Primary steroid-resistant nephrotic syndrome (SRNS) is characterized by childhood onset of proteinuria and progression to end-stage renal disease. In 26% of cases it is caused by recessive mutations in NPHS2 (podocin). Congenital nephrotic syndrome (CNS) is caused by mutations in NPHS1 (nephrin) or NPHS2. In three families mutations in NPHS1 and NPHS2 had been reported to occur together, and these tri-allelic mutations were implicated in genotype/phenotype correlations. To further test the hypothesis of tri-allelism, we examined a group of 62 unrelated patients for NPHS1 mutations, who were previously shown to have NPHS2 mutations; 15 of 62 patients had CNS. In addition, 12 CNS patients without NPHS2 mutation were examined for NPHS1 mutations. Mutational analysis yielded three different groups. (1) In 48 patients with two recessive NPHS2 mutations (11 with CNS), no NPHS1 mutation was detected, except for 1 patient, who had one NPHS1 mutation only. This patient was indistinguishable clinically and did not have CNS. (2) In 14 patients with one NPHS2 mutation only (4 with CNS), we detected two additional recessive NPHS1 mutations in the 4 patients with CNS. They all carried the R229Q variant of NPHS2. The CNS phenotype may be sufficiently explained by the presence of two NPHS1 mutations. (3) In 12 patients without NPHS2 mutation (all with CNS), we detected two recessive NPHS1 mutations in 11 patients, explaining their CNS phenotype. We report ten novel mutations in the nephrin gene. Our data do not suggest any genotype/phenotype correlation in the 5 patients with mutations in both the NPHS1 and the NPHS2 genes.