Medieval Money Mystery Solved

[Romismatist]

Hello Everyone,

Interesting research out from researchers from the Universities of Oxford, Cambridge, and the Vrije Universiteit Amsterdam who analyzed early medieval coinage from the Fitzwilliam Museum in Cambridge. They discovered that Byzantine bullion made up Europe's early adoption of silver coinage in the 3d to 7th centuries AD, but then the Frankish mine in Melle rose to prominence from 750-820 AD, supplying silver coinage in the UK and continental Europe.

https://phys.org/news/2024-04-early-medieval-money-mystery.html

[John Conduitt]

Interesting. A little dogmatic as usual for these things. I didn't follow why, if the silver wasn't from a western mine or recycled Roman coins, it had to be Byzantine. Especially if links were not strong at that time and those in the west must have been 'sitting on it for decades'. That also goes against their argument that the 'dramatic surge' in the use of silver coins was due to the increase in the supply of silver - which they apparently had already. The goths, huns and vandals went into Africa and Asia, so there were other sources.

The Melle mine makes sense. But it is also odd that "people in England would have been very aware that their silver was coming from Francia and that they depended on it." It was only a couple of centuries before that much of Europe's silver came from Britain. So why didn't they use that? True, it was actually from Wales, but Offa seemed so disinterested in gaining control of there he built a huge dyke across it.

Edited April 9 by John Conduitt

[JeandAcre]

Huge thanks for the head's up and the link, @Romismatist!!!  You could bet money you didn't have that it was summarily bookmarked.

Regarding the transition to Melle from Byzantine sources (for those tuning in late, no, you Really Have to see the webpage), there are these.  As you might expect, Melle was one of the commonest mints for Carolingian issues, which (ditto) were immobilized into the 12th century.  William IX, Duke of Aquitaine (grandfather of the eponymous Eleanor) continued the trend, specifically for the ones of Charles III, with the mint featured very prominently on the reverse.

 

Charles II (King of West Francia 840-877; Emperor from 875.)  Denier of Melle, c. 840-864.  Funly, my only Carolingian with Viking 'peck marks.'

Obv. +CARLVS REX.  Rev. 'CAR<>LVS' monogram; +METxVLLO.  Depeyrot (3rd ed., 2008), 627 (very minor variant; 'x' instead of '+' in the mint signature).

Charles III (898-922).  Initiating the 'MET ALO' legend in two lines.  Depeyrot 629; cf. 630.

Immobilization perpetuating the type of Charles III.  As such, on the grounds of style and module, likely as early as William IX (Duke of Aquitaine 1086-1127), although Duplessy doesn't list a variant with only one central pellet on the reverse.  Cf. Duplessy 906-910.

 

Edited April 9 by JeandAcre
Just, Lose that redundant pic at the bottom! ...3, No, 5 tries later, What. is if Going. to Cost me to Lose That (expletive of choice) last Picture. At this point, if that would be a question, it would be rhetorical. So why bother?

[John Conduitt]

I have a fair few coins made of this silver. Every early Saxon coin is Byzantine, apparently (although, ironically, silver Byzantine coins are not all that common), while the late Saxon coins are from Melle.

Series A Sceatta, 673-685

East Kent. Silver, 12mm, 0.88g. Radiate bust with curved exergual drapery; TIC in front, A and annulets behind. Degenerate votive standard with seriffed letters TOTII, tufa above containing trefoil of pellets, rounded horns, seriffed letters in margin, cross below (S 775; BMC Type 2a). Possibly attributed to Hlothere of Kent.

Offa Group II Light Coinage Cut Halfpenny, 780-792

London. Silver, 16mm, 0.43g. OFFA with chevron-barred A (over REX), separated by line of pellets with forked ends; various pellets in the field. AEDEL (over PALD, moneyer Aethelweald), divided by line of pellets with forked ends; various pellets in the field (S 904).

Offa, meanwhile, seemed to do a lot of trade with the Abbasids.

