The Big Book of Torcs

Tess and I have recently created a brand new website called The Big Book of Torcs where we will be sharing a lot of our torc work from now on.


Replication by Ford Hallam (Image © Ford Hallam)

We wanted somewhere informal where we could bounce around ideas and share our current thoughts on various torc topics. Although academic peer review underpins everything we do, the nature of our work means we are often pursuing a number of leads at any one time, and these may not reach a state where they can be published for several years, so we wanted an instant means of sharing information.

Current research themes we are exploring involve – torc decorating techniques; the re-use, re-purposing and repair of torcs; recurring decorative themes; why and how torc neck rings break; the replication of torcs; sources of Iron Age gold; numbers of Iron Age goldsmiths; the origins of the Iron Age goldsmithing tradition and weight and numbers patterns associated with torcs – and thoughts about all of these will appear in due course.

We’ll also be using this space to provide our thoughts on recently published research and why we, working with goldsmiths and silversmiths and from a craft perspective, sometimes come up with ideas that conflict with the mainstream archaeological narrative for these artefacts.

Tom Rout Great torc

Tom Rout finds the Great torc (Image © Eastern Daily Press)

Oh, and we also have some rather light-hearted thoughts about gold and torcs, which range from the number of goldfinches you’d need to boil to make a Snettisham Great torc (did you know that every living creature is 0.00001% gold?) through to some terrible torc puns (there are many) and jokes and maybe even a few chocolate torcs (Tess makes them!).


Chocolate Netherurd terminal (Image © Tess Machling)

We also have a collection of torc novels which may have to be shared and the odd plug or two for our favourite torc beer. We are also interested in a number of topics that are more tangential to our main torc research: for example, the people who find torcs, be they Norfolk plough boys from the 1940s or detectorists of the late 20th and early 21st centuries.

So yes, a veritable smorgasbord of torc related ephemera will be coming your way, some serious, some not so: we do hope you will come to love ‘All Things Torc’ as much as we do! 

You can find The Bog Book of Torcs HERE

A bucket for Brighton…..

I was honoured to be asked to make a replica Saxon bucket for the new Elaine Evans Archaeology Gallery at Brighton Museum in East Sussex.


The bucket was to be used as a mount for the surviving fittings of a large bucket found during the Howletts cemetery excavations in Kent in the early 20th century.  Little remained of the original fittings and wood but it is known that the vessel had a rather large rim diameter of almost 300mm – far bigger than any I have made before and would have held approximately 32 pints!

The bucket was stave built from yew wood and comprised 16 staves (2mm thick at the rim and c. 10mm thick at the bottom), slotted around a circular yew base, and held in place with copper strips, riveted together to form bands.

Using the pressblech technique, I also made some foils of the pendant fittings which were attached to one of the metal bands.

The bucket was finished and delivered ready for the Museum gallery opening in early 2019. It looks rather impressive in its new home (…although I’m not sure that the original handle would have ever supported the bucket when it was full!)

And how did I know it really held 32 pints? Well……


Did Vikings ‘nick’ the Newark torc…and the West Lindsey and Towton torcs too?

In a previous blog we wrote about the  possibility that the Iron Age Newark torc had been redeposited during the Viking period, as evidenced by the 5mm by 2mm gash which can be seen on the interior of one of the Newark torc’s terminals. This gash, as previously mentioned is purposeful and deliberate, and shows none of the evidence likely to be present had the gash been post-depositional, or the result of a percussive blow. Well, the really good news is: we think we’ve found more torcs with gashes and, intriguingly, these seem to have been deposited at various spots along the line of the route of the Great Viking Army’s movement south in 872, along the Ouse and Trent Rivers.

As previously discussed, the Newark torc was found by a detectorist in 2005 in a location close to the River Trent and in an odd spot for an Iron Age torc (which are often found on hill slopes, away from rivers or settlement activity). It was also unusual in that it was found as a solitary object whereas torcs are often discovered buried in hoards (for example, at Leekfrith, Blair Drummond and Snettisham, where four or more torcs were found deposited together). As has been shown in the previous blog the gash on the Newark torc appears to have been made deliberately and the incision made closely matches other examples of Viking nicked metalwork.

