Allan,

I have to wholeheartedly endorse what Vlad has said to you.

It's easier to dismiss this as a "bug" and move on to other code, but lets face it, we all would be going crazy if this were truly a bug in VFP. Lots of VFP drives lots of us crazy, but not THAT one!

It most likely *will* get you sometime later, so best to kill it while its still "fresh".

Good luck,
Jim N

>>It doesn't make sense that the first thisform.refresh moves the record pointer to EOF() and the second thisform.refresh does not!
>
>Apparently not. But you can be sure that there is a very good reason for it, ie: there's a place in your application where the record pointer is moved and this happens when Refresh is executed in certain conditions.
>
>I understand your workaround and this will probably work correctly and it solves your problem now. But, from my experience, a situation like yours (when the program behaves in a weird way and you don't know why) hides a bug. My advice is to find the real cause now, even if you have a perfect good workaround. Usually, if you don't, this bug will take its revenge latter and you might have an apparently different weird behavior/bug. This is very bad because a simple bug may "generate" many workarounds in one application. Like all workarounds, this means supplementary code that must be maintained and a lot of comments to explain why you needed that code.
>
>Also, from my experience (which may not be very representative:)), after some time, you will find the real bug. Of course, you will fix it. But the fix will almost always generate bugs in the workaround code. And these are one of most nasty bugs because it's very difficult to fix a bug in an workaround. Even if the workaround is commented, the comment will not address the right bug (because you didn't know it at that moment) and most probably you'll just fix the workaround code, etc, etc. It's like an avalanche.
>
>Also, bugs related to weird rec pointer movements are in my top of "the nastiest /serious bugs on long term". Just next to the "selected area" bugs. :)
>
>No need to say how difficult it is to fix these things when you're not the original programmer... :(
>
>Sorry for being so long... :) But, IMHO, when you have a bug like this, it doesn't matter how much time it takes you to fix it. It cannot be longer than the sum of the times needed to maintain the workarounds. (Just in case you need to justify this time to your boss :))
>
>Vlad