The reported prevalence of localized juvenile periodontitis (LJP) amongst teenagers and young adults varies greatly. The etiology of LJP has been related to Actinobacillus actinomycetemcomitans (Aa), and it has also been suggested that there may be a transmission of Aa within families resulting in the familial distribution of the disease. This study describes the high prevalence of LJP in adolescents, 12-20 years of age, from a group of nuclear families living and functioning in a closed, closely knit community. The survey was carried out on a population of teenagers that had attended the same school and their siblings. All students attending that school and their siblings were examined. They were given a periodontal examination and a questionnaire relating to their demographic details and their personal oral hygiene habits. The periodontal examination was limited to the incisors and first molar teeth. Plaque index (PlI), gingival index (GI), the presence or absence of bleeding on probing (BOP), probing pocket depth (PPD) and recession were measured. All patients having at least two of the examined sites with probing pocket depth > or =5 mm or one site > or =6 mm were considered as possible sufferers from LJP and had a full mouth periapical radiographic survey carried out using a paralleling technique to confirm the diagnosis. At the sites with probing pocket depth > or =5 mm, a Shei ruler was used to measure the % of the root coronal to the alveolar bone. A cut off point of > or =20% was used as a measure of true bone loss confirming the clinical diagnosis of LJP. 86 individuals from 30 families comprised the population of interest. There were 44 males and 42 females with a mean age of 14.7+/-2.3. Of the 86 individuals examined, 33 individuals from 15 families were diagnosed as having LJP (38.4%). None of the individuals examined showed any evidence of the generalized form of juvenile periodontitis. The mean age of the LJP patients was 15+/-2.3 yrs. with a 1:1.75 male to female ratio. Except for 2 pairs of families with genetic ties, no familial connections could be traced between the different nuclear families affected by LJP despite repeated and intensive questioning. There were no significant differences in the PlI and the GI between the groups while the LJP group had significantly higher BOP, PPD and PAL than the non-LJP group. These finding strongly suggest an environmental influence in the etiology of the disease.