The Underlying Structure of a Coin

Style is the art of the die used to strike coins. 'Fabric' is the sum of technical details that make up everything else about the way our coins were made. These are mostly details of planchet preparation that have left artifacts on the final product. Most commonly, we see reference to 'thick' fabric or 'thin' fabric but we shall see that the matter go far beyond a simple measurement.

Most ancient coins were struck from dies on blanks that were prepared for use by a number of different methods. Most commonly, the blanks were cast in some sort of mold producing a coin shaped disk lacking only the design that would be added by the dies. Some mints worked the castings to make them more regular or more easily struck. This could be as simple as filing off rough edges and casting sprues or could take the form of chasing the surfaces on some sort of tool. Other mints might have cuts blanks from sheet metal or hacked irregular pieces from flattened lumps of metal. In many cases, our understanding of the techniques used is incomplete. Perhaps the most important thing to understand is that each mint was individual and developed these techniques as they saw fit. What is true for one time and place will not be true for another. The little clues of 'fabric' sum together a signature of the mint that identifies it and allows the student to recognize what is normal, correct and genuine. This can be invaluable when confronting coins that may be questionable or counterfeit.

This page will take the format of query by example. We will examine a few coins that show some 'fabric' characteristic that is distinctive. Remember, however, that most other coins that are less distinctive will show these clues to some degree. Fabric signatures, like handwriting signatures, vary in style, beauty, individuality and the ease with which they are read. It is hoped that this page will cause some readers to examine their coins for these clues and appreciate the work that went into their manufacture.

Most simple of fabrics is the selection of a lump of metal to use as a blank. Our example stater of Aegina (6th century BC?) seems to be quite irregular and unworked but the extreme accuracy of weight control shown on these issues proves that the pieces were trimmed down to size. Evidence of trimming remaining on the coin would seem to be the flat surface on the edge shown on the upper left of the reverse photo. Most turtles are found on smooth and round lumps so this one may be unusual or a casting error that was used anyway. The flan is quite thick and was struck with small enough force that the basic 'nugget' shape was not erased. Relatively few turtles were struck well enough that the flan spread into all of the die and shows head, tail and four legs. This tendency is also part of the 'fabric' of the series. Individual coins can not be examined in isolation but must be seen as a part of the larger picture seen only by looking at a few thousand similar examples.

Another Greek coin, a tetradrachm of Syracuse in Sicily of the 5th century BC, is also of 'thick' fabric. It was stuck on a ball of metal that was cast (possibly in sand?) as part of a group connected by thin channels (sprues). On this coin the remnants of one of these sprues shows at the lower right of the reverse photo. Not all coins of this type will show this irregularity. Since the blank was a globe it was possible that the flan would fall on the die with the sprues up and down so they would be erased with the blow of the hammer. Only when, as here, they fell to the side would a trace be left after striking. The reverse die used for this coin (Boehringer 703 (V345/R481) late state) was badly broken (behind the head, upper left) but was kept in use long after most mints would have thrown it out. It amazes me that people who could produce such beautiful art of the die showed so little care in the technical aspects of coin production. Perhaps the die cutter was busily cutting a new die but rushed demands of production required the strikers to keep using this one. This coin is struck very unevenly. Is it too much to suggest that the workers were intentionally favoring the broken side hoping to avoid a catastrophic break? The coin is very flatly struck in many places but still has a nice face on the head of Arathusa. Part of the fabric of this series is the failure of the striking to spread the metal into all parts of the obverse die. Coins showing both the horse heads and the charioteer are worth a premium. Transferring the high relief beauty of the die to each individual flan was beyond the skills or priorities of the mint. This failure must also be considered part of the fabric of the issue.

Diametrically opposed to the thick fabric, high relief Greek coins is this Sassanian silver of Shapur I (241-270 AD). The ornate design is very low relief and struck on a flan with minimal (1mm) thickness. The flan was trimmed to be perfectly round. Equal in diameter to the 16.7g Syracuse tetradrachm, this thin coin weighs only 3.9g. Careful attention to detail and selecting techniques that were within their abilities enabled the Sassanian mint to produce a fine product. Later, the desire to make the coins look even larger without actually being heavier would lead to very thin flans that usually show severe flat spots in the design. Our example was produced in a time of a technical peak that would rarely be equaled until the advent of modern coinage techniques.

