ancient metalworking technology

Talking torcs…two years on!

Almost to the day two years ago, we blogged here about a day spent at the National Civil War Centre looking at the Newark torc. What started out as a replica commission has, over the last two years, turned into a wider investigation into the technology and method of manufacture of later Iron Age gold torus torcs and which has produced results that no-one, least of all us, expected to find. It seems fitting to share a summary of these results here, two years after our quest began.

The Newark torc

When asked to replicate something, one of the first things you have to do is to try to understand how it was made – the accepted theory for these types of torcs is that wires were hammered and twisted into a neck ring, before having the hollow terminals cast on to the wires, using a process called lost wax casting.

However, having looked at the Newark torc, we both felt that the cast-on terminal construction method was unlikely as a fault in the process – common in such a difficult manufacturing method – would potentially force the entire procedure to be re-started. In such a precious and rare metal as a high-content gold alloy, this would clearly be undesirable. So our initial thought was that the terminals were cast separately, before being soldered to the neck ring. But it would only be by looking at torcs that we would find out what was really happening.

The Newark torc is complete which, although making it very beautiful, means it is very difficult to see inside and we could not test whether our ‘terminals cast separately’ theory was borne out in the evidence. What we needed was a torc of a similar type….but broken! A quick search soon provided such a thing, in the shape of the Netherurd torc terminal. Found in Peebles in 1806, and housed in the National Museum of Scotland this lone terminal had been removed from the neck ring in antiquity and offered the chance to look inside it. So, off to Scotland we went!

The Netherurd Torc terminal

We expected to see dendrites, dribbles and other signs of casting. What we did not expect to see was evidence of repoussé- many indentations on the interior showing the opposite of the relief decoration on the exterior! The Netherurd terminal was not cast, or cast on: it was made from gold sheet, hammered into shape with the decoration added from the rear before being finished and chased on the surface of the torc. In addition, there was evidence of seams marking the joining of several metal sheets, which appeared to show that it had been constructed from a donut shaped piece of sheet gold which then had a central ‘apple core’ (we do like our food metaphors!) of material added to create the central hole. Only after this was the collar fixed to the terminal, and the complete terminal then soldered to the wire neck ring.

A working model of the opened Netherurd terminal, showing its method of manufacture.

This was not at all as we expected. Furthermore, if Netherurd was made of sheet, were any of the others? Newark we could not be sure of but it soon became apparent we needed to look at more torcs! So it was back to London and to The British Museum to look first at Sedgeford, and then at the Snettisham Grotesque Torc.

The Sedgeford terminal

Examination of Sedgeford made clear that this torc was cast and it looked as if the terminals were cast on. However, recent x-ray work by Dan O’Flynn of the British Museum Science Department has shown that in fact the Sedgeford torc was manufactured using separately cast terminals, attached using a very precise fixing method of snuggly fitting cast sections secured by rods and possibly, although this has yet to be confirmed, without the use of solder.

The Grotesque torc from Snettisham

The Grotesque torc, although known to be sheet is assumed to be much earlier than the other hollow torus torcs. However, it too showed evidence of a donut construction method, with a centrally added ‘apple core’ as seen in Netherurd. So we now had Netherurd and the Grotesque showing the same technique, Sedgeford as cast, and Newark as still uncertain.

At this point, we were offered help by the Materials group at The National Physical Laboratory, who subjected the Newark Torc to X-ray Computed Tomography (XCT). Although the results, due to the difficulties of working with such a dense material as gold, were not as clear as we’d hoped, there was enough evidence to show that the Newark torc was also constructed in sheet, with 3D microscopy showing evidence of metal smearing over the area where we would expect the ‘apple core’ join to be.

Detail of the Snettisham Great Torc

The list of sheet work torcs kept growing: Netherurd, the Grotesque and now Newark. Where next? The next obvious torc target was one of the grandest gold torcs in the world: The Snettisham Great Torc. Weighing it at 1084g of 22kt gold, this torc was found by farmhand, Tom Rout, on the Ken Hill estate in Snettisham in 1950. Having recently installed new x-ray equipment at the British Museum, it was agreed that the Great Torc would be examined.

And this brings us bang up to date with this year’s Day of Archaeology. Yesterday, we went to the British Museum to see the Great Torc x-ray results and they were better than we could ever imagine: The Great Torc terminals are definitely made from sheet gold, with repoussé decoration, and were also constructed using the same ‘apple core’ and donut method predicted by Roland and I in 2016.

The Snettisham Great Torc. (Image courtesy of Jody Joy and The British Museum).

This brings the total to three previously unrecognised sheet work gold torcs (Netherurd, Newark and the Great Torc), a fourth whose sheet construction method is now understood (the Grotesque) and a fifth torc (Sedgeford) whose cast terminals seem to be more cleverly achieved than was previously assumed. So where now?

For us, what started as a simple question with an apparently simple answer has turned into an ever increasing list of questions with research implications far beyond the Newark torc. The answers to these questions are likely to have knock on effects on the current thinking regarding Iron Age metalworking technology, trade/exchange and dating. In addition to the technological aspects we have also been investigating the decoration and tooling on the torcs and have found further intriguing results which would appear to suggest that the Netherurd and Newark torcs were made – not just in the same workshop – but also by the same craftsperson! In torcs found some 200 miles apart in Nottinghamshire and Scotland, this is unexpected and very exciting.

We are currently in the process of writing up our findings for peer reviewed publication, are preparing conference papers and are continuing to investigate various other aspects of these fabulous torcs. We suspect, if we post here again in two years time, we will have much more to tell – but we also suspect that these enigmatic objects will be posing yet more questions for us for many years to come.

And who knows, one day we may even get to do what we originally planned…and replicate the Newark torc!

Tess Machling & Roland Williamson


Entirely self funded, we have benefited hugely from the generosity and kindness of many people over the past two years: the curators of the various torcs – Glyn Hughes, Fraser Hunter, Julia Farley, Jody Joy, Tim Pestell and John Davies are greatly thanked. In addition, we are extremely grateful for the kind and generous help given by the Materials team at NPL -Rob Brooks, Tony Fry, Hannah Corcoran, Stephen Brown and Eric Bennett. Dan O’Flynn and members of the Science Department at the British Museum have offered great assistance in carrying out the x-rays of the Sedgeford and Great Torc and are thanked for their help.