 

Edited April 9 by John Conduitt

[ela126]

I enjoyed the article, the Byzantines though, weren’t know for their vast amounts of silver coinage. Although the hexagrams had become rather commonplace by the middle of the 7th century. 
 

I don’t have the details in front of me, but since I see some Carolingian coins from Melle being shared, this is my Charles the Bald (I think, possibly Charles the simple, I get them confused. Apologize that it’s slabbed, one of my early acquisitions that I felt had to be slabbed.

[Rand]

What happened to Roman silver, and what could have been the context of the silver influx from Byzantine to English? Were at least some coins produced from Roman coin silver?

There was a prolonged time gap between Byzantine silver reaching England and minting coins. The silver was likely to undergo several rounds of recycling for jewellery and other items before being used for the coins. Is it not surprising to see such a uniform pattern of findings, with the metal content remaining consistent across a range of coin types?

[JeandAcre]

Brilliant stuff, @John Conduitt.  An Offa Anything?  Just, never mind.  I have to love how the name just leaps out at you, as if it was a neon sign.

The sceatta is absolutely brilliant.  The one representative example I gave myself permission to buy was this one, with the same combination of radiate crown and 4th-c. standard.  

Not least for this period, I need the distinctive combination of economic dynamics that were in play.  Yes, the mercantile classes were a miniscule socio-economic elite.  Given which, trade, especially by maritime routes, was faster, more efficient, and more routine than it's often given credit for.  Stuff could really Jump.  ...I wish it was easier to find (not bookmarked; Slap Both Hands), but there was an absolutely brilliant documentary (BBC or, I doubt it, PBS) on excavations around Tintagel Castle, establishing that, vaguely during the Arthurian period, there was a polity there that was effectively in constant contact with the Mediterranean, including Byzantines.  My uncle (a.k.a. Dad's Smarter Younger Brother), who was a professor of physical anthropology, liked to emphasize this same point, in the context of First Nations across North America --thank you, even by overland routes (including, thank you again, rivers, including upstream routes.)

Moral: We don't get to underestimate these people's intelligence.  Just for one, they didn't spend hours parked in front of the tv.

 

[JeandAcre]

24 minutes ago, ela126 said:

I enjoyed the article, the Byzantines though, weren’t know for their vast amounts of silver coinage. Although the hexagrams had become rather commonplace by the middle of the 7th century. 
 

I don’t have the details in front of me, but since I see some Carolingian coins from Melle being shared, this is my Charles the Bald (I think, possibly Charles the simple, I get them confused. Apologize that it’s slabbed, one of my early acquisitions that I felt had to be slabbed.

Yike, @ela126, that's a Very close match to the last one in my post.  ...Right, with minor variants, but during this phase, that's effectively the default mode.  Nowhere near enough of this is published in the first place.

[John Conduitt]

31 minutes ago, Rand said:

What happened to Roman silver, and what could have been the context of the silver influx from Byzantine to English? Were at least some coins produced from Roman coin silver?

There was a lot of Roman silver in Britain, much of which had come from British mines. A huge amount of it was simply buried in the ground when the Romans left. The late Romano-British clipped the remaining siliquae and used them as smaller coins for a while, as they slowly reverted to using hacksilver.

Flavius Victor Clipped Siliqua, 387-388

Milan. Silver, 10mm, 0.68g (cut down from 16-17mm, 1.25-1.8g). Pearl-diademed, draped and cuirassed bust right; D N FL VIC-(TOR P F AVG). Roma seated left, holding globe in right hand, reversed spear in left, (VIRTVS RO-M)ANORVM; mintmark MDPS (RIC IX, 19b). From the South Ferriby (Lincolnshire) Hoard 1909. Portable Antiquities Scheme IARCH-1C7D3F.

It seems odd that they would need Byzantine silver specifically or that they would receive a sudden influx of it. But a flow of silver from the East in general seems quite likely given the trade routes to Persia.

