Crystallisation - is it a guarantee of authenticity?

[AncientOne]

Heavy embrittlement(crystallization) on a coin from Larissa.

[rasiel]

8 hours ago, John Conduitt said:

Here's some pseudo-science of my own. This is by sampling ACSearch to see if purer silver produces more crystallised coins. ...

There probably is a good seed here for a paper that could shed some light on this. Here's some points I think you could consider if you're going to go down that road (and I hope you do publish your results)

1) The purity tables are incorrect. Severan denarii might be, as an average, 55% silver but the average is meaningless given the wide (the very wide) distribution pattern during these years. Some denarii in this period test well above 90%.

2) Auctioneers preferentially sell high value samples. A crystallized portrait Julius Caesar is highly sought after while a Gordian III is going to get tossed in the group lots so the tally would need to account for this skewing.

3) How or who is determining these coins to be crystallized?

4) The Republican and Julio-Claudian as well as the tetrarchy/late 4th periods are reportedly the same purity yet have two very different results. This seems to undermine your thesis. Propose a hypothesis that can explain this.

5) Module thicknesses play a role. Thicker coins will hold it together much better than thinner ones. If siliquae are disproportionately rarer because they broke in half and never made it to auction this also affects the overall results.

Rather than relying on factors you can't control I think a better study design would simply have you buy a bunch of culls of these various periods, analyze them individually for purity and level of crystallization then tabulate the results. It's obviously costlier and more arduous but that would be real science. If you want to pursue it I'd be happy to collaborate with XRF analyses.

Rasiel