I still have a bit of a problem with a clash of edges as I consider it implausible you could get that depth of impression from 28g of silver falling onto another edge, perfectly aligned so as to produce the incuse A as if it were part of the edge.
It isn't the same situation as a brockage, where you have a coin in situ to make the incuse detail on one side, all made using one of Boulton's presses utilising much more force than that obtained by gravity. If there was another coin only partially ejected, then I expect the collar to malfunction as the first coin would have to be in the same plane as this coin, and that would surely jam the mechanism. It doesn't add up to my way of thinking.
The collars used for this issue are 3 part (or at least my type example - 1819 no stops is so), with their ends joining in the middle of the gaps after DECUS, TUTAMEN and REGNI. That would be consistent with the clear joint line after TUTAMEN, but the other two are not obvious. Where the join should be after REGNI, is that a trace of a full height cross pattee, and in the gap after DECUS a smaller cross - or am I just seeing things? Crosses were used on some edges during William III and Anne on halfcrowns, but my crowns of these two reigns didn't use them (only used on halfcrowns?). I don't have a late Geo. II crown to say what was used on the issues prior to the new coinage. Help someone?
The earlier silver used the Castaign edging process, which in the case of a couple of my coins left a very sharp vertical cutoff in one place (with a step in height on the edge), as seen after TUTAMEN above and coincidentally also after TVTAMEN on some of the earlier coins in my trays. Was this consistently the starting or terminal legend on the strip and I'm wondering if could this be a one part collar? I can't see any trace of a vertical line in the appropriate place, but this could be due to wear. Check?
I don't think we have the answer yet, so more research required.