INTRODUCTION: Integration of positron emission tomography (PET) and magnetic resonance (MR) has become a topic of increasing interest to the imaging community over the past two years. OBJECTIVES: In this text, the authors attempt to distinguish facts from fiction concerning such integrated systems. Analysis of existing information of combined imaging on existing brain PET/MR systems and imaging experience with PET-computed tomography (CT) is reviewed. Various types of system integration of PET and MR are discussed with completely independent systems on one hand and completely integrated systems with the possibility of simultaneous data acquisition on the other hand. Furthermore, it is discussed, what simultaneous data acquisition with nuclear imaging systems combined with MR or CT really means, as technical simultaneity may not be relevant in light of the pharmacokinetics of the nuclear tracers used. DISCUSSION: The authors conclude that combining PET/MR is an interesting research endeavor with uncertain outcome. They argue that, while completely simultaneous brain applications are of research interest immediately, clinical applications do not currently warrant the construction of fully integrated systems. Systems adjacent to each other, where imaging tables are linked with a patient "shuttle" thereby requiring only patient translation but no repositioning, may be a good start to assess the value of integrated PET/MR.