Fermat's Enigma

by Simon Singh

Could it really be the case that this minor modification turns Pythagoras’s equation, one with an infinite number of solutions, into an equation with no solutions?