> Don't you want to just test that the number of "("s equals the number of > ")"s or am I missing the point? I had this idea too, but there is additional requirement that any beginning must have greater or equal number of '(' than ')'. -- Adrian