Recently, whilst looking at evidence for other torcs, Roland and myself noticed that there was another Iron Age torc, the Caistor/West Lindsey torc which shows evidence of Viking misuse! In his PAS report, Martin Foreman notes that there are ‘nicks noted on the band [which] are a familiar feature of hacksilver ingots and objects’. In addition, the torc has been broken and the remaining piece weighs an amount that corresponds with 8.06, Viking, Dublin weights. Furthermore, Foreman notes that a domed lead weight (or potentially a gaming piece) was found close to the torc findspot, on the same day. Such weights are often Viking in origin. Although, due to the sensitive nature of ‘treasure’ finds, the precise find spot has not been released, the location in West Lindsey would lie close to the River Trent, and interestingly, only some 30 miles to the north of Newark.

Suddenly we had two Iron Age, Viking nicked, torcs, some thirty miles apart, and both close to the Trent. So we carried on looking…..

Having plotted all known torc finds on a map, the only other torc finds in the vicinity were Ulceby and Towton. Ulceby was an apparently undisturbed and secure Iron Age hoard find, with gold torcs and bronze horse bits etc, buried together and with no nicking recorded (although it should be noted that these items have not been checked by the authors). One of the Towton torcs, however, proved to be more interesting. In images from The Yorkshire Museum website, a nick appears to be visible on the back of the right hand terminal of one of the torcs (incidentally, in the same place that the Newark torc is nicked) .


(Image courtesy of York Museums Trust :: :: CC BY-SA 4.0)

This was confirmed by an image previously take by Roland, which shows the nick much clearer and shows that it has very close similarities to that seen on the Newark torc. Towton is on the left, and Newark on the right.

The Towton torcs, although again having an undisclosed location due their ‘treasure’ status, appear to have been found in Selby, which lies close to the River Ouse. Again the find site is unusual, the torcs were found -one in 2010 and one in 2011- in a stream bed, and have always been assumed to have washed there from their original deposition site.

The evidence above amounts to three torcs, all with apparent Viking nicking, found in unusual locations not typical of Iron Age torc hoard deposition and most interestingly, all appear to have been found in areas close to the route of the Great Viking Army as it moved south from York, via Torksey and Newark to Repton in 872. So, do these torcs map the route of the army, via the Ouse and then the Trent? and how on earth did they get into the hands of the Vikings if they do?

Our suggestion is that these four torcs may well have started life as a single hoard which was found somewhere to the north, and then was broken up and gifted/deposited at various points on the route south. This would fit the evidence for a number of reasons: firstly, Iron Age torc hoards are not common and we believe there are no other Iron Age hoards as yet recorded which show evidence of having been nicked (there are however two other torc pieces, one from nearly Roxby cum Risby and the second from Telford that Foreman suggests conform to Viking weights). In addition, all the nicked torcs come from a defined geographical area some 60 miles long with York at the northern end and Newark at its southern extent.

Perhaps more convincing  is the evidence that both the Newark and Netherurd torcs appear to have been made/finished by the same maker (see here for our recent Proceedings of the Prehistoric Society paper) . The Netherurd torc was deposited in a hoard just south of Edinburgh and so it is not impossible that the Newark torc originally came from a hoard location to the north of its current find spot.

If it is assumed that all four torcs came from the same hoard which was located by the Vikings and broken up, this would make a certain amount of sense: two or three neck torcs and one or two bracelets would not be unusual, and would have been a similar composition to that seen in several torc hoards, including that at Leekfrith and the later Winchester hoard. Weights may also hold a clue as the complete hoard (if it is assumed that the West Lindsey torc was once complete, and proxied by the extremely similar Leekfrith torc weighing 230.6g) would give a total hoard weight for both Towton torcs, the West Lindsey torc and the Newark torc of c.1060g. This weight has previously been noted by Roland and myself as of relevance in later Iron Age torcs and hoards and can be seen time and again in the weight of the complete Netherurd hoard which was c.1092g , the weight of the Snettisham Great torc which is 1084g and similar weights also appear to be occurring in the Essendon hoard also. These weights will be explored in more detail in a forthcoming Historical Metallurgy Society paper.

Also, the fact that the West Lindsey torc may be of an earlier date (c.400BC) than the other three torcs (c.200-100BC) does not preclude their having been found together as it is known from sites such as Snettisham that torcs could be curated/collected/used for a number of years prior to deposition. In addition, dates of IA torcs are currently insecure and mainly based on art historical criteria and related coin finds, and so it may be that the manufacturing dates of all four torcs are not so widely different after all.

Of further interest is that evidence that only one of the nicked torcs was broken up and apparently cut down as bullion: this could be explained by the West Lindsey torc being the least showy of the neck torcs with the two armring Towton torcs and the Newark torc potentially having a greater value as complete items, perhaps as wearable armrings and as a necklace of exquisite goldwork. However, this can never be proved with certainty…. although it may cause trouble for the authenticity officers in a number of re-enactment groups should it be proved that the Vikings were wearing IA torcs!!