Much earlier but just as perfectly round is this 36mm bronze of Ptolemy III (246-221 BC) of Egypt. The flan was cast as a flat disk of considerable thickness and tooled to be perfectly smooth and round before striking. Remnants of this tooling process, centration dimples, have been discussed previously as they were used on Roman Provincial coins. Notably different in this Ptolemaic version is the care with which the coin was made perfectly round. The edge view of this example shows a groove running around the edge which is not seen on every coin but is evidence that the coin was trimmed in a rotary fashion. Smoothing the surface allowed much less striking force to be used than if the hammer were to be required to erase surface irregularities as well as to transfer the design. Many large Ptolemaic bronzes are weakly struck; producing so large a bronze coin required careful controls and heating of flans. Considering the technology available at that time, these coins are of amazing quality.

Another very round coin shows a different method of edge smoothing. Struck at Antioch in Syria (possibly from dies prepared at Rome??) this 30mm bronze coin of Philip II (247-248 AD) not only is perfectly round but shows a flat edge with squared corners (light line on the diagonal view). Most noticeable, certainly, is the lack of a centration dimple. The close up shows the edges of this coin (bottom) and coins produced by the same technique (moving up the pile: Gordian, Singara; Macrinus, Antioch; Philip I, Zeugma). Note that each shows file marks at a slight angle across the edge of the flan. The marks all seem to go the same direction and strike me as what would result if a right handed person were applying a file to a stack of coins held in a vise. In truth, I have no idea how these flans were prepared and the evidence presented here could probably be interpreted in other ways. Coins of this fabric are common to the Syrian and Mesopotamian cities of the 3rd century AD. Compared to the irregularly shaped sestertii often seen from Rome, the workmanship of the coins is impressive. Note that all are of low enough relief that stacking is quite possible. Just try to stack your 12 Caesars sestertii and you can appreciate why bankers must have appreciated this technology. Please don't tell me that coins with file marks on the edge are cast fakes. These coins are all quite genuine and definitely produced by striking. The same scratch evidence found on Athenian silver would be a good sign of a bad coin. The rules are different for the two issues and collectors must learn which standards apply to which coins. That is all part of the fun of being a student numismatist!

How were the flans of ancient coins cast? By now you should know I will not be giving just one answer since many different methods were used across the time and space that made up what we call antiquity. This space can not cover them all but we will visit some of more than the usual interest.

A bronze drachm of Antoninus Pius was issued from the mint of Alexandria, Egypt, in 153-154 AD. It shows the typical fabric of Alexandrian bronze coins. Flans were cast by pouring puddles of metal into shallow cups in an open top mold (probable carved in stone). The extreme slope to the sides of the cups allowed easy removal of the cooled blanks from the reusable mold. The resulting coin is considerable more broad on the reverse than on the obverse. The lack of metal around the obverse edge of these flans rarely allowed proper transfer of the entire obverse legend. Few Alexandrian bronze designs had reverse legends beyond the date (here LZ=year 7).

A similarly cast tapered flan was used to produce this silver tetradrachm of Macedon (under Roman rule) c. 150 BC. The added feature of note is that the die was cupped to match the taper of the blank so the details of the fancy border wrap around the edge of the coin. This produces a beautiful result compared to the poorly struck coins resulting from the flat dies used on the Alexandrian dies. The cupped dies made centering of the blanks easily accomplished so off center coins of this type are not common. This seemed like such a good system, it is strange that it died out after this one use. The work involved must have been more than I realize. Again, perhaps, the ancient mind did not expect coins to be perfectly round and smooth so the advantages of this fabric would have seemed insignificant.

Returning to Alexandria we see a bronze (potin) tetradrachm of Probus (year 2=277 AD). By the time of this coin the bronze drachms were no longer being made and the 'silver' tetradrachms contained only a trace of the precious metal. The fabric used for these late tetradrachms is quite unusual. The flans were quite thick and show a particularly rough edge. It appears that the edge is a remnant of the coarse sand used to mold the blanks. Fields of the coins are usually smooth with no sign of the texture. One answer would be if round cast blanks were hammered into disks before being struck by the dies. This would smooth the faces of the coins without changing the edges. I am not aware of scientific studies on the techniques used to produce these coins.

A discussion on varieties of fabric could go on forever and provide a fascinating area of specialty for collectors with a 'technical' bias. Various pages on this site have shown other oddities of fabric but even the most normal coins (not shown on this page!) show clues of fabric that are important to identifying and authenticating various series of coins. Understanding which methods were used for which coins will help the collector appreciate what to expect from the coins. These characteristics are not necessarily plusses or minuses but simply parts of what makes these coins worthy of our study.

Back to Main page

(c) 1999 Doug Smith