 

31 minutes ago, Rand said:

There was a prolonged time gap between Byzantine silver reaching England and minting coins. The silver was likely to undergo several rounds of recycling for jewellery and other items before being used for the coins. Is it not surprising to see such a uniform pattern of findings, with the metal content remaining consistent across a range of coin types?

Yes at that point, with no mines open, the recycling of all the existing silver in western Europe would make it homogenous with no regional variations, as it was found to be. The high proportion of gold reflected the coinage at the time, which started off gold c600 and slowly debased to silver c700. I don't know if that has been considered in the study when claiming that a 0.6%-2% gold content aligned it with Byzantine silver.

As they say, many of the coins in Britain in the 600s weren't made in Britain anyway. Britain was in a sort of monetary union with Frisia, and a lot of the coins were struck there, or at the edges of the Merovingian kingdom. Most of the English coinage was struck south of the Thames.

Series E, Variety D, Op den Velde Sub-Variety K, Anglo-Saxon Continental Phase Sceat, 695-740

Frisia. Silver, 1.33g. Quilled crescent on wheel enclosing cross pommée with pellet in three quarters. Central pellet-in-annulet in line-beaded square, unusual geometric symbols around (S 790B; SCBI 69, 225 this coin). Ex Tony Abramson. Found by A Wicks at Amesbury, Wiltshire (not far from Stonehenge).

Edited April 9 by John Conduitt

[JeandAcre]

@John Conduitt, it took a minute to find the pics, but here's my clipped, sole representative, correspondingly prized siliqua of Honorius, also of Milan.

 

[John Conduitt]

3 minutes ago, JeandAcre said:

@John Conduitt, it took a minute to find the pics, but here's my clipped, sole representative, correspondingly prized siliqua of Honorius, also of Milan.

 

Nice example. Since most of the portraits are the same, you need just enough legend to work out who it is.

[JeandAcre]

@John Conduitt, thank you!  In this instance, that was my good fortune.

(Instant edit:) now that I'm newly retired, I wouldn't dare to go looking for anything comparable.  That one was cheap in the first place.

Edited April 9 by JeandAcre

[Hrefn]

I felt the article deserved an additional illustration.

The first two coins are Byzantine Hexagrams featuring the legionary battle cry, “Deus adiuta Romanis!”  Almost always poorly struck and appearing like the emergency coinage they probably were, the first coin is a brockage with just a few letters of the reverse inscription ROMANIS seen retrograde and incuse in the middle of the flan.  Nevertheless, they are both chunky bits of good silver which would gladden the heart of any barbarian.  Possibly they are similar to the feedstock of the coins in the following row, since it is thought that they were made of melted church plate, among other things, during the crises of the Persian and Muslim threats to the empire. 

The second row shows examples of the earliest North European silver coinage since the departure of the Romans.  From left to right, the first is from Northumbria circa AD 710, a type “J” penny attributed to Osric.  The obverse shows two facing profiles flanking a standing cross;  the reverse a whorl of birds.  

The next coin is from Kent, circa AD 690, attributed to Wihtred.  The obverse is a diademed bust;  the reverse is a bird on a cross.  

The third coin is from Mercia (or Essex?) circa AD 735, attributed to Aethelbald.  A female centaur graces the obverse, and a wolf whorl the reverse.  

The fourth coin is from Denmark, around AD 700, with a bearded visage usually believed to be Wodan, and a fantastic beast on the reverse.  Believed to be the first Danish/Viking coin.  The crosses on the obverse may not be intended to be Christian symbols, but be merely apotropaic.  The area where these were struck was highly pagan when these were produced.  (I always wanted to use the word apotropaic in a sentence, and now I have!)

The third row is comprised of coins of Louis the Pious.  The first three are Class 2 deniers.  The first is from Bourges;  the second is from Tours where the great battle was fought by Charles Martel only 66 years before.  The third came from the mining town of Melle (Metallum).  If the article is correct the silver from which all these deniers were minted may have originated there.  Because these are Class 2 deniers, they can be precisely dated to AD 819-822.      