To sum up, we think we have four torcs, possibly from the same hoard, that came into Viking hands, perhaps at York, and which may have come from an original deposition site to the north of York. The evidence currently is by no means secure, and we will be carrying out further research to try and prove or disprove our theory. However, we feel that the evidence so far makes a tentative case that, in the late 9th century, the Viking Army found themselves an Iron Age hoard…..and nicked it!


Please let us know what you think! We’re on twitter @Tess_Machling @bodgit_bendit  or can be emailed at



Museum replicas… and chocolate??

A guest post from Tess Machling

I’m an archaeologist and Roland is a Museum replica maker. Roland and I usually work together on museum replica projects and also we research Iron Age gold torcs. Roland makes the most amazing replicas for many of the biggest and best museums in Britain, and his work has recently featured in the Celts exhibition at the British Museum. He works in a variety of media including metal, wood, bone and cloth and makes all manner of things from swords to brooches to bone combs. And if he makes something that needs casting, he makes a model of that thing first, so it can be moulded and cast… And that’s where the chocolate comes in!!

If you have a model, you can as easily mould it and use that mould to make chocolate, in the same way as you would make a metal replica! And that’s where the whole chocolate replicas thing started…..

Over the last few years, I have been taking some of Roland’s models and replicas (along with some of the rather marvellous teaching collection from St Albans Museum) covering them with food-grade silicon putty and making moulds of them, which I then fill with chocolate!

So far, I’ve made handaxes, axes, torcs, arrowheads, coins, combs, brooches and various other bits and bobs. My only rule for what I make is that it has to be 100% edible: no props, inner strengthening material or tricks of the eye. And so when it came to make them as good as the original replicas or artefacts, it was really important to me to keep up that one rule. But how to make them look good, and also be edible?

Well, it turns out you can buy edible paint, the sort of thing sold in craft shops to decorate cakes and icing. So then I started painting them, and below you can see the result. Everything you see is 100% edible…..

Roland and I are now running workshops to talk about how I make chocolate, and how Roland makes his amazing museum replicas. At each workshop we bring a large number of Roland’s replicas for people to handle (and yes…swords!), also many of my completed choccy versions. At the workshop you can also paint some chocolate replicas (all materials are provided) and take them home with you! We hope to see you at one of our events. The next is at the gorgeous new St Albans Museum and Art Gallery on Saturday 8th December: do book for it. We guarantee it will be interesting and fun!

P.S. And for those not of an entirely archaeological bent, I also provide a number of non-archaeological objects to paint, including chocolate tiles, flowers, shells and cats!

We were wrong!

We all make mistakes. In academia, it’s sometimes tricky to hold your hands up and admit you got it wrong. But we did. We got it wrong. The long and short of it is that the Newark torc is not sheet-work as we previously thought and wrote about, but is instead almost certainly a hybrid of sheet and cast technologies!!! Yup, you read that right. Cast and sheet!

Yesterday, we went to GE Inspection technologies with the Newark torc. They had kindly offered to x-ray the torc for us with a high-powered machine (it’s amazing what you can do with 3/4 million pound’s worth of 18 tonne machine). Prior to this, preliminary, lower power, scans had suggested that the relief elements of the terminals were hollow, as would be expected in a sheet-work torc. We had also noticed what appeared to be ‘apple core’ seams running around the terminal. However, once we got the torc into a high powered x-ray, although it was not immediately obvious, after looking at several different perspectives, it slowly dawned on us what we were actually seeing.

The relief was not hollow, but instead more filled and typical of a cast terminal. In addition, what we had assumed were the ‘apple core’ seams visible on the torc, in fact turned out to be a narrower ring of material, added to each terminal either prior to or after the casting of the main piece.

Rather more exciting, was the wire attachment to the collar and torus, which proved to be intricate and highly skilled and created a seamless joint to complete the terminal.

So where does this leave us? Were we wrong?

Well yes, we were, because Newark is not made of gold sheet-work (although the core element may well be more related to sheet than cast technology). What we had thought to be an ‘apple core’ was more complicated. However, the decorative link with Netherurd remains and we are still confident that the decoration was carried out by the same hand in both torcs (could we be looking at the same finisher? or was the Netherurd/Newark  maker skilled in both sheet and cast?).

