The recent paper by Cohen in the August issue of Pediatisics relies on the use of "one or more cultures of over 100,000 organisms per millimeter of urine" to make the diagnosis of urinary tract infections upon which the remainder of the study is dependent. Obviously, the colony count was intended to read "per milliliter," and yet the results of the study must be seriously questioned since the criterion for diagnosis of urinary tract infection was not rigid enough to exclude a large number of false-positives— patients who never really had urinary tract infections.