The mere presence (or absence) of any given microorganism is not adequate for ruling it in or out as a cause of clinical urologic infection. A distinction between true bacteriuria and contamination of the urine during the procedure of collection must always he made in the evaluation of a positive urine culture. The bacterial count in a given urine sample has offered a means of defining this distinction. The limited data in children, supported by more extensive data in adults, suggest that urines containing less than 1,000 colonies/ml are indicative of contamination; urines containing between 1,000 and 100,000 colonies/ml are to be suspected of infection and studies repeated, and urines containing more than 100,000 colonies/ml are indicative of infection. The proper collection and handling of specimens after collection is mandatory, and should not be entrusted to untrained personnel. Conditions under which low bacterial counts may be encountered in the presence of active infection are discussed. Special consideration must be given to those patients with chronic or recurrent urinary tract infection who frequently, in the presence or absence of symptoms, may show bacterial counts in the urine below the range considered to be significant. Reliable clinical means for resolving this problem remain to be explored. Suprapubic aspiration of bladder urine for culture and bacterial counts may prove helpful in given cases. Pyuria, said to be the hallmark of pyelonephritis, is not of itself satisfactory for making a diagnosis of urinary tract infection. In the first place, there is a lack of general agreement as to what constitutes pyuria; secondly, pyuria may be seen with many extra-renal infections; and finally, pyuria may be absent with infection of the urinary tract. A Gram stain of an uncentrifuged specimen of urine will usually reveal bacteria if infection is present. This has been subjected to quantitative evaluation, with almost all urine specimens with bacterial counts of 100,000 or more/ml showing organisms when stained by Gram's method. Thus, the Gram stain of a smear of urine provides a useful screening technique for detecting patients with urinary tract infection. While it has been taught, generally, that only the study of catheter specimens in girls is reliable for the diagnosis of urinary tract infection, recent studies refute this. While there is always a risk of introducing infection during the procedure of catheterization, this is minimal if proper precautions are exercised, and such fear should not interdict catheterization under certain circumstances which have been mentioned. While bacteriuria is not to be considered synonymous with pyelonephritis, several histologic studies (at necropsy, at nephrectomy, and after renal biopsy) show that bacteriuria is frequently associated with involvement of the renal parenchyma. While reliable clinical means of determining which of the patients with true bacteriuria have renal involvement are not presently available, prudence dictates that each case of bacteriuria be regarded as being associated with active renal disease, or very likely to become so. The staining characteristics of leukocytes excreted in pyelonephritis may be helpful in determining the presence of renal involvement; the leukocytes of patients with cystitis or urethritis apparently do not show the staining characteristics of those excreted in pyelonephritis. A wider use of this supravital staining technique is needed to elucidate further its usefulness. In addition, renal function studies and roentgenographic examinations may be helpful. Bacteriologic culture of renal tissue obtained by needle biopsy has not proved fruitful, in general. Histopathologic study of such tissue has proved more rewarding. The limitations of the renal biopsy technique by needle are in large measure due to the focal nature of pyelonephritis. While the open biopsy technique, permitting direct visualization of the kidney and selection of site(s) for biopsy, may provide more adequate tissue for examination than the blind needle technique, satisfactory comparative data are lacking.