Figured it out - I can use requests to select all files starting with 91xx and 92xx etc to discriminate between those which are higher up in the number series, then delete the opposite in the other file. Maybe not the most elogant way to do it but it seems to work.
If this is a number field or if it's text and all your codes have the same number of digits, you can use Value 1...Value 2 as a search criterion to find a range of values. You can also use inequality operators such as > and < .
How is it too large? Does it take too long to load, or filling the hard drive?
I believe Filemaker 6 has maximum file size limitation.
You might consider making a backup
(make a backup)
duplicate the file in the OS
Open the duplicate
Save As Smaller
This can sometimes speed up a database (depending on several things)
(A severely fragmented database can cause speed problems, depending on the OS)