This site is meant to provide ordinary collectors of ancient coins with a basic guide to the earliest coins in the western world, namely the electrum issues of Ionia and Lydia, and the gold
and silver coins of Croesus (Kroisos).
Early Electrum Coinage.
Most collectors of ancient coins are naturally interested in the early electrum issues of Ionia and Lydia. However, they are faced with the problem that we still have very limited knowledge of exactly
where and when these types were struck, so that a definitive study of them is not as yet possible, and hence there is no single comprehensive and up to date catalog of these types to which collectors can turn*. As a
result ordinary collectors often see these coins as a bewildering confusion of unattributed types, and hence they hardly know where to start when trying to put together a meaningful collection of the period. Not
surprisingly this is enough to discourage the interest of many collectors, particularly as even the smaller examples of these types are not exactly cheap.
Therefore, in order to give collectors a better overall understanding of these types and how they relate together I have taken the limited evidence that we do have to go on and put together
some charts of images of the key early electrum issues, ordered in space and time, down to roughly the fall of Croesus. These charts can be found on the "Early Electrum Coins" page.
Ultimately of course the basic aim of the study of these early types is to determine when and where (and why) coinage was first introduced. I'm not going to try to answer these questions
definitively here myself, but I hope that the data as presented here will help readers assess the various theories that have been put forward over the years, and to understand the problems still to be solved in this
key area of research. (For more details see the Appendix below "The Introduction of Electrum Coinage").
I will say however that, taking the charts at face value, it would seem likely that coinage was introduced in two phases. Initially we have unfigured staters from Samos (or at least a mint
on the Samian standard) and Ionia B (probably Miletus**), together with a limited range of associated binary fractions - 1/2 and 1/4 staters, plus the odd 1/8 stater^. Then after a short time it seems that a
new series of denominations was introduced, based on the 1/3 stater, with fractions ranging down to 1/96 stater (and even 1/192 stater). Just where and when these new denominations originated is not clear, but they
were soon incorporated into the striated series at Miletus and the issues of Samos, and, as the "smooth (and rough) unfigured types", they make up what seem to be the earliest regular issues of Ionia A (Ephesus?).
At Miletus it seems that for a time 1/2 staters were still issued alongside the new 1/3 and 1/6 staters, as they share punches with the new types, but we notice an interesting fact -
although most of these 1/2 staters have three reverse punches, some feature only two^^. What does this mean? One possible interpretation is that the two-punch 1/2 staters represent the earliest versions of
the type at Miletus, and that the change to three punches was made in order to distinguish the later 1/2 staters from the new 1/3 staters, which now appropriated the two-punch reverse. (At Samos there seems to have
been less of a problem, although one early series did in fact mix a 1/3 stater with very similar 1/2 and 1/4 staters, but after that the 1/3 stater was dropped and new fractional denominations were based on the
1/6th stater, which was easily distinguished from the earlier types).
In any case, and whatever may have been the exact sequence of events, after a time it is clear that the 1/2 and 1/4 stater fractions were discontinued everywhere, and the 1/3 stater (or in
some areas the 1/6 stater) became the effective basis of the currency throughout Ionia - full staters were still issued, but except at Miletus these were probably struck mainly for prestige reasons, or perhaps in
some cases as commemorative types.
Given this progression of types, it is difficult to see how coinage could have started with the lion-head types in Lydia, as has often been assumed, as all the "Lydian" lion-head series
were clearly based on the 1/3rd stater (including the putative Lydia C and D series, the rare full staters of those mints notwithstanding). In fact the earliest Lydian lion-heads would seem, on the basis of their
general style (and as we now know, similar gold content), to be roughly contemporary with the "panther" face type of (possibly) Miletus, which was hardly the first issue of Ionia#.
(In any case the Lydian types can hardly have been primary, as then the earliest Ionian types would most likely have been based on the 1/3 stater from the start).
So, where did the 1/3 stater types and their derivatives first appear? It's tempting to think that they first appear as the smooth unfigured types of Ionia A, but there are problems with
this idea, as these types would seem not to have been the earliest issue of Ionia A, but rather the successors of the similar but different (and much less common) rough unfigured three-punch staters and 1/2 staters
referred to in the charts here as the Boston MFA types. These latter types, which probably derive from Ephesus or Ionia A, presumably coincide with the similar striated three-punch 1/2 staters of Miletus, which in
turn, as we have seen above, seem to mark the appearance of the 1/3 staters at that mint. In other words this (admittedly rather tenuous) line of argument seems to suggest that the 1/3 staters appear at Miletus no
later than at Ionia A/Ephesus, although this conclusion can only be tentative at best.
