The aim of this study was to assess by means of an ELISA technique, the occurrence of 3 putative periodontopathogens, Porphyromonas gingivalis, Bacteroides forsythus, and Treponema denticola, in 3 clinically-defined adult periodontal conditions. Thirty systemically-healthy subjects were selected and grouped into 3 categories according to their periodontal health: 1) 10 periodontitis subjects (PS), having moderate adult chronic periodontitis; 2) 10 untreated gingivitis subjects (UGS), exhibiting no signs of periodontal destruction but presenting with clinical signs of mild gingivitis; and, 3) 10 treated gingivitis subjects (TGS), having the same clinical status as UGS, but who received a thorough prophylaxis treatment within the past 7 to 14 days prior to the baseline examination. A total of 60 samples were collected subgingivally from the six Ramfjord teeth per subject in each group and ELISA analysis was carried out to give a semiquantitative estimate of P. gingivalis. B. forsythus, and T. denticola. The immunologic detection method suggested the presence of antigens of P. gingivalis, B. forsythus, and T. denticola in subjects from each of the 3 groups. When a global analysis for the 3 disease groups was performed at one time, statistically significant differences were found among the ELISA scores of the 3 bacterial species. For example, comparisons of the ELISA scores showed that the concentrations of P. gingivalis differed significantly when comparing TGS to UGS and PS, but not when examining UGS/PS. The ELISA scores for B. forsythus were significantly different between TGS and PS. Mean concentrations of T. denticola were significantly different when comparing PS to TGS or UGS, whereas no difference was found between the latter categories. Within the limited scope of this study, the concentration of antigens detectable from putative periodontopathogens like P. gingivalis, B. forsythus, and T. denticola differed among the 3 diseased groups, with periodontitis subjects often showing the greatest level of antigens. Thus, it is reasonable to expect that, when using sensitive immunological detection methods, antigens of suspected periodontal pathogens can be found irrespective of the individual's clinical status. However, while detectable in the periodontal sites, the concentrations of these microorganisms are most likely to be above the threshold necessary to induce clinically-significant disease. Studies with larger sample size and standardized antigens are necessary to determine if the groups we found not to differ, were, in fact, different.