The final coin is part of the so-called Temple coinage (AD 822-840) with the legend XRISTIANA RELIGIO, also called Class 3.  These coins of Louis the Pious were of almost identical design throughout the Carolingian empire, which doubtless served as a powerful statement of unity and centralized power.  Unfortunately the lack of mint marks renders them less valuable to archeologists and less interesting to collectors.  

The fourth row contains coins of Charles the Bald, AD 840-877.  Here, mint designations have returned, and we see examples from Paris (PARISII CIVITAS), Le Mans (CINOMANIS CIVITAS), Quentovic, an important trading port (QVVENTOVVICI), and St-Denis, the burial site of the French kings (SCI DIONVSII M)  for the mint of Saint Dionysus who is buried there.  

[DLTcoins]

The article is poorly worded, I think. As I read it, coins dating from 660-820 were tested. Coins dating after 750 matched the isotope signature of silver produced from the mine at Melle. The pre-750 coins, however, matched the isotope signatures of silver objects (not necessarily coins) known to have been produced in the "Byzantine Empire" (i.e. Eastern Mediterranean) from the 3rd through 7th centuries. This earlier silver does not match any known "European" source, the article says, and shares "no meaningful overlap with late Western Roman silver coins". The upshot seems to be that when trade blossomed in Western Europe in the mid-7th century with silver replacing gold as the primary coinage metal, prestige (church?) objects originating in the Eastern Mediterranean were sacrificed to provide the necessary silver for nearly a century, after which the mine at Melle became the dominant source. Fascinating stuff!

Edited April 10 by DLTcoins

[John Conduitt]

53 minutes ago, DLTcoins said:

The article is poorly worded, I think. As I read it, coins dating from 660-820 were tested. Coins dating after 750 matched the isotope signature of silver produced from the mine at Melle. The pre-750 coins, however, matched the isotope signatures of silver objects (not necessarily coins) known to have been produced in the "Byzantine Empire" (i.e. Eastern Mediterranean) from the 3rd through 7th centuries. This earlier silver does not match any known "European" source, the article says, and shares "no meaningful overlap with late Western Roman silver coins". The upshot seems to be that when trade blossomed in Western Europe in the mid-7th century with silver replacing gold as the primary coinage metal, prestige (church?) objects originating in the Eastern Mediterranean were sacrificed to provide the necessary silver for nearly a century, after which the mine at Melle became the dominant source. Fascinating stuff!

Yes indeed. The reasoning behind Byzantium as the source is particularly unclear. Especially if the Byzantines were themselves melting down the church silver to produce their own coins. It doesn't sound like there was all that much to spare for the millions of sceattas produced. There are only so many candelabras in churches. Although I'm sure they had silver mines in Greece. Of course, the Saxons would use whatever silver they could get. As the article says, that would only be Byzantine if it had been lying around a few decades, which then doesn't make sense as the reason why so many coins were suddenly produced. Could it be silver from further east?

The cut off dates they've used specifically relate to the type of coin produced, which were made under quite different regimes. Saxon cut off dates are somewhat arbirary because the dating of the coins isn't any more precise. They rely on a few hoards. The start date of 660 (based on the Crondall Hoard) relates to the period when the Saxons stopped using gold for coins (which they'd been striking since around 610). So for coins to have a higher gold content just after that point would be normal. All the coins produced between 660 and 750 are sceattas. They were the ones produced in Kent, Essex and Frisia, with a few in 'York', Hampshire and East Anglia. For these to have a particular silver source makes sense, as I imagine the British coins were melted down continental coins.

750 is the traditional 'end' of sceattas. This was when Offa came to power and changed the whole currency system to one based on Charlemagne's. These were the first pennies of the sort then used until Henry VII. The mints were under centralised control. Presumably at that point, having based the currency on Charlemagne's, Offa sourced his silver from him too.

[Rand]

The accepted refining processes could have influenced the metal content, potentially leading to contaminants. The analysis heavily relies on isotopes of lead, which is not the primary coin metal—I am unsure if some of it could have been introduced/lost during the alloy preparation. 