Why were the 1/3 staters introduced? Again we don't really know, but one possibility is that it was connected with the weights of the earliest coins. Thus the earliest Milesian staters
weighed c.14.5 gm, i.e, they seem to have been didrachms on the Euboic scale, weighing 1/30 of a Euboic mina of 436 gm##. Thus the Milesian stater would have had a weight of 12
Euboic obols. A trite would then weigh 4 obols and hence the weights of its binary fractions would have been convenient binary multiples or fractions of an obol (in which case the coins could also do double duty as
weights). Similarly, if the Samian electrum stater of 17.45 gm is taken to be an Attic style tetradrachm (1/25 of a Euboic Mina divided into 100 drachms), then it would have had a weight of 24 Samian obols, so that
the key Samian denomination of 1/6th stater would have weighed 4 Samian obols.
As well of course, if the coins were treated as fractions of a stater then ternary fractions like 1/3 and 1/6 can be combined to yield binary fractions like 1/2 and 1/4, but not vice versa.
One thing we can say is that the introduction of the 1/3 stater and its fractions probably entrenched coinage (rather than bullion or other goods) as the basis of the money economy. The
earlier 1/2 and 1/4 staters were still too large to be useful in everyday transactions for most people, but the smaller denominations of the 1/3 stater series could well have been used in the market, at least at the
wholesale level, and would presumably also have been more convenient for the clearance of debts by ordinary people at the retail level. (At this stage households would generally have run accounts with local
shopkeepers and tradesmen which were settled periodically. Electrum coins could then have been used to settle accounts, while odd purchases for small amounts would still have been covered by uncoined silver or
bronze.). Thus the need for a simpler method of retail debt clearance may well have been the reason for the introduction of the smaller coin denominations in the first place, which may in turn explain, as we have
seen, the shift to the 1/3 stater.
* Although a project to produce just such a catalog, the "Catalogue of Archaic Electrum Coinage" (CAEC), is now underway. It is unclear however whether this new catalog will be
anything more than an updated version of Weidauer, i.e, basically a compilation of current museum holdings, or whether it will include the large amount of new material that has appeared in the market in recent years.
For the present collectors have to make do with Michener's "Ancient Trade and Early Coinage" (ATEC), which is fairly comprehensive but somewhat out of date and rather speculative in many
areas (and not easily accessible to most collectors).
** The actual origins of many of the types listed here are not known with certainty, so that mints are designated as Ionia A, Lydia B, and so on in the charts.
For examples of full, 1/2 and 1/4 staters from Samos see Triton VIII 438-441. Other examples from Samos and Miletus are shown in Linzalone's "Electrum and the Invention of Coinage", noting that Linz. 1009 and 1010 are actually 1/4 staters of Samos.
^^ For a two-punch 1/2 stater and the corresponding (punch-linked) full stater see Triton VIII-433 and Triton VII-242 (shown here as the top two coins of Ionia B in the "Basic Electrum Types" chart). Triton IX-941 is presumably a 1/4 stater from the same period. There seem to be no corresponding 1/3rd staters (although these are all rare types, and CNG 72-760 could be a 1/6 stater from this period, or immediately afterwards). For the same or similar coins see Linzalone 1026, 1018 & 1032. For another two-punch 1/2 stater and a linked full stater see the oddly styled Heritage NYINC Jan. 2013-21229 and Baldwin's 70-20.
I would guess that that the earliest opposing lion-head types of "Lydia A" (attributed mostly to "Walwet") were inspired by the similar but much less common lion-head types assigned here to Ionia J, which is possibly a Lydian, or at least Lydian controlled, mint in Ionia. These latter types presumably evolved from the later striated issues of Ionia B where opposing lion and human heads were engraved on the large size striated obverse die plates used for the earliest staters. These obverse dies could perhaps then be the ultimate model for the oversized obverse dies used for the later two-headed types of "Walwet".
## As noted elsewhere, after not too long the weight of the Milesian stater declined to the Lydian standard of 14.15 gm, and eventually even less.
The first Gold and Silver Coinage.
By the middle of the 6th century B.C, and quite possibly earlier, the Lydians had mastered the technique of separating electrum into gold and silver, and it is usually thought that the
world's first gold and silver coins were issued at Sardis in Lydia during the reign of Croesus (c.560-546)*. These coins, with the facing lion and bull foreparts on the obverse, were evidently issued in various
denominations at various times, and in fact their production was probably continued by the Persians for some time after Cyrus the Great had conquered Lydia. Eventually they were replaced in the late 6th century by
the gold daric and silver siglos of Darius featuring the sitting or running king.
Several writers, notably Berk, Naster and Nimchuk, have attempted to classify the various issues of the Croeseid types, with, implicitly at least, the ultimate aim of arranging them in time. The
results of these efforts have been limited, but nonetheless it's fair to say that significant progress has been made in defining the various issues, at least in broad terms. Unfortunately the key literature on these
types is not easily available to the average collector, and therefore I have put together a chart of images of the different issues, arranged in time, which can be found on the "The Coins of Croesus" page. This chart is based on the ideas of various writers, but ultimately it reflects my own assessment of the available evidence.
