Two hundred and eighteen urine specimens were cultured in duplicate to compare the reliability of a miniature culture technique with a more standard, calibrated loop-method for quantitative urinary bacteriology. The new method appears to be reliable: the rate of false negative results was 2.5 percent and of false positive 3.6 percent. The correlation with the standard method was best in the ranges of 0 to 10,000 and greater than 100,000 colonies per milliliter. The technique is simple, quick, and easily applicable to the physician's office laboratory.