible
steel, and an excellent metallographic study of these by Michael Wayman is
included here as an appendix. The second feature was the discovery of an
entirely new type of iron-smelting furnace. As reconstructed by Juleff (and
the data presented here allow no doubt as to the accuracy of her
reconstruction) these were low subrectangular structures, 1.5 - 2 m in
length, 0.4-0.8 m wide and (particularly suprising) only 0.5 m high. Large
numbers of these were found, invariably placed near the crest of
west-facing hills, with the front long wall, bearing a single line of up to
a dozen tuyeres, facing downslope. Juleff argued that these were wind
powered furnaces utilizing the force of th seasonal monsoon (July to
September), which (as she shows in an innovative chapter packed with
wind-velocity measurements) achieve sustained wind speeds of 40 km/h, with
periodic peaks up to 60 km/h.

Since Juleff was not an archaeometallurgist (at least not yet!) and there
was no precedent for the technology that she proposed, her reconstruction
encountered intense scepticism from the archaeometalurgical community. She
countered this in the most effective way - by building full-scale replicas
and smelting iron in them successfully on four separate occasions, using
only the force of the monsoon wind. There can be no doubt that she is
correct and that the Sri Lankan furnaces, for which available dates run
from the seventh through the eleventh centuries AD, are a significant new
chapter in the history of metallurgy. Mathematical modelling of the
windflow patterns by David Wilson, an aeronautical engineer, explains why
these furnaces work. A complex pattern of boundary layer separation occurs
where the pasees over the lip of the front walls, producing a low pressure
zone that draws air in through the tuyeres. This is NOT a natural draft
furnace - Wilson's calculations suggest that the pressure drop achieved in
these 0.5 m furnaces is equivalent to that in natural draft furnaces 3 to 6
m tall.

This is the kind of publication that sets new standards for an entire
field. The quality of the fieldwork is very high, it is superbly
documented, and it is all woven into a complex and extremely coherent
argument. Furthermore, unlike much contemporary archaeometallurgy (and I am
thinking here particularly of European and Latin American
archaeometallurgy) this study stands out for its wide-ranging use of
comparative material - African, European, Near Eastern, Indian and
Japanese. In summary, this is about as good as it gets in our field.

----------------------
David J Killick
Associate Professor, Department of Anthropology
University of Arizona, Tucson, AZ, USA


>It is quite possible that some Damascus swords were made from Sri Lankan
steel,
>but they would not have been of top quality. The two essentials of
suitable steel
>are freedom from slag inclusions, and high carbon content in the order of
1.5 to
>1.8 percent. The former requires a crucible steel such as Wootz which was
normaly
>used, and not a bloomery steel.
> Ned.
>