I don't actually know the details behind eaolson's particular bug. Just to be clear, all the way back to 8.2, when you load a class, even dynamically, the member VIs do load into memory. I believe that something about this one particular case was causing the problem. I don't know the details, so I can't be more helpful, but I didn't want people who had 8.2 to avoid a dynamic class loading architecture under the theory that it just didn't work ever.