One of the holy grails of contemporary science has been to establish the possibility of preferentially breaking one of several bonds in a molecule. For instance, the two O-H bonds in water are equivalent: given sufficient energy, either one of them is equally likely to break. We report bond-selective molecular fragmentation upon application of intense, 2-cycle pulses of 800 nm laser light: we demonstrate up to three-fold enhancement for preferential bond breaking in isotopically substituted water (HOD). Our experimental observations are rationalized by means of ab initio computations of the potential energy surfaces of HOD, HOD(+), and HOD(2+) and explorations of the dissociation limits resulting from either O-H or O-D bond rupture. The observations we report present a formidable theoretical challenge that need to be taken up in order to gain insights into molecular dynamics, strong field physics, chemical physics, non-adiabatic processes, mass spectrometry, and time-dependent quantum chemistry.