The German Wikipedia [1] has a slightly different variation than MW and you which makes a lot of sense to me:
The letters which were also words were prefixed with "per se" (it was not bracketed like MW says "I per se I" - why?).
So, when schoolchildren recited the alphabet, they'd say
"per se A, B, C, D, E, F, G, H, per se I, ...".
Now, the "&" was at the time considered the last letter, and it was introduced by "and", as the last item of a list often is ("Peter, Paul and Mary"), thus the schoolchildren would end
"... W, X, Y, Z, and per se &." [1]
French, similarly, ended in "X, Y, Z, et per lui &."
That seems to me a very plausible theory for ampersand and esperluette.
Note, however, that the English Wikipedia mirrors the "A per se A" explanation from MW.
The letters which were also words were prefixed with "per se" (it was not bracketed like MW says "I per se I" - why?).
So, when schoolchildren recited the alphabet, they'd say
"per se A, B, C, D, E, F, G, H, per se I, ...".
Now, the "&" was at the time considered the last letter, and it was introduced by "and", as the last item of a list often is ("Peter, Paul and Mary"), thus the schoolchildren would end
"... W, X, Y, Z, and per se &." [1]
French, similarly, ended in "X, Y, Z, et per lui &."
That seems to me a very plausible theory for ampersand and esperluette.
Note, however, that the English Wikipedia mirrors the "A per se A" explanation from MW.
[1] https://de.wikipedia.org/wiki/Et-Zeichen#Englische_und_franz...