Maybe it was Pi, which is to say, there is no answer.

I suppose "the inscriber started to add a third saying, realized this would violate Meden agan, and stopped after the first letter" is not a satisfying explanation?

A minor nitpick, but isn't the reference to "a capital epsilon" anachronistic? My understanding is that the upper-case/lower-case distinction in the Greek alphabet didn't arise until centuries later.

The fact that all movies were black-and-white for decades doesn't make it an error to say that early movies were black-and-white.

I suspect it wasn't originally a letter at all.

KCinDC, I don't think the analogy is a particularly apt one. The color/B&W distinction has an external reality to it which upper-case/lower-case does not. The latter is an aspect of the symbol system in question, and has no reality outside of that.

Let me present a hypothetical situation. Suppose that, at some point in the future, English develops a dual number, and that the word "oxen" is assigned to that number, the plural form being "oxes". Would it be accurate to speak of (pre-change) "oxen" as being dual, especially in a context where there was nothing to distinguish whether the reference was to two or to more?

Jim: While I understand your nitpick (and believe me, I love nitpicks), I think in this instance your persnicketiness is misplaced. The point of adding "capital" is not to imply that the distinction existed back then but to help modern readers, who are primarily familiar with the lower-case Greek letters and thus think of ε as "epsilon" (and think of E as "capital E," so it's helpful to remind them that what looks like an E is actually a capital epsilon).

That explanation did eventually occur to me, yes.

In another work, Plutarch gives the third maxim as a whole sentence: 'Give a pledge, and disaster follows'. I posted about it here.

Short for Έιςοδος / Έξοδος?