Well, the first book I scanned had what K1000 called 'long dashes' in it. When I translated it into rtf, and looked at it with JAWS and Word, those became em dashes; I presume that 'long dash' is just K1000's name for em dash, although I cannot be sure. Anyway, I transfered the file over to my Pac Mate to fix it up in Braille, because I tend to pick up errors better that way. I thought something was strange, because the em dashes were displayed as dashes on the Braille display, whereas, when I scanned books in OpenBook, the em dashes were not displayed properly, containing a letter t, of all things. Well, I went ahead and made some corrections, and when I transfered the file back to my PC, all the em dashes were replaced by dashes. I just checked it by checking the original rtf file with the em dashes in it to the Pac Mate, making a small change and transfering it back. Sure enough, the em dashes are now dashes. I am going to have to go through with a search and replace and fix them one by one. I cannot just do a global search and replace for obvious reasons. The major irony here is that just a few months ago, I was arguing with Pratik about whether or not to keep em dashes, I being on the dump them side because they weren't coming out correctly in the brf versions, and now I am complaining about losing them. Feel free to gloat, Pratik, if you like; but not too much. That would be unprofessional. But if you find the temptation simply impossible to resist, then I suppose that you are entitled to a little gloating. <grin> I cannot see that this is not a problem with K1000, as that is the only thing that has changed. Something in the translation from kes to rtf, perhaps, but the character that Word calls em dash has the same number as the one I get when using JAWS insert 4. Something weird. I didn't lose the em dashes when using OpenBook, but I have no idea what could be causing it, other than it having to be K1000. Well, I gotta go; I have a lot of work ahead of me today. If I had known I was going to lose the em dashes, I could have put in some character(s) and done a global search and replace to fix them in a flash. Since I have no reason to think that this is some kind of fluke, I will have to remember to do that in the future when scanning with K1000.