The results of standardized 8 h lasting exposures of n = 18 volunteers to ethylbenzene (EthBz) at levels of 25 and 100% of the maximum allowable concentrations at the workplace (MAK) value of 100 ppm as well as the results of field studies are considered to evaluate a biological tolerance (BAT) value for EthBz. On the basis of the relationship between the external and internal exposure a BAT value of 1.5 mg/l has been set for the EthBz concentration in blood as the most sensitive and specific parameter of exposure to this aromatic hydrocarbon. The interpretation of EthBz blood values has to take into account the short half-life of t1/2 = 0.5 +/- 0.08 h in the first hour after the end of exposure in which this aromatic hydrocarbon is eliminated from the blood. The additional determination of the EthBz metabolites mandelic acid (MA) and phenylglyoxylic acid (PGA), respectively, excreted in post shift urine as well as in urine samples at the beginning of the next shift shows good correlations with the external exposure. The biological half-life of MA was calculated to t1/2 = 5.3 +/- 1.1 h. Because the time of sampling can vary the relationship between the levels of MA to PGA the total concentration of the excreted metabolites depends less on this influence and is therefore better suited for monitoring exposed persons. On the basis of the standardized experiments a BAT value has been proposed of 2 g MA plus PGA corrected per gram creatinine. Both BAT values are adjusted to data which result from earlier standardized exposures during 30 min to EthBz under physical activity of 50 watt on a bicycle ergometer.

The increased risk of lung cancer among foundry workers is assumed to be associated with the inhalation of gaseous and particle bound polycyclic aromatic hydrocarbons (PAH). These compounds are produced during pyrolysis of carbon containing loading material in the moulding sand.

The results are in accordance with relevant field and laboratory studies. Taken into consideration with the 95th percentile, the evaluated BAT values should be set at levels of 300 microg MCB/l blood, 175 mg 4-ClCat/g creatinine or alternatively at 30 mg p-ClPh/g creatinine in urine after the end of a shift. At the beginning of the next shift, the BAT values of the metabolites should be 35 and 15 mg/g creatinine, respectively.

The places of work where the air contains polycyclic aromatic hydrocarbons (PAH) include the production and processing of high boiling residues of the dry distillation (pyrolysis) of fossil fuels. Those worthy of mention include coke works, gas works, iron foundries, and the manufacture of aluminium."' Here it is mainly preparations based on coal tar or pitch that are used.The risk of cancer to people inhaling such substances has been established56 whereas in animal experiments tars and pitches are among the classic carcinogenic materials.7-9Tar bitumen is used as a binder for bituminous The aim of our investigation was to analyse the

Air sampled from the breathing zone of chimney sweeps during "dirty work" and soot samples were analysed for polycyclic aromatic hydrocarbons (PAH). A total of 20 PAH were quantified by gas chromatography-mass spectrometry in 115 air samples and 18 soot samples. These included benzo(b)fluoranthene, benzo(a)pyrene (BaP), chrysene, dibenz(a,h)anthracene, and indeno (1,2,3-cd)pyrene, all of which are animal carcinogens. The summed atmospheric concentration of these compounds depended on the type of fuel used and averaged 2 27 yg/m3 for oil fuel. If a mixture of oil and solid fuel was used the concentration was 5 06 ug/m3; pure solid fuel heating yielded 5 08Mg/m3. The air concentrations of BaP were 0-36, 0-83, and 0-82 pg/M3 respectively. The soot samples recovered after using the three different fuel types were 10 50, 109-10, and 51-25 mg BaP/kg. The maximum total concentrations of the five carcinogenic PAH were 243-70, 691 06, and 213-94 mg/kg respectively. The time weighted, shift mean concentrations of 0-02 to 0-21 pg/M3 benzo(a)pyrene obtained on 11 days form the basis for the industrial medical estimation of risk.