* It is not impossible that the first "Croeseid" issues may have actually preceded the reign of Croesus, but for present purposes we will adopt the conventional dating.
Appendix - The Introduction of Electrum Coinage.
For a clear and reasonably up-to-date discussion of the basic problems of the early electrum coinage see Velde's "On the Origin of Specie", which can be found in various forms on the
internet via the usual search engines. (The most usable version is probably the pdf file from the Institut de hautes études internationale at Geneva). For a more general view of the origin and role of coinage
readers should also familiarise themselves with the ongoing "chartalism" versus "metallism" debate (don't be put off by the academic terminology - it just means token or fiduciary money versus real-value money).
More specifically, for the moment the most popular explanation of the introduction of electrum coinage, due originally to Wallace, is that electrum coins likely originated as a way of
standardising the use of natural electrum as a means of exchange. As more or less pure materials, gold and silver could be simply exchanged by weight, and silver in particular had for centuries been the prime medium
of exchange in the market. The varying composition and hence uncertain value of natural electrum created obvious difficulties in using it as a medium of exchange, but nonetheless it seems that the richer forms of
electrum were customarily assumed, at least in some places, to contain the equivalent of 75% gold, and were exchanged with silver at a rate of 10:1, and gold at a rate of 3:4 (implying a gold/silver value ratio of
40/3)*. However, at some stage this system seems to have broken down, possibly due to dilution of the natural electrum with added silver, so that a new more formalised approach was needed to restore the reputation
of electrum as a medium of exchange.
A system of certification was then introduced (so Wallace suggested) by which electrum lumps of fixed weight but uncertain actual composition were stamped and issued by the state** at some
defined nominal value (in terms of gold and silver bullion). But what value? In his initial exposition of this theory Wallace doesn't specify this but clearly he assumes that the certified lumps will be token coins,
i.e, that they will be significantly overvalued against their real value (whatever that might be). Thus the actual composition of the alloy, and hence the intrinsic real value of the coins, is largely immaterial –
all that counts is the issuer's guarantee of nominal value, presumably underwritten by a promise of redeemability, and certified by the stamping of the lumps.
This is of course basically an updated version of Sture Bolin's famous thesis that the debased electrum coins were essentially a state swindle, although in this new version the people were
apparently supposed to knowingly accept and use them as credit tokens.
But would ordinary people have accepted such an advanced concept at this stage in the development of coinage? Possibly, although my impression is that overtly token coins were very much the
exception rather than the rule in ancient coinage, particularly in the early stages (for the primary coinages in precious metal at any rate - at a somewhat later stage the secondary issues in bronze quickly became
tokens in most places).
Because of this I incline to a slightly modified form of Wallace's idea. The earliest coins must surely have been intended to be valued in terms of gold and/or silver bullion, and the most
obvious way to do this would have been to issue certified coins of (natural) electrum, tariffed at the traditional value of 10 times silver, or 3/4 times gold. The Milesian stater/didrachm would then have been
(nominally) worth 20 drachms of silver, and the trite a 1/2 drachm of gold (Euboic drachms at first, declining to the Lydian standard). The stamping of the coins would then then have been done to guarantee the
quality of the alloy as well as the weight, i.e, to reinforce the idea of the standardised intrinsic value of the electrum, and not just the redeemability of the coins.
But is this what actually happened - were the first coins (i.e, the earliest issues of Samos and Miletus) really made from natural electrum, or perhaps some artificial equivalent? Recent
measurements have partly answered that question, at least in the case of Samos, as two of the unfigured types of Samos have yielded gold content figures of 74 and 79%, consistent with natural electrum.^
But if the earliest issues of Samos did use natural alloy the situation soon changed. We also have figures for the second round of issues from Samos (the pebbly types) which show a wide
spread in the alloy composition, with the gold content ranging from c.80% (as for natural electrum) down to less than 50%. Clearly while some of these types may have used raw electrum, for the most part they were
made of a diluted alloy, although with rather poor quality control.
Overall, the variation of the alloy here suggests that the coins became fiduciary tokens, although still retaining the value of "standard" electrum.
Now whether the Samos situation was typical of other early mints or not we don't yet know (we need more data on the alloy content of the other mints), but in any case after not too long we
find that in most places the coin alloy appears to have been more or less standardised at values well below that of natural electrum, but with the average gold content of the alloy varying significantly from mint to
As well the alloys varied over time, and thus even for coins struck on similar weight standards we find quite different gold contents, varying, for example, from an average of c.61% for the
early striated types of Miletus(?), to 54% at Sardis and 41% for the later framed lion issues of, perhaps again, Miletus^. Moreover, in Lydia we find that by the time of the lion-head issues the gold content of the
alloy seems to have been quite tightly controlled, suggesting that, at Sardis at least, a completely artificial electrum alloy was being used, meaning presumably that the Lydians were now routinely separating their
electrum into gold and silver. (Samos, it should be noted, ultimately took a different approach and seems to have eventually largely given up issuing electrum coinage).