Were there any technological changes in silver refinery/minting practised around the same time the Melle became active, which could have contributed to the alloy composition?

While appearance of Melle silver in coins is expected, the uniformity of ealier silver and the lack of obvious transition to new silver still puzzles me.

[John Conduitt]

8 minutes ago, Rand said:

The accepted refining processes could have influenced the metal content, potentially leading to contaminants. The analysis heavily relies on isotopes of lead, which is not the primary coin metal—I am unsure if some of it could have been introduced/lost during the alloy preparation. 

Were there any technological changes in silver refinery/minting practised around the same time the Melle became active, which could have contributed to the alloy composition?

While appearance of Melle silver in coins is expected, the uniformity of ealier silver and the lack of obvious transition to new silver still puzzles me.

The big change at this point was Charlemagne (or to be precise, his father, Pepin the Short, who reigned from 751). Pepin and Charlemagne revolutionised the coinage and so it seems likely all sorts of things changed or were rationalised, including technology, silver sources, supply chains etc. They standardised, centralised and harmonised the currency, using only the denier. They abandoned the gold standard because of a shortage of gold, itself caused by peace with Byzantium in which Venice and Sicily (and therefore trade routes to Africa) were given to the Byzantines. They weren't happy relying on gold from distant lands and knew they had silver deposits in their own territory, so switched to those. That is why Melle came into being and the old method of melting down whatever you had (by then very uniform) suddenly ceased.

It seems to me it could be that the Byzantine silver had the same make up as the early Saxon silver because it was also recycled, not because it was the main source of Saxon silver. If no-one was mining silver in Europe, of course it would all be the same. Charlemagne addressed that and the silver content changed.

[Rand]

5 minutes ago, John Conduitt said:

It seems to me it could be that the Byzantine silver had the same make up as the early Saxon silver because it was also recycled, not because it was the main source of Saxon silver.

This makes sense to me, but would lead to a different conclusion for the study: the coins were minted from old recycled silver rather than Byzantine silver. 

It would be good to see analyses of Italian silver of the period which would also be recycled and contemporary eastern coin silver from Persia.

[John Conduitt]

9 hours ago, Hrefn said:

I felt the article deserved an additional illustration.

The first two coins are Byzantine Hexagrams featuring the legionary battle cry, “Deus adiuta Romanis!”  Almost always poorly struck and appearing like the emergency coinage they probably were, the first coin is a brockage with just a few letters of the reverse inscription ROMANIS seen retrograde and incuse in the middle of the flan.  Nevertheless, they are both chunky bits of good silver which would gladden the heart of any barbarian.  Possibly they are similar to the feedstock of the coins in the following row, since it is thought that they were made of melted church plate, among other things, during the crises of the Persian and Muslim threats to the empire. 

The second row shows examples of the earliest North European silver coinage since the departure of the Romans.  From left to right, the first is from Northumbria circa AD 710, a type “J” penny attributed to Osric.  The obverse shows two facing profiles flanking a standing cross;  the reverse a whorl of birds.  

The next coin is from Kent, circa AD 690, attributed to Wihtred.  The obverse is a diademed bust;  the reverse is a bird on a cross.  

The third coin is from Mercia (or Essex?) circa AD 735, attributed to Aethelbald.  A female centaur graces the obverse, and a wolf whorl the reverse.  

The fourth coin is from Denmark, around AD 700, with a bearded visage usually believed to be Wodan, and a fantastic beast on the reverse.  Believed to be the first Danish/Viking coin.  The crosses on the obverse may not be intended to be Christian symbols, but be merely apotropaic.  The area where these were struck was highly pagan when these were produced.  (I always wanted to use the word apotropaic in a sentence, and now I have!)

