View the full J.E.A. Crake collection of Greek and Roman coins on Flickr

The collection consists of 24 Greek and Roman era coins and one holed “medallion” of as yet unknown origin. If genuine, the manufacture of this set of coins spans a time frame of approximately eight centuries (500 BC - 300 AD). Roman coins include 15 silver denarii, 1 silver antoninianus and 2 copper (or possibly bronze) coins. Greek coins are all silver and include 2 drachmas, 1 didrachm, 2 tetradrachms and 1 stater.

The coins and medallion are in the process of being non-destructively examined at the Digital Microscopy Facility, Mount Allison University. High-resolution spatially calibration color scans were produced in order to facilitate detailed examination of the coins as well as measurements of diameter and features without the need for subsequent handling of the actual pieces. An example is shown below.


High-resolution calibrated scan of Coin #22 - ostensibly a Corinthian stater dating to the early 5th century BC; weight 8.4 g. Open red square indicates region of SEM image of the same coin shown below.


Specimens were weighed to the nearest 0.1 gram on a digital scale. Elemental composition of each coin as well as that of inclusions in the metal and surface contaminants was determined using a JEOL JSM-5600 scanning electron microscope (SEM) equipped with an Oxford Inca 200 energy dispersive x-ray microanalysis system (EDS). Briefly, SEM/EDS allows the quantitative determination of elemental composition of selected regions of the specimen surface down to volumes as small as 1 cubic micrometer (µm3) and detection limits on the order of 1 part per thousand by weight (0.1%).

The 15,000 volt electron beam generated by the SEM interacts with the element(s) present in the specimen and generates x-rays with energies that are characteristic for those element(s). The EDS system collects the x-rays emitted from the specimen, and plots their number as a function of energy. An example SEM image and EDS spectrum are shown below. This process is completely non-destructive and does not add or remove any material from the coins.

SEM image

SEM image of a portion of the Corinth stater shown above. Open red square in image corresponds to area scanned by electron beam to produce EDS spectrum shown below.

EDS spectrum

EDS spectrum collected from area indicated in SEM image above. Silver (Ag) is the only metallic element present in this area of the specimen. Carbon (C), oxygen (O), sodium (Na), silicon (Si) and chlorine (Cl) are present either as impurities in the original metal used to produce the coin or as contaminants on the surface of the coin.


This work aims to address three main areas of potentially intertwined inquiry:

1) Authenticity. As the provenance of the specimens in the collection is apparently unknown, a first step in authentication is to determine whether the coins physically match the characteristics of known genuine specimens of the same type. Comparison of size, weight, principal metallic composition as well as details of the artwork on the coins to known genuine examples can eliminate fairly obvious counterfeits. For example, a coin known to be produced in silver by the Greco-Romans but revealed to be composed of primarily lead and/or tin would obviously disprove authenticity. Similarly, coins that vary significantly from published size and weight ranges would be immediately suspect. Note that such discrepancies may serve to disprove authenticity, but physical characteristics that match genuine coins do not necessarily guarantee authenticity. Once these physical characteristic requirements have been passed, however, the coins can then be considered for more formal authentication and preservation by third party services if warranted.

2) Origin of Metal in Coins. An intriguing possibility in ancient coinage is the investigation of sources of ores used in their production through examination of minor and trace elements found through SEM/EDS analysis. While the refining techniques of Greek and Roman metallurgists was surprisingly advanced, for large-scale production of coinage metal, some practical limits as to fineness could be expected and differences possibly related to metal origin. This may be particularly true for copper, bronze or similar base-metal coinages where the determination of purity and fineness of the metal was of less concern than for silver and gold coins.

3) Post-minting History. A potential problem in non-destructive SEM/EDS analyses is that only the surface composition of the coin (the outer few micrometers) is determined. There is some evidence that over time, the composition of outer layers of coin alloys is changed by preferential leaching of some elements over another. While this can complicate the goals outlined in points 1) and 2) above, it also provides the opportunity to investigate post-minting history in terms of not only loss of elements, but also addition of elements through exposure to various environments (e.g. burial in soil, exposure to seawater).

As of this writing, imaging and weighing of all the coins has been completed, and SEM/EDS data has been collected. At the major element level of composition, all coins have been determined to be plausibly correct for coins of their age and type. Positive identification of minor elements present and weight percent quantification requires further data analysis that will be performed in the next few weeks. This is particularly true in the case of trace metals in close proximity to the major elements (silver and copper) as major x-ray lines for these elements often overlap and require careful spectral reconstruction in order to determine their presence or absence and subsequent quantification.