The behavioural manipulation hypothesis posits that parasites can change the behaviour of hosts to increase the reproductive fitness of the parasite. The protozoan parasite Toxoplasma gondii fits this description well. Sexual reproduction occurs in the cat intestine, from which highly stable oocysts are excreted in faeces. Grazing animals, including rodents, can then ingest these oocysts. The parasite has evolved the capacity to abolish the innate fear that rodents have of the odours of cats, and to convert that fear into an attraction. This presumably increases the likelihood of the rodent being predated, thereby completing the parasite's life cycle. The behavioural syndrome produced by T. gondii does not have any precedent in neuroscience research. This is not a case where the normal functioning of fear system have been altered. This is not even the case of the altering of fear towards predator odours, while leaving other kinds of fear intact. This is an unprecedented example of one component of the fear being eliminated (and replaced by a novel attraction), while appearing to leave other domains unchanged. An understanding of the neurobiological effects of T. gondii is beginning to emerge. One possibility is T. gondii's preferential localisation to, and effects within the amygdala; this is particularly intriguing, given the role of this brain structure in the normal fear response. Obviously, far more must be understood, and the unique behavioural effects of T. gondii put very demanding constraints on any hypothesis we formulate to explain proximate neurobiological mechanisms.