Inside my custom Problem implementation, I had an assert that was failing, but no output was printed to the screen. The evolutionary run just crashed without explanation. Debugging was tricky, because all I knew was that it crashed with "-ea" and didn't crash without it. Does ECJ somehow suppress assert messages even when asserts are enabled? Or is it more likely that I somehow managed to do so myself? Siggy -- Ph.D student in Computer Science George Mason University http://mason.gmu.edu/~escott8/