I did a quick survey of the 2073's in the
CNG archives, and found 4 funny
types out of 92 examples (one the same as the 2nd
Obryzum coin above), the earliest from 2006. So if these are indeed
medieval types then they were not particularly
rare.
[Revised] I now have 7 examples of these
types from 5
obverse and 5
reverse dies, so there were probably a few dozen dies for each
side. These figures may well be consistent with the frequency ratio above, although I have not worked out the die statistics properly.
In any case these
types certainly constitute a well defined group of coins, quite different in
style and fabrication to the normal Nicaean issues of
John III.
It would be interesting to know the composition of their
alloy, and whether it matches the normal
John III types.
Ross G.
P.S. Below is a normal example of a "Latin"
type, this one with a diamond of dots. Note the double strike and inverted
reverse.