Broiler carcasses were removed from commercial processing lines immediately after defeathering, before chilling, and after chilling to determine whether any relationship exists between aerobic bacteria and the human enteropathogens salmonellae and Campylobacter. In two experiments, a whole carcass rinse procedure was used to sample 30 carcasses after defeathering, 90 carcasses before chilling, and 90 carcasses after chilling, for a total of 210 different carcasses. Aerobic bacteria and Campylobacter spp. were enumerated and the incidence of salmonellae was determined. Salmonellae and Campylobacter incidences were 20 and 94%, respectively, for all carcasses sampled. After picking, neither salmonellae-positive nor Campylobacter-positive carcasses had mean aerobic most probable number (MPN) values that were different from carcasses negative for those organisms. Immediately before chilling, aerobic and Campylobacter counts were 7.12 and 5.33 log10 cfu per carcass, respectively. Immersion chilling reduced aerobic counts by approximately 1.8 log and Campylobacter by 1.5 log, with no change in salmonellae-positive carcasses. There was no difference in aerobic or Campylobacter counts between carcasses that were positive or negative for salmonellae at any of the sampling locations, nor was any correlation found between levels of aerobic organisms and Campylobacter. Carcasses with aerobic counts above the mean or more than one standard deviation above the mean also failed to show any correlation. Discriminant analysis indicated error rates as high as 50% when numbers of aerobic bacteria were used to predict incidence of salmonellae or Campylobacter on individual carcasses. Aerobic bacteria are not suitable as index organisms for salmonellae or Campylobacter on broiler carcasses.