This also opens up an intriguing connection with the Sedgeford torc: as we’d previously noted, this torc is almost identical in size to Newark (interestingly, the central terminal hole in the Newark torc is a little larger than Sedgeford and this may link with the central added ring we have witnessed in the Newark torc.)…but is very differently decorated, of a lower gold content and the wires appear to be attached to the collar using a complicated rivet attachment, which is not present on the Newark torc. In addition, as with all the other castings we have so far seen, the Sedgeford terminal casts are faulty, on both terminals whereas, even under high powered x-ray, both the Newark terminals appear to be flawless.

So, despite the fact we were wrong, intriguingly we now have a link, not just between Netherurd and Newark….but a kind of torc venn diagram that links Netherurd and Newark’s decoration, and Newark and Sedgeford’s technology. Watch this space. Torcs just got interesting…..Again!

If you want to read more on our torc work:

‘Up Close and Personal’: The later Iron Age Torcs from Newark, Nottinghamshire and Netherurd, Peebleshire by Tess Machling and Roland Williamson 

Did vikings ‘nick’ the Newark torc?

We’ve spent a lot of time researching Iron Age torcs of the sort most famously represented by the Snettisham Great Torc and we have now been able to prove that many of these torcs were made using a previously unknown gold sheet-work technique (for more on this see here). We have also been able to show that it is likely that the Netherurd and Newark torcs were made or finished by the same maker, probably somewhere beyond the traditionally assumed East Anglian heartland (peer reviewed paper will be out soon, so watch this space….).


The Newark torc (Image © Roland Williamson & Tess Machling)

After three years of looking at these torcs, we know them pretty well – but there is one thing that has continued to puzzle us. The Newark torc (which is on display at the National Civil War Centre in Newark) is to all intents and purposes an ideal example of torc making. But it does have one fault: a large gash (some 5mm by 2mm) on the interior face of one of the terminals.

Gash on Newark torc terminal (Image © Glyn Hughes, Roland Williamson & Tess Machling)

This gash has always been of concern to us. It is clearly a very deliberate mark, made with force, but which did not distort or dent the surrounding gold. It does not look like post-depositional damage and is unlikely to be accidental. But what is it? Various ideas – including a rather bloodthirsty thought of the torc wearer’s neck being cut – have come up over the years, but none was entirely satisfactory.

However, in the last few days, we have been considering a new idea which is slightly off beam but, if correct, will set the Newark torc at the centre of a story that we may never be able to tell in entirety, but which is persuasive and exciting nonetheless.


The Newark Torc was found by a detectorist in 2005 in a field just outside Newark on flat ground, close to the River Trent. Despite an excavation in the vicinity, no other Iron Age items were located. The location itself was unusual: in many cases (for example, Snettisham, Netherurd, Leekfrith etc) torcs have been found on gentle hill slopes, and have been part of hoards of usually two to four items. The topographically flat location and single nature of the find were unusual. In fact, a find of a Snettisham type torc, so far beyond the confines of East Anglia was deemed, at the time, to be suspicious enough for some to even go so far as to initially suggest it might be an antiquarian fake.

Recent work by us at the National Physical Laboratory has shown that the terminals of this torc, like the Netherurd terminal and the Snettisham Great Torc, are almost certainly made of gold sheet. However, the gash in the torc has always caused us much head scratching: how could a blow have left such a mark without denting or distorting the gold if the torc terminal was made of sheet gold? It just didn’t make sense. However, by looking closely at this mark, and comparing it with other examples, the find location, and the history of the area we believe we may have come up with a solution.

The gash on the torc is large, some 5mm by 2mm and around 1mm deep. It comprises a deliberate linear incision – the cut of which has come in at slight angle from a direction going towards the wires. It has caused a bulbous area of gold to develop on the opposing side of the cut and must have been achieved by a very thin, and very, sharp edge. The bulbous area of gold suggests that the cut was opened as the incision was made. In profile the cut is ‘v’ shaped.

Such an incision is difficult to create from a percussive blow, and neither is it correct for a sawing action. The nature of the cut suggests more that the gash was made by a very controlled action: the incision of a blade, which was then deliberately twisted or turned to open the cut to a ‘v’ and which then caused the thickening of the gold on the opposite side of the entry point of the incision. The next question is why?

In searching for information, we could find no evidence to suggest that this gash was likely to have been created in prehistory, nor could we find evidence for such practices in the Roman or Saxon periods. What did become evident is that the Vikings regularly nicked metal items and, although a far more common practice with silver, there are also examples of nicked gold work.

