Technically Interesting Seleukid Coins
Unstruck Serrate Flan
Coin • Basic Observations • References
| Identification Number SXX-AE-01 | |
| Ruler: | Uncertain Seleukid king1 |
| Mint: | probably Ake-Ptolemais or Antioch2 |
| Period: | 2nd century BC |
| Denomination: | AE Unit |
| Weight: | 5.52 g |
| Diameter: | 15 - 16 mm |
| Obverse: | – |
| Reverse: | – |
| Die Position: | – |
| References: | Hoover, Seleucid Bronze Coinage Production |
Many Seleukid bronze coins produced at mints in Syria and Phoenicia bear small central cavities (dimples), sometimes accompanied by surrounding concentric circular scratches. Similar central cavities and concentric scratches are also found on many Ptolemaic bronze coins and in the bronze civic coinage issued by many cities in Moesia, Thrace and Asia Minor under the Roman Empire. Several possible explanations of these marks were published.3 The most plausible theory argues that cast flans were installed on a lathe in order to smooth its faces with a tool. This tooling was responsible for the concentric circular marks. By some authors, the cavities were either drilled or hammered into the cast flan so that it could be easily installed on the lathe. According to Welsh, Lathe Machining of Bronze Coin Flans, a centering pin of the lathe was forcibly impressed into the flan, creating the dimple. As the central cavities seldom match from obverse to reverse, a lathe with two centering pins was not used. The flans were probably turned on a rotating platen, similar to a potter’s wheel, and held in place by a single centering pin positioned over one of the central cavities.4 Nevertheless, as for the Seleukid bronzes, the true purpose of lathe turning is unclear yet.5
The following four observations are either in accord with the theory of lathe processing mentioned in the previous paragraph or at least do not contradict this theory. All these visual observations refer only to the above unstruck flan (ID Number SXX-AE-01). A reliable judgment of the first observation would probably require a destructive metallurgical microstructural analysis. The second and third observations are well-known from many other coins too. The fourth observation should be verified on other specimens to exclude a possible mistake.
Observation 1: The central cavities have raised edges, see Figures 1 and 2 (sides 1 and 2, respectively). It indicates that these cavities were forcibly impressed by a centering pin of a lathe or, less probably, somehow hammered into the cast flan.6 In any case, it seems to be sure that they are not artifacts of a casting process.
 |
 |
| Figure 1 | Figure 2 |
Observation 2: The center of concentric circular scratches do not coincide with the center of the flan. Compare white and yellow circles on Figures 3b and 4b. On each side of the flan, the white circle represents an approximation of the border of the flan without the serrate edges, whereas the yellow circles represent some of the clearest visible scratches. The white and yellow crosses represent the centers of the white and yellow circles, respectively. It is clearly visible that the white and yellow crosses do not coincide.
These auxiliary graphical objects were constructed as follows: Several points were chosen on the clearest visible scratches on large photos of each side of the flan. These points are denoted by red crosses on Figures 2a and 3a. Thereafter, these selected scratches were approximated by concentric circles via a computer program in such a way that the circles optimally fit the chosen points (the distance of each point from the corresponding circle is less than 1% of the diameter of the circle). The white circles representing the border of the flan were positioned manually.
It is necessary to emphasise, as Welsh notes on p. 5 of his study Lathe Machining of Bronze Coin Flans, that it is not clear whether the scratches are really concentric circles (such as would be cut by irregularities in a fixed tool) or a continuous spiral (such as would be cut by a tool moving across a rotating flan).
 |
 |
| Figure 3a | Figure 3b |
 |
 |
| Figure 4a | Figure 4b |
Observation 3: The cavities do not match each other from one side to the other, see Figures 5 and 6. The white circle represents borders of the central cavity on the given side of the flan, whereas the blue circle represents borders of the central cavity on the opposite side (imagine that the flan is transparent). It means that a lathe with two centering pins was certainly not used to smooth the faces with a tool.
These auxiliary graphical objects were constructed as follows: Three reference points were chosen on the edge of the flan such that these points are visible both form the obverse and from the reverse. Thereafter a composition of a rotation, of a translation and of a dilation was found by a computer program so that the reference points on the image of the obverse coincide with the reference points on the image of the reverse which was specularly reversed with respect to the vertical axis. Afterwards this planar transformation was used for computation of the position of the unvisible opposite central cavity.
 |
 |
| Figure 5 | Figure 6 |
Observation 4: It seems that the concentric circular scratches do not perfectly match with the central cavities (their centers do not coincide), see Figures 3b and 4b. To demonstrate this fact more clearly, the yellow circle on the first side of the flan was manually moved to coincide with a geometrical center of the cavity, see Figure 7. It is visible that the shifted yellow circle does not fit the points chosen on the circular scratch. Figure 8 shows the same experiment for the second side of the flan. The shifted yellow circles again do not fit some of the red crosses, although it is not so evident as on the first side.
The importance of this observation is not clear. It is possible that the bottoms of the central cavities are not exactly conical, so that a single centering pin of a lathe did not press the flan exactly in the geometrical center of the cavity. In any case, this phenomenon should be proved on other specimens with visible circular scratches and central cavities before any conclusion is made.
 |
 |
| Figure 7 | Figure 8 |
1 Serrate coins were minted from Seleukos IV to Alexander II. See Hoover, Seleucid Bronze Coinage Production, Table of Seleucid bronze fabric by ruler and mint.
2 Serrate coins were produced nearly exclusively at Ake-Ptolemais mint and Antioch mint, but some serrate issues are also known from Seleukeia on the Tigris, Tyre and Apamea. See Hoover, Seleucid Bronze Coinage Production.
3 See Hoover, Seleucid Bronze Coinage Production.
4 For a detailed description, see Welsh, Lathe Machining of Bronze Coin Flans, pp. 7-8.
5 See Hoover, Seleucid Bronze Coinage Production.
6 Drilling the cavity would tend to create a recess without a noticeably raised edge. (Welsh, personal communication)