I enjoy how these two coins together illustrate the variety of acceptable forms of the letter A in the Latin alphabet at the end of the third century. That on each specimen two very different forms of the same letter co-exist warns the epigrapher against using these letter forms alone as a dating criteria. It also suggests to me that certain names were rendered in particular ways habitually. Compare for instance the VAR ligature of RRC 126/1 with that C.VAR ligature of RRC 74/1 (links to BM specimen). ROMA has an open single bar A because that’s just how the word looks right.
I was thinking about the use of legends on the republican series to label or describe the images. The above type is a nice example of the use of abbreviation to do so. We can resolve it based on longer inscriptions in other contexts.
I wanted to think about the phenomenon over time and by type of usage. So I created a little color coded chart. No promises I didn’t miss a few or mis-transcribe a couple, I didn’t want to give this too much time.
My groupings are pretty subjective, pink for hard to recognize gods, peach for divine qualities/virtues — aren’t those two pretty much the same thing? Green for Kings, but then I threw in Faustulus because where else would he belong in these categories? Blue for the accomplishments of individuals. This can be hard to separate from the moneyer’s name and titular. A slightly darker blue for military accomplishments and a slightly lighter blue for religious acts. Purple is for buildings and monuments, but not statues. Mud is for personifications of place. I’ve left uncolored others that are not readily paralleled elsewhere in the series.
I’d say 63 BC onwards is the real ramp up in this coin epigraphy if we can call it that. The trend is towards longer more complete legends, rather than just ‘helpful hints’. It’s still a long way from the Imperial habit of labeling most reverse types.
[Of course, I’m also ignoring the question of when ROMA is labeling the goddess and when its indicating the minting authority. And, yes, I just gave up transcribing at 431/1. I’ll save the file and come back to it later, if it should prove useful.]
Update 1/11/16: The Φ on RRC 293/1 should have been on the above chart!