Integer division by zero causes a runtime error
floating point division by zero returns an error code meaning "not a number"(NAN)
soo
a#=1.0/0.0 //runs completely fine
b=1/0 //causes a runtime error
[edit]
What's really bad is that most floating point operations involving a #NAN return #NAN, so you can get cascading floats that aren't numbers.
Also, I have a thread in [code snippets] with functions to determine if a float is a number or not.

Tell me if there's a broken link to images in a thread I post, and I'll fix 'em.