The third row is comprised of coins of Louis the Pious.  The first three are Class 2 deniers.  The first is from Bourges;  the second is from Tours where the great battle was fought by Charles Martel only 66 years before.  The third came from the mining town of Melle (Metallum).  If the article is correct the silver from which all these deniers were minted may have originated there.  Because these are Class 2 deniers, they can be precisely dated to AD 819-822.      

The final coin is part of the so-called Temple coinage (AD 822-840) with the legend XRISTIANA RELIGIO, also called Class 3.  These coins of Louis the Pious were of almost identical design throughout the Carolingian empire, which doubtless served as a powerful statement of unity and centralized power.  Unfortunately the lack of mint marks renders them less valuable to archeologists and less interesting to collectors.  

The fourth row contains coins of Charles the Bald, AD 840-877.  Here, mint designations have returned, and we see examples from Paris (PARISII CIVITAS), Le Mans (CINOMANIS CIVITAS), Quentovic, an important trading port (QVVENTOVVICI), and St-Denis, the burial site of the French kings (SCI DIONVSII M)  for the mint of Saint Dionysus who is buried there.  

Some very nice coins. The sceatta with the centaur (Series S Type 47, possibly from Essex) is difficult to get well struck and in good condition.

[Hrefn]

;@Rand, Yes.  I was trying to make that precise point, that the Byzantine hexagrams and the Saxon sceats came from the same feedstock, which was old plate.   I doubt roughly contemporaneous hexagrams were melted for sceat production.  Rather, when both East and West turned to silver coin production, they both drew from the only available and substantial source.  And that was old Roman silver.  One object from the Mildenhall treasure alone would have yielded material for about 7,000 sceats.  

Given the mobile nature of the Roman upper classes, who might spend time in Numidia, Britannia, Judaea, and Armenia during the course of a professional career, silver objects they accumulated could come from anywhere. Some of them would be awarded by the emperor as donatives.  The empire’s silver would have have been fashioned, pooled, and remelted, coined, disbursed empire wide, then recollected, etc.  Over the course of centuries it would not be surprising to discover Roman silver acquired a degree of homogeneity.  

Collectors of antique silver often bemoan the tendency of households in the past to scrap their old-fashioned or out-of -style silver objects to have them refashioned into the latest styles, but this was a regular occurrence and still occurs.  There is no reason to expect the Romans differed on this score.  Most collectors of antique silver today would point with pride to any object in their collection which was 200 years old.  Very few persons have any silver which is 300 years old, and hardly a single private person would possess a silver object which is 4 centuries old (coins excepted.)  The old silver, with the exception of objects concealed and forgotten, has all been melted, as indeed has an incalculable amount of obsolete silver coinage, all furthering the process of homogenization.  In the days of the Roman Empire, it conceivable that an Armenian spoon could end up in Gaul, and be flung into a crucible with a broken harness fitting from Hispania and a handful of obsolete Gaulic silver coins.  And this process of recombination persisted for centuries.  

[John Conduitt]

This statement is interesting:

"Between 660 and 750 AD, Anglo-Saxon England witnessed a profound revival in trade involving a dramatic surge in the use of silver coins, breaking from a reliance on gold. Around 7,000 of these silver 'pennies' have been recorded, a huge number, about as many as we have for the rest of the entire Anglo-Saxon period (5th century–1066)."

They are right. The PAS database has about 2,500 coins called 'sceat' or 'sceatta'. I don't doubt 7,000 have been found. Maybe vastly more - AC Search has 4,400. Series E is very abundant. Given the wide range of styles and dies, there must have been hundreds of millions of them struck. That's a lot of Great Plates of Bacchus. Tens of thousands of them, in fact. It seems inconceivable that they were not using old Roman silver (and anything else they could find or steal) to get to that number.

It is also a bit misleading. The reason there was as much produced in the 90 years between 660-750 as the rest of 410-1066 is because nothing was produced between 410-660 and next to nothing between 800-980. Aethelred II, Cnut and Edward the Confessor were at least as productive in the last 90 years to 1066. The question might not be why so many coins were struck between 660-750 but why they weren't between 410-660 (because of the collapse of the Roman Empire) and 800-980 (because of the collapse of Mercia's rule over the other Saxon kingdoms). The fact that coins were produced where there was a stronger regime is reinforced by Northumbria, where more coins were produced between 800-900 than everywhere else in England put together, albeit in debased silver.

