I have a few questions. Why is it that detection of a virtual particle would mean that it cannot be virtual? What is it exactly about the detection that destroys the virtual nature of a particle? I have read in several places that the idea of virtual particles is just an artefact of perturbation theory - if we could solve things exactly we wouldnt need to speak of virtual particles. So does this mean virtual particles dont actually exist, but rather are a mathematical tool that is necessary in perturbation theory? Are there really particles flying about that violate the relativistic energy equation? Thirdly, virtual particles can have an indefinite energy for a short period of time due to the energy time uncertainty relation. However in SR time is observer specific - does this then mean that in some frames of reference a particle is virtual, but in others it is not? Does this make sense?