Conrad uses Larmissiau in the genealogy section of his St John the Baptist Civil records for Marie-Josephe, wife of Urbain Picou. The very person we're speaking of.
I think the point I was trying to make is that merges are being held off on while what seems to be an arbitrary decision as to the best spelling is researched. Given that professional genealogists differ on the best spelling just goes to show how arbitrary it is. Just like with the Acadians, spelling varied, so does the spelling in Belgium.
I looked at the spellings in Belgium, Hainaut, Civil Registrations 1600-1913 at familysearch.org from the early 1800's are transcribed as
The earliest record is a 1796 death were the name is spelled Lermusiaux. The records all have images and are quite legible. There are a total of 14 records ranging in dates from 1796-1845 with most being in the 1800-1810 range.
So, no there wasn't one spelling in Hainaut, at least not in the turn of the next century of her life. I wouldn't expect there to be one earlier, but if more research is desired (and gets results) I suggest that the origins of the name can be added as part of the biography later on.