Blyth and Ockenden [1971] assigned patients with autosomal recessive polycystic kidney disease (ARPCKD) to 4 discrete groups (perinatal, neonatal, infantile, juvenile) on the basis of the age of presentation. They and others speculated that at least 4 genes were responsible for what they considered to be closely related, but different conditions. These views have gained wide but not universal acceptance. Some workers have insisted that the perinatal and neonatal "forms" of ARPCKD differ fundamentally from the juvenile "form." However, others have proposed that ARPCKD-CHF (congenital hepatic fibrosis) and CHF-ARPCKD are manifestations of the same disease with variation of expression in a kindred. We report on a patient who presented at birth (1979) with ARPCKD and respiratory distress. He died at 18 hr. An older sib presented at 16 yr in 1984. She had no symptoms, but her mother wanted reassurance that the daughter did not have a condition similar to that of the deceased sib. Blood pressure was 120/80 mm Hg and there was hepatosplenomegaly. A diagnosis of renal tubular ectasia and CHF was made by ultrasonography, radiologic studies, and a liver biopsy. The evidence from families such as this favors the concept that ARPCKD and CHF presenting as Blyth and Ockenden's perinatal form, and CHF and renal tubular ectasia as their juvenile form, are manifestations of the same genetic disorder, and that the different manifestations are more likely variations in expression than the results of different mutant genes. The manifestations in this family add weight to the growing body of evidence that intrafamilial variability may occur, not only in autosomal dominant conditions, but also in autosomal recessive disorders.