Summary: Parietal and occipital seizures have been investigated relatively little. Recent interest in seizures of frontal lobe origin has led to a definition of several well-identifiable clinical patterns of frontal lobe seizures. A similar delineation of the clinical and EEG pattern of parietal and occipital epilepsy, using modern investigations, appears appropriate, not least because the few reported series of surgical treatment in parietal and occipital epilepsy have indicated that operation may be relevant in selected cases, not only for removal of space-occupying lesions. Such a study is currently being undertaken at the National Hospital for Neurology and Neurosurgery and The Chalfont Centre for Epilepsy with the aim of developing and evaluating diagnostic criteria for these seizures. From the existing literature we can conclude that precise incidence and prevalence are largely unknown. A recent community-based study of 252 subjects with partial epileptic seizures in an epileptic population of 594, showed that parietal seizures and seizures of posterior origin each comprised 6.3% and central or sensory-motor seizures comprised 32.5% of focal seizures in the 160 cases in which seizures could be subclassified (Manford et al., 1992). This incidence seems low for occipital seizures as compared with the 1953 study by Gibbs and Gibbs, who observed occipital epileptiform activity in 8% of subjects with focal epilepsy. The most prominent clinical manifestations of parietal epilepsy are elementary sensory phenomena at the beginning of seizures and elementary visual hallucinations in occipital epilepsy. These symptoms are not associated solely with posterior hemisphere epilepsy, however, and more studies are obviously needed to define how close this relation is. Scalp EEG is frequently negative or maybe misleading; further-more, spread of epileptic discharges from the parietal and occipital lobes to frontal and temporal regions may obscure seizure origin. Because of these controversial symptoms, diagnostic criteria may be difficult to define. The wide difference in clinical and EEG manifestations between reported series of parietal and occipital epilepsy also reflects a considerable problem with patient sampling. Classification of epilepsy according to the anatomic division of the brain may be arbitrary, and it may be appropriate to define epileptic syndromes such as sensorimotor seizures or occipitotemporal seizures that cross such artificial divides.