Recent surgical reports indicate that for patients with secondary bacterial peritonitis, surgeons do not routinely use the identification of the bacterial pathogens and determination of their antimicrobial susceptibilities in choosing antimicrobial therapy. Further, some surgeons advocate abandoning the routine practice of obtaining culture specimens from patients with complicated appendicitis, because the data from the clinical laboratory have not been found to have an impact on postoperative care. We review the rationale for continued surveillance of and implementation of bacteriological data in treatment of secondary peritonitis. We also describe in detail the anaerobic flora of secondary peritonitis, the unique susceptibility patterns of these organisms, and the specific virulence factors of anaerobes, particularly Bacteroides fragilis. The fact that clinical investigations sometimes result in treatment failure when gram-negative anaerobes are resistant to the antimicrobials used or when complete antimicrobial susceptibility data are not available emphasizes the need for accurate and early knowledge of the bacteriologic characteristic of the flora of the operative site. We emphasize the relationship of in vitro susceptibility of intraoperative isolates with clinical outcome. We propose a cooperative trial that would demonstrate that successful antimicrobial therapy should be based on the susceptibility of the flora of the operative site, which correlates with clinical outcome.