CO₂ has been used for many years to model panic anxiety. Much of this research has used the 35% