The aim of this study was to investigate to which extent 6th grade school children are exposed to tobacco smoke by others. As biomarker for the exposure to tobacco smoke nicotine and cotinine were measured in the urine. Our study population consisted of 771 schoolchildren aged 11-14 years who according to a questionnaire did not smoke. In addition we analysed the data of 459 school children who were not exposed to tobacco smoke at home. The nicotine and cotinine concentrations in the spontaneous urine sample were determined by HPLC methods.On average in about 20% of all non-smoking children, who were not exposed to tobacco smoke at home, biomarker (nicotine or cotinine) were detected in the urine. The percentage of the detected biomarker values (nicotine and/or cotinine) in the urine of the school children varied between 0% and 50% between schools. In addition we determined the proportion of smoking classmates per school. No positive association was found between the detected biomarker values of the non-smoking school children not exposed to tobacco smoke at home and the proportion of smokers per school. The concentration of biomarker depending on the time of day the urine samples were collected showed higher nicotine and cotinine values when the urine sample was collected between 10 and 12 o'clock in the morning compared to urine samples collected between 7 and 10 a.m.In spite of the limitations our study provides some evidence that children are exposed involuntarily to tobacco smoke by others at school. That is why our results support the requirement of a general legal ban on smoking for teachers, the school staff and students.