To the Editor.— We enjoyed reading the article by Komaroff et al in the JulyArchives(138:1069-1073, 1978) and agree that the urinalysis, particularly the examination for bacteriuria, can be helpful in discriminating between a urinary tract infection and vaginitis. In their study, Kunin's criterion 1 was used, which considers the counting of one to ten bacteria in an unstained urine sediment in a high-power dry microscopic field as a positive result. With 10 5 /cc bacteria denoting an important level of bacteriuria, Kunin's method results in approximately 12% false-positives and about that many false-negatives. We favor checking for bacteriuria in an unspun urine specimen stained with methylene blue. This technique eliminates variabilities of urine centrifugation and affords better visualization of bacilli; the dye aids in distinguishing cocciform bacteria from urinary debris. 1 Approximately 1/30,000 cc is contained in one highpower field, so that seeing one bacterium per field correlates