The Capgras syndrome, first described by Capgras and Rebould-Lachaux (1923), is the name applied to a peculiar delusional system in which the patients come to believe that persons well known to them—usually close relatives—are impostors who have assumed the exact appearances of those whom they have supplanted. The majority of examples of this condition have been observed in schizophrenic patients but, like most eponymous titles in medicine, the term was applied to a disease pattern that failed to fit conveniently into existing classification. Most branches of medicine, with the gradual accumulation of new information, pass through an ‘identification’ phase when new syndromes are being discovered. A further 'synthesizing’ phase is eventually reached when researchers, equipped with greater knowledge regarding likely aetiology, pattern of symptoms and treatment, are able to appreciate the wider implications and ramifications of the problem, as well as uncertainties surrounding the initially identified disease pattern. In psychiatry, titles that emphasize specific patterns of symptomatology tend to be the rule rather than the exception, but advances in psychopathology, neuropsychology and neuropathology imply a need for reassessing the contribution of a purely phenomenological approach to the problem. Such an approach may well be overdue as far as the Capgras and other syndromes are concerned. The following case is presented in the hope that the condition, when it is recognized, will be evaluated in terms of its aetiology, and that attention will be paid to the more fundamental questions of psychopathology and related patterns of neuropsychological function.