For the purpose of clinical management, any woman with an acute rise in blood pressure in the latter half of pregnancy must be regarded as having preeclampsia with the possibility of progression to eclampsia. Unfortunately, such diagnoses have been accepted uncritically in the selection of cases for clinical and laboratory studies of preeclampsia, with inevitably erroneous and contradictory conclusions about the disorder. The diagnosis of mild preeclampsia may be correct in roughly one-half of cases, but others may be latent or frank essential hypertension or any of a variety of renal diseases. In selecting cases for research, the diagnostic errors can be greatly reduced by the exclusion of all multiparas and all primigravidas without abundant proteinuria. The primigravidas should have a reliable history of normality or follow-up studies proving it, be aged 25 or less, and have hyperuricemia. The selection of cases for the study of preeclampsia demands far more rigid criteria for diagnosis than does the diagnosis for clinical management.