Asphalt fumes have been reported to cause irritation of the mucous membranes of the eyes, nose, and respiratory tract. While other symptoms such as coughing and headaches were reported recently, there was no statistical association with asphalt fume exposure. Results from experimental studies with anirnals indicate that roofing asphalt fume condensates generated in the laboratory and applied dermally cause benign and malignant skin tumors in several strains of mice. Differences in chemical composition and physical characteristics have been noted between roofing asphalt fumes collected in the field and those generated in the laboratory. However, the significance of these differences in ascribing health effects to humans is unknown. Furthermore, no published data exist that examine the carcinogenic potential of field-generated roofing asphalt fumes in animals. Since the health risks from asphalt exposure are not yet fully defined, NIOSH, labor, and industry are working together to better characterize these risks while continuing their effort to reduce worker exposures to asphalt fumes. In the roofing industry, exposure to asphalt fumes and other related exposures is well documented and studies still continue. Several studies have identified increased polycyclic aromatic compounds (PACs) exposure to the kettle operators versus other categories ofroofers. Due to the nature of the kettle operator's job, this 'appears to be an obvious conclusion; however, few controls have been utilized to minimize these exposures. A survey was conducted on January 23-25,2001, at the Blue Valley West High School in Stilwell, Kansas, where a new 3-ply roof with a mineral surface fiber glass cap sheet was being applied. This was a new building under construction. The engineering control used during this evaluation was low fuming asphalt; other existing engineering controls for this industry will be evaluated in subsequent surveys. The kettle operator's exposures to TP, BSF, and total PAC were reduced by 38%, 56%, and 45%, respectively, when comparing exposures when low fuming asphalt was used to when conventional asphalt was used. When exposure concentrations were adjusted for NTP the percent reductions for the kettle operator were 36%,55%, and 43%, changing the reduction in exposure slightly. When the kettle operator's exposure data was normalized exposures to TP, BSF, and total PAC increased by 85%, 28%, and 60% respectively when using low fuming asphalt. When the normalized data was adjusted to NTP the increase in exposures was slightly greater. These increases in exposure seen when the data was normalized to the amount of asphalt use each day suggests that to some degree the exposure reductions seen when low fuming asphalt was used may be effected by the amount of asphalt used. Another possible explanation for this increase in exposure when the data was normalized is that on the third day of sampling when the low fuming asphalt was used, three hours passed before a complete crust of polymer was formed over the surface of the asphalt thereby allowing more asphalt fume to escape in a similar manner as conventional asphalt. None of these reductions were statistically significant at 95% confidence. Similar to the results seen for the kettle operator, the mean exposure concentration for TP, BSF, and total PAC were reduced by 46%, 57%, and 49%, respectively, in the area air samples collected around the asphalt roofing kettle when comparing exposure results when low fuming asphalt was used to when conventional asphalt was used. When the exposure data was adjusted to NTP, the reductions were slightly less than the unadjusted results. Like the kettle operator's results, when the area air sample results were normalized there was increase in exposure when low fuming was used and the increase was slightly more when the normalized area air exposure data was adjusted to NTP. None of the reduction were statistically significant at 95% confidence. The mean exposure results for TP and BSF were reduced by 14% and 13%, respectively, in the roof level workers' exposure results when comparing low fuming asphalt exposure data to conventional asphalt exposure data. There was an increase of 21 % in total PAC exposure results for the roof level workers when using low fuming asphalt. NTP data adjustments lowered the TP and BSF reductions in exposure to 11 % and 10%, respectively, while the increase in total PAC exposure seen when low fuming asphalt was used was slightly greater at 26%. Normalizing the roof level workers' exposure data increased mean TP, BSF, and total PAC exposure concentrations by 208%, 206%, and 323%, respectively, when using low fuming asphalt. When the normalized data was adjusted to NTP the increases seen in the roof level workers' exposures were slightly greater when low fuming asphalt was used. None of the reduction were statistically significant at 95% confidence. The results of measurements taken when both conventional and low fuming asphalt were used at the same asphalt roofing site indicate that using low fuming asphalt reduces the exposure of the kettle operator to asphalt compounds. The area samples and personal samples taken on the kettle operator showed reductions, but none of the reductions measured were statistically significant at 95% confidence. Two reasons that the reductions were not statistically significant are that the sample size was too small, and on the day that the TruMeltTM was used, the polymer crust was not fully formed on the asphalt surface until three hours after air sampling had begun. When the results were nonnalized by dividing by the amount of asphalt used exposures for the kettle operator, area air samples collected around the kettle, and roof level workers all slightly increased indicating that the reduction measured may have been influenced by the amount of asphalt used that day. From the results of this survey, one conclusion that may be drawn is that in order to collect quality data, the researcher must have some control over the amount of asphalt used each day and the transition between conventional and low fuming asphalt. Also, while it appears that using low fuming asphalt reduces operator exposure to the components of asphalt fume, due to the many variables encountered during sampling, the results are not statistically significant.