I don't know what your real design considerations are here but, from I understand from your post, this is a really bad way to go about the process of logging data -- IF you want to be able to do significant ad hoc or stored procedures analyses after it has been collected. Using separate MDB files for data that ONLY differs by one field (namely that date) is not the most efficient way to organize it. What would be much more efficient would a joined table including the date and a reference ID of some sort for the various measurements that were done. That way your stored procedures for looking at ALL measurements of type X would be very simple going across ALL dates. Making such a comparison across multiple MDB files is a much more challenging process AND doing the original data collection in that way doesn't really gain you anything.
Generally, if something is difficult to do in the DCT (Database Connectivity Toolkit) it's because it's a "not good thing" to do within MDBs. I know that others probably disagree with that but I've worked with Access since it's initial release and other RDBMs prior to that both through compiled tools, Unix scripts, etc. You may, of course, still choose to proceed in the way you've described and that may work excellently for you.