The practice appears to have been carried out as a test of the item’s metal purity – by opening up a nick in the metal, any evidence of gilding, or plating, could be easily seen and the texture of the metal felt. It is also possible that such a technique became a kind of hallmark of quality that could be easily examined if the item was traded. Others have suggested that the marks may have been a kind of trading mark, made each time the item changed hands, or that the nicking became a kind of ritualised action whose practical necessity was the least important element of the nicking. Indeed, there are several silver ingots that show large amounts of nicking far beyond that which would be practically needed.

By comparing the nature of nicks in known Viking examples from the links above and from, for example, items recorded by the Portable Antiquities Scheme there does seem to be a close similarity between nicked Viking examples and the gash seen on the torc. Enough to convince us that this theory was well worth exploring further.

We next looked at the known Viking activity in the vicinity of Newark, and this too provided good evidence. It is known that the Great Heathen Army passed through Newark en route to Repton, after overwintering in Torksey in 872/873.

Time for a flight of fancy…..

WARNING: What follows is an interlude of entirely unsupported speculation  – but we just could not ignore that name. Torksey. Tork-sey. Turc’s island in Old English, although the etymology is insecure. Torksey is a fascinating place: a rare raised area of land in an incredibly flat landscape, made effectively an island by the River Trent on one side and marshes on the other. The perfect place for Iron Age people to bury a torc….

Torksey is also the place from where the Viking army set off from for Repton. And on the way to Repton, following the River Trent, they would have had to pass by the spot almost exactly where the Newark torc was buried. Not such a perfect place for Iron Age people to bury a torc….

And yes, we know we’re really pushing the bounds of credibility, but did the torc come from Torksey? And even if it didn’t then an original site of deposition further north (putting the Newark torc closer to its Netherurd sibling) is a distinct possibility with the Vikings obtaining the torc somewhere along the route from the North.

But back to reality…..

With the Viking army in such close proximity to the torc’s find spot by the River Trent at Newark, and with the Newark torc’s probable northern attribution shown by its similarities to the Scottish example from Netherurd, it is possible that it could have been ‘acquired’ by the Vikings either in Newark or further North, and that it could have been deposited in Newark as a bullion cache.

This would be supported by the nick, the odd context of the find’s deposition and the presence of the Viking army in the right place at the right time. This is, we fully admit, an outlandish theory, based on not much more than an unusual torc find location and a gash that cannot be easily explained. But we have to say that we like the idea. It works. We want to explore it further. Because, after all, they did tell us these torcs certainly couldn’t be sheet gold…..!

So what do you think: An unusual Iron Age torc find or Viking plundered bling?

Please let us know!

Tess Machling & Roland Williamson, 17th May 2018.



Making loops…

Recently myself and Tess Machling have been involved in making more Bronze Age Sussex loops, this time using more traditional techniques. We wanted to see what the limitations of the techniques were, and whether known archaeological finds could be used to make a loop.

Giovanna Fregni, in her groundbreaking PhD thesis, ‘The Compleat Metalsmith: Craft and Technology in the British Bronze Age’ (Fregni 2014) describes the many types of tools available to the Bronze Age craftsperson. Of interest to our work, are Fregni’s Type 1, 1a and 3 hammers. However, hammers of this weight proved insufficient to move the 8mm cross-section bronze (although they would certainly have been sufficient to bend the finer 4mm cross-section examples such as the East Dean example -Loop 9 in Friend’s catalogue (Friend 2013)) and so a work around was sought. This involved using hammerstones of sufficient weight, 1.8kg, used against a wooden punch or drift, to avoid marking of the metalwork. This proved capable of moving the metal. These would have been available to BA craftspeople. It is, however, possible as Fregni notes, that heavier bronze hammers were available at the time and have been recycled for their bronze once they became no longer usable, but without extant examples, this will always be speculative.


Fig 1: Artist’s representation of how a loop might have been clamped using a split beam (Image © Roland Williamson)

We also think that, without the use of modern clamps, holding the loop during construction would have been extremely difficult. Our suggestion is that a suitably robust tree, split and wedged, could provide a solution comparable to a modern vice. However, it proved difficult to locate a suitably robust tree with a correct branch angle (see Fig 1). This must have also been a problem in the Bronze Age, and would have perhaps been solved by husbanding trees at a young age, and encouraging them to grow in a certain manner by using careful binding and training of branches). This, in turn, has implications for the importance of such trees which may have had a special status as metalworking sites.


Fig 2: The use of a large pebble and wooden drift.

