>>Conclusion, which we may try to check: if an object has a .parent, it is created in parent's DS and operates in it. If it doesn't, then it's the DS in which it was created.
>
>Dragan,
>if a object has a .parent, it is not a VFP's object, but an .parent's member and it's datasession is that of the .parent

So I have concluded, yes. We have posted a different conclusion here at some time, that any object will operate in the DS in which it was created. We didn't have this "but in the .parent's DS if it was .addObject()ed or .newObject()ed" addition to it.