FROM RECORDS OF COURT APPEARANCES, DELINQUENCY RATES WERE DETERMINED FOR BOYS' SECONDARY MODERN SCHOOLS IN A NORTHERN CITY IN ENGLAND. USING CENSUS DATA, THEIR NEIGHBORHOOD AREAS WERE COMPARED, AND TWO PAIRS OF SCHOOLS, EACH HAVING SIMILAR NEIGHBORHOODS BUT DIFFERENT DELINQUENCY RATES, WERE SELECTED. A ROLE DESCRIPTION INSTRUMENT WAS ADMINISTERED TO ALL FOURTH-YEAR PUPILS MEASURING THEIR PERCEPTIONS OF THE 'TASK ORIENTATION' AND 'EMOTIONAL TONE' OF THEIR SCHOOLMATES AND THE 'CONCERN' AND 'SOCIAL CONTROL' OF THEIR TEACHERS. ALTHOUGH THE SCALE OF THE STUDY WAS SMALL. IT PRODUCED CONSISTENT RELATIONSHIPS BETWEEN THE DELINQUENCY RATE OF A SCHOOL AND THE PERCEIVED BEHAVIOR OF ITS TEACHERS AND PUPILS. IN THE HIGH DELINQUENCY SCHOOLS, THE PUPILS WERE SEEN TO BE SIGNIFICANTLY LESS ORIENTED TOWARDS THE WORK TASK GIVEN THEM BY THE SCHOOL THAN WERE THE PUPILS IN THE LOW DELINQUENCY SCHOOLS. IT IS CONCLUDED THAT THESE TRENDS SUGGEST THAT THE HIGH DELINQUENCY SCHOOLS FALL FAR SHORT OF FULFILLING ANY POSITIVE, SUPPORTIVE ROLE IN THE LIVES OF THEIR DELINQUENT PUPILS. FROM THE PUPILS' PERCEPTIONS OF TEACHER BEHAVIOR, IT IS CONCLUDED THAT THE WAY IN WHICH TEACHERS ARE SEEN TO EXERCISE THEIR AUTHORITY DIFFERS IN HIGH AND LOW DELINQUENCY SCHOOLS. IN THE LATTER THEY ARE SEEN TO BE LESS HOSTILE TOWARDS PUPILS THAN ARE THEIR COLLEAGUES IN HIGH DELINQUENCY SCHOOLS. REFERENCES AND TABLES SHOWING MEASUREMENT RESULTS ARE INCLUDED. (RCB)