The fabrics associated with protective clothing affect heat stress, which influences productivity and risks of heat-related disorders. This study compared the work limiting effects of five protective coveralls and a semiclothed condition (t-shirt and shorts). Two fabric characteristics determined from bench tests, moisture vapor transmission rate (MVTR), and air permeability were also examined as possible predictors of ensemble performance. A progressive metabolic rate protocol was used where environmentalconditions (T(db) = 32 degrees C; T(pwb) = 26 degrees C) were held constant while treadmill speed was slowly increased. The limiting metabolic rate to just maintain thermal equilibrium was the critical point. At this point, critical speed and critical metabolic rate were noted and total evaporative resistance was calculated for each ensemble. Five acclimatized subjects wore each of the six clothing conditions in a random order. Statistically significant differences were found among the five protective garments and a semiclothed ensemble for critical treadmill speed (S(crit)), critical metabolic rate (M(crit)), and total evaporative resistance (R(e-t)). The semiclothed condition (S(crit) = 1.77 m/sec; M(crit) = 580 W; R(e-t) = 0.0099 kPa m2/W) and ensembles made from spunbonded, melt blown, spunbonded polypropylene (SMS) (1.72 m/sec; 560 W; 0.0135 kPa m2/W) and spunbonded polypropylene (1.67 m/sec; 550 W; 0.0126 kPa m2/W) were able to support higher work rates than fabrics made from Tyvek 1422-A (a nonwoven spunbonded olefin) (1.48 m/sec; 470 W; 0.0183 kPa m2/W) and a microporous film supported by spunbonded polypropylene (1.34 m/sec; 420 W; 0.0231 kPa m2/W). A tightly woven polyester ensemble (1.59 m/sec; 510 W; 0.0130 kPa m2/W) had intermediate values and was not significantly different from either group. Air permeability was a better predictor of fabric work limiting performance than MVTR. An air permeability on the order of 10,000 L/min cm2 bar would have little effect on maximum sustainable work.