For our purposes, a suitably authentic set up was created (Fig. 2) which did not utilise clamps or other modern equipment. Interestingly, working in this way proved impossible for just one person to achieve, with a minimum of two being necessary to complete the work. This could be suggestive of a craftsperson and apprentice working together and would also suggest that Bronze Age workshops might have been staffed by more than one craftsperson? However, we do believe that the split beam solution would have allowed for a single person to construct a loop, thus negating the need for a helper. Further experimentation will be necessary. The one downside to the two-person method used above was a certain amount of slipping of the drift from the loop.

In addition, closer study of the examples held in the Sussex museums, and previous replication of the loops, suggests that there are several different techniques used to make loops, including a method which results in a flattened loop on the ‘wrist band’ and another which results in a loop which continues the circle of the ‘wrist band’shape.

Decoration of the loops, as seen on examples such as the Handcross loops (Friend’s loops 10 and 11) proved straightforward with an observable difference being noted between the bronze chisels and modern steel examples (Figs 3 and 4): the bronze chisel became worn with time and created a less precise punched mark in the bronze loop as use increased. This would suggest that chisels were often re-sharpened regularly as the decoration on BA examples does not show a reduction in precision of the marks.

Fig 3 and 4: Chisel marks on the bronze loop (left), and wear of chisel (right).

This work has shown how the manufacture of Sussex loops may have been achieved in the Bronze Age, has offered some thoughts about how many people might have been involved, and which tools could have been used. Further research will be necessary to refine these ideas, but our work has provided a few suggestions for the techniques utilised.


Fregni, E.G. 2014. The Compleat Metalsmith: Craft and Technology in the British Bronze Age. PhD thesis, University of Sheffield.

Friend, A. 2013. The Sussex Loops: An investigation into their form function and significance. MA thesis. Bangor University.

We are extremely grateful to the Prehistoric Society for awarding us the James Dyer Prize, which enabled us to carry out this work.

Buckets, bowls and bits and bobs…..

It seems a long time since I last wrote here but – fear not – I have busy making all manner of things from a copy of a Viking beard comb from Lund in Sweden and a tiny Saxon bucket from a grave from Worthy Park through to a copy of the Germundbju helmet and a shield and cauldron or two! I’ve also updated my gallery, so do go have a look!

The Cirencester Cockerel

As mentioned in a previous blog, I was commissioned to make two replicas of the Cirencester Cockerel found on an excavation in 2011. The brief for this commission was unusual in that the client not only wanted a replica of what the cockerel would have looked like back in Roman times, but also wanted a copy of the Cockerel as he is today – in all his aged glory!

Both cockerels started life as an epoxy resin model, before being cast in bronze.

Once cast, one model needed to be polished before having the coloured enamels added. The second model needed to be treated with a little less honour: whilst still needing a lot of work, my aim was to make him look as old and battered as the original looks today! This involved some vinegar, a little bit of car paint and an awful lot of trial and error!

But I hope you’ll agree the finished results for both cockerels give a very good ‘before and after’ impression. I’m very pleased with both.



The large Knutsford brooch

The Knutsford brooch was found in 2012 as part of a large hoard of 103 coins, three brooches and two finger rings. I was approached by Museum of Liverpool to replicate the largest brooch of the three, and one of the finger rings…and it has been one of my most difficult commissions to date.

(images above, copyright Museum of Liverpool and PAS)

Made in several parts, and cast in silver with gilding added afterwards, the brooch started life as a piece of plastic card which then had epoxy resin added before being carved, having more resin added and being re-carved. Due to the intricacy of the piece, much of this work was carried out whilst wearing magnifiers!

When the carving was complete, the piece was then cast in silver. This proved to be as difficult as the carving, with three casts being made before the brooch was right. It makes me wonder about the original brooch which was broken close to the foot: was this a weakness that the Roman caster also had issues with?

Once cast, the brooch had to have all the casting material removed, the clasp bent to take the pin and then it was polished before being gilded. The areas not to be gilded need to be masked off, and I use a brightly coloured nail polish (not mine I hasten to add!) to do this.

After gilding, the brooch was cleaned and polished before having the spring, rosette and clasp attached. The brooch was finished …and I could finally breath a sigh of relief for another job completed!

The finger ring proved to be a far more straightforward undertaking: in silver, this ring also included a carved carnelian intaglio. Again moulded in resin, the ring was cast in silver. Then the intaglio was carved and attached.

I hope you like the finished result. The brooch and ring will now be used in workshops by the museum.