Nicholas Bodley wrote:

> My father's first language was Russian, and his neat English
> handwriting used overscores for [m] and [n]. [...]
> After he passed on, I later noticed that (iirc) Bulgarian and
> (definitely) Serbian script (even typeset text?) uses the
> overscore for (m), at least

Actualy, handwritten Cyrillic uses overscores over <т> (t) and <п> (p),
which look identical to Latin handwritten <m> and <n>, respectively.

The overscores are functional to distinguish these two letters from <ш> (sh)
and <и> (i), which look just the same to the two above, when handwritten.

In Latin script, a similar confusion may arise between <m> and <w> and
between <n> and <u>, and that's probably why your father borrowed the
overscores from his Cyrillic hand.

--
Cingar