>Looking for general ideas in where to place the
>coding for cascading deletes (and other pieces
>of code like this).

IMHO, these kind of things should always be defined in the data tier (back end). Here's why:

Let's assume you switch to an entirely different back and. Something that works really different from VFP tables. One example would be XML files. In XML files, we may not have child records, but we would have a hierarchy. If we were to delete the parent node, we also deleted the childs. For this reason, any code in any other tier but the data tier would plain fail.

Conclusion: Cascading deletes belong to the data tier.

Makes sense?

Markus