Now all this sounds reasonable at first sight, as it accounts for the generally diluted alloy of the coins, and overvaluation of the token coinage means that the variability of the natural
alloy is no longer a problem. But it leaves us with the odd situation of different mints using different alloy compositions for coins with (according to the theory considered here) the same token value. We have to
wonder whether this may have caused problems with the exchangeability of the coins from different mints.
In fact, there are still a few more problems with Wallace's theory. Firstly, as noted above, after not too long a time it seems that the coin metal (in Lydia for example) was actually made
by combining refined gold and silver to produce a purely artificial electrum, possibly because the natural electrum had run out, but more likely to improve the control of the alloy's composition. But if you now have
regular supplies of gold and silver, why go the trouble of making an artificial electrum - why not take the obvious step and produce gold and silver coins?
And finally, the prime mover in the introduction of electrum coinage is usually taken to be Lydia, which had, it is generally assumed, copious supplies of natural electrum, and hence a
vested interest in utilising them. However, the earliest actual electrum coins seem to have been issued, not in Lydia, but in Samos and Miletus. Why? – we don't know, and Wallace's theory gives us no clue.
Clearly we have much to learn about the rise of the electrum coinage, but another question is why did it ultimately fail, or to put it another way, why did the gold and silver coinage take
over? Without getting too specific, in general terms the problem may have been that ultimately people simply lost faith in the electrum coinage system, with its varying gold content and token value, and hence they
tried, like the Samians, to avoid it wherever possible (by reverting to weighed bullion for example). In the end then it is not surprising that most coin issuers were forced to finally switch to a more convincing
and overtly real value system, which meant, in practice, gold and silver coinage^.
Nonetheless, despite the ultimate failure of the electrum coinage system, the electrum pieces were the first to have their value certified by stamping, and as such they became the first
true coins, while gold and silver were still being exchanged as bullion.
Behind all this of course is an interesting further problem - what was the source or sources of the natural electrum in the first place (it now seems doubtful that the Pactolus river was
the main source of the Lydian electrum), and who controlled those sources - i.e, who would benefit from utilising them. These basic questions are yet to be answered.
* Note that an "equivalent gold content" of 75% includes the value of the silver content, and hence implies a theoretical actual gold content of somewhat less than 75%. In practice
though the ancients wouldn't have been too concerned about the exact composition of the alloy as long as the coins maintained their nominal market value in terms of gold or silver bullion.
It is possible that entities other than the traditional polis were responsible for the introduction of coinage. For example it has been suggested that coinage was used by wealthy merchants or populist tyrants to free up markets and trade, and hence wrest control of the economy from the domination of the aristocratic elites. On this theory electrum was used precisely because it was not favoured by the elites as a store of value or medium of exchange, and hence gave the ordinary people an alternative to the gold and silver "monopolised" (in this theory) by the elites. Just how much truth there is in this idea is hard to determine, but the spread of coinage would no doubt have benefited the "new men", i.e, those "merchants and bankers" whose wealth derived from trade and finance, rather than the land of the traditional aristocracy.
^ For more information on the composition of the alloy of the various types see Gitler et al, "XRF Analysis of Several Groups of Electrum Coins" in "White Gold - Studies in Early Electrum
Coinage" ed. van Alfen & Wartenberg, ANS New York 2020 (available on-line).
^^ Nonetheless some mints, such as Phokea and particularly Kyzikos, did persist successfully with electrum issues until the establishment of the Macedonian supremacy in the 4th century. It appears that while the gold content of these types may have declined somewhat over time, it seems to have been fairly stable in the short term, so that they could be traded internationally as real value coins.
N.B: Given the peculiarities of the various browsers, viewers may need to adjust the displayed text size for comfortable viewing.
29 Mar. '13: Chart of Northern Ionian types added.
16 July '13: Section on the Coins of Croesus added.
9 Jan. '15: Discussion of introduction of 1/3 staters revised.
9 Sep. '16: Discussion of earliest electrum coinage revised and clarified.
18 May '17: Possible role of tyrants in coin issuance noted.
31 May '17: Discussion of introduction of 1/3 staters further revised.
23 July '17: Appendix on the introduction of electrum coinage revised again.
14 Oct. '21: Some recent gold content data incorporated.
3 Nov. '21: Northern J mint revised to note Artemision coin.
12 Apr. '22: Some revision of Basic Electrum Types chart.
Contact me by email at: email@example.com