The design and construction of a radiant furnace are described. Acute toxicity experiments were conducted using the radiant furnace as a fire model to produce smoke from Douglas fir (D. fir), southern yellow pine (SYP), and tempered hard- board (THB). The correlation of blood carboxyhemoglobin (COHb) and mortality to irradiation time, smoke concentration, mass loss, carbon monoxide (CO) con centrations, and carbon dioxide (CO2) concentrations is reported. At 2.5 w/cm 2, toxicity was characterized by post-exposure lethality with sublethal COHb ex posure levels. LC 50's (median lethal concentrations) were 59.5 (D. fir), 66.6 (SYP), and 86.5 mg/L (THB) and the IT50's (median lethal irradiation times) were 6.8 (D. fir), 6.9 (SYP), and 9.3 minutes (THB). For 5.0 w/cm 2, most deaths occurred dur ing the 30 minute exposure and COHb levels were in a lethal range. LC50's at 5.0 w/cm2 were 101.6 (D. fir), 100.1 (SYP), and 58.1 mg/L (THB), and the IT50's were 2.7 (D. fir, SYP) and 2.6 minutes (THB). Data were compared to those produced in other combustion toxicity test methods, and the advantages of the radiant furnace with continuous sample mass-loss monitoring are discussed.