Recently published studies not only demonstrated that laser printers are often significant sources of ultrafine particles, but they also shed light on particle formation mechanisms. While the role of fuser roller temperature as a factor affecting particle formation rate has been postulated, its impact has never been quantified. To address this gap in knowledge, this study measured emissions from 30 laser printers in chamber using a standardized printing sequence, as well as monitoring fuser roller temperature. Based on a simplified mass balance equation, the average emission rates of particle number, PM(2.5) and O(3) were calculated. The results showed that: almost all printers were found to be high particle number emitters (i.e. > 1.01 x 10(10) particles/min); colour printing generated more PM(2.5) than monochrome printing; and all printers generated significant amounts of O(3). Particle number emissions varied significantly during printing and followed the cycle of fuser roller temperature variation, which points to temperature being the strongest factor controlling emissions. For two subgroups of printers using the same technology (heating lamps), systematic positive correlations, in the form of a power law, were found between average particle number emission rate and average roller temperature. Other factors, such as fuser material and structure, are also thought to play a role, since no such correlation was found for the remaining two sub-groups of printers using heating lamps, or for the printers using heating strips. In addition, O(3) and total PM(2.5) were not found to be statistically correlated with fuser temperature. (C) 2010 Elsevier Ltd. All rights reserved.