In order to assess the validity of Apo E phenotyping from stored specimens, phenotypes determined from fresh serum samples were compared with those from stored (8 years at -20 degrees C) samples from the same individuals (n = 42). The effect of early postmortem period on Apo E phenotype determinability was studied by taking four duplicate blood samples from eight cadavers 2, 6, 12, and 24 h after death. Apo E phenotyping was performed directly from serum by isoelectric focusing and immunoblotting. From the cadavers, the same Apo E phenotypes were obtained 2, 6, 12, and 24 h after death. After eight years' storage five out of ten Apo E4/4 phenotypes were falsely recorded as Apo E4/3 and one out of six Apo E4/3, one out of 12 Apo E4/2 were falsely interpreted as Apo E3/3. Phenotypes Apo E2/2 (n = 2), Apo E3/2 (n = 10), and Apo E3/3 (n = 2) were correctly assessed after 8 years of storage. In the total material, 17% (7/42) of Apo E phenotypes were incorrectly assessed after the storage.