Again CP, point us all to the exact language of the "margin" as there is many ways to include margin in a trial, not just on MOS. Moreover, why do you assume they only used those three trials for their design assumptions? FFTT JBAIN