Forty-two children out of 20 sibships with autosomal recessive polycystic kidney disease were observed pro- and retrospectively over a mean period of 3.7 years in a long-term study on cystic kidney diseases in children. The intra- and interfamilial variability in terms of age at diagnosis, administration of antihypertensive therapy, liver affection, and renal function were evaluated. According to the 1971 subclassification of Blyth & Ockenden, defining different grades of severity, 12 patients were assigned to the perinatal, nine to the neonatal, 13 to the infantile, and eight to the juvenile subtype of autosomal recessive polycystic kidney disease. In 11 of the 20 families different subtypes were observed among affected siblings. In seven families, affected sibs belonged to adjacent subtypes, while major intrafamilial differences were observed in only four families. The defined subtypes, therefore, cannot be regarded as appropriate in distinguishing genetic groups of autosomal recessive polycystic kidney disease. With respect to the severity of autosomal recessive polycystic kidney disease, there is a wide spectrum of phenotypic manifestations, ranging from stillbirths to mildly affected of phenotypic manifestations, ranging from stillbirths to mildly affected adults, while intrafamilial variability of the clinical picture is generally small with multiple allelism as the most likely genetic explanation. Age at death, however, showed gross variation in eight sibships. Differences in the clinical course between several siblings cannot be explained by a sex influence in autosomal recessive polycystic kidney disease.