> the simple && solutions works great. > > I will try a benchmark against mapinfo intersect, after running the more > sofisticated solution with the gist index. The distance function isnt very accurate when things either 'just' touch or are really-really close to touching (you might get zero or you might get a very small number). dave