The reliability of the bacteriologic identification of β-streptococci in private offices was compared with a university laboratory. Results were identical for 85.6% of the specimens processed. Only 4% of cultures reported positive by the reference laboratory were found to be negative by the private offices. A greater discrepancy, 10.5%, was found with specimens positive in the office but negative in the laboratory. The majority of discrepancies occurred in cultures which were only 1 to 2 + positive. These data suggest that (a) the correlation between five private offices and a reference laboratory was relatively good, (b) relatively few positive streptococcus cultures were missed, (c) where errors were found, it was more often in the direction of overreading for streptococcus, and (d) with quality control, bacteriologic identification can be a practical, useful, and reliable office procedure.