So there might not have been a surge in silver at all. Just in economies that might make use of it.

[Rand]

These are certainly interesting contexts.

I hope (from a collector's perspective) that a small number of coins was still produced in Britain in 410-660, perhaps after 500, when there was some trade with the continent and some Frankish settlers were moving to Kent. Anglo-Saxons may even have participated in Italian Ostogothic wars and wars in Gaul in the next decades. This would explain some very rare gold coins of the period found in England. Any local produce would follow that of Franks/Visigoths (so gold coins) and thus may be difficult to distinguish. Still, a number of coins found in England are not known from the continent.

[Rand]

Today, I had a chance to get further insights into this study, which was presented and discussed by the author, Professor Rory Naismith, at the 9th International Symposium in Early Medieval Coinage in Cambridge.

The interesting points from the discussion were:
- The team did have information about old Roman silver from other analyses and the reported coins matched Byzantine (e.g., Syria) and not old Roman silver.
- A small amount of gold in the alloy is explained by melting gold-plated silver.
- There is an ongoing study by another group showing that Saxon leaders sent soldiers to Byzantine for training (which is relevant to my interest in whether mercenaries from the British Islands could have participated in Theoderic’s Italian wars).
- There were economics-driven reasons why Byzantine silver was not melted for coins earlier, and gold continental and later English coins were used instead.

I feel more convinced now that the silver was from Byzantine.

[John Conduitt]

59 minutes ago, Rand said:

Today, I had a chance to get further insights into this study, which was presented and discussed by the author, Professor Rory Naismith, at the 9th International Symposium in Early Medieval Coinage in Cambridge.

The interesting points from the discussion were:
- The team did have information about old Roman silver from other analyses and the reported coins matched Byzantine (e.g., Syria) and not old Roman silver.
- A small amount of gold in the alloy is explained by melting gold-plated silver.
- There is an ongoing study by another group showing that Saxon leaders sent soldiers to Byzantine for training (which is relevant to my interest in whether mercenaries from the British Islands could have participated in Theoderic’s Italian wars).
- There were economics-driven reasons why Byzantine silver was not melted for coins earlier, and gold continental and later English coins were used instead.

I feel more convinced now that the silver was from Byzantine.

Thank you for the update. Did they give a reason for the sudden use of Byzantine silver? What were the reasons for hoarding specifically Byzantine silver until that point? It seems difficult to believe the Saxons had a sudden change of heart, switching from only melting down Roman silver to only melting down Byzantine silver. Why would the care whose silver it was?

I don't think the question was whether they literally melted Roman silver, but that by then any Roman silver would have been recycled with everything else and become unidentifiable as Roman silver. This would indeed give you silver with gold in it. The same recycled silver you'd find in Byzantium.

If the Saxons did send soldiers to Byzantium for training, it surely wasn't in huge numbers. They may have fought in the Italian wars, but that wouldn't significantly impact the supply of silver in northern Europe to the extent they could suddenly strike millions of coins.

Edited April 20 by John Conduitt

[Rand]

I am afraid this was information that I got. I did not ask questions myself. 
There was a specific comment on the sudden use of Byzantine silver, which could be due to the changes in economic relationships, now involving smaller agricultural landowners and other increases in trade, which needed silver rather than gold coinage, and it became profitable to melt chunky silver items. There was a comment about a large amount of stored Byzantine silver items when several big silver plates could provide silver for c. 10,000 coins. The sudden trend of melting Byzantine silver may explain why many coins were produced from Byzantine rather than mixed silver. This does not explain what happened with recycled Roman silver. The overall numbers of the produced coins were estimated as very large, but not in millions as was mentioned on our thread (I am no expert in these coins to have a strong personal opinion).