Hi I have two tables, say A and B, that have a many-to-many relationship, implemented in the usual way with a join table A_B. How can I economically find all the rows in table A whose id's are not in A_B at all (i.e. they have zero instances of B associated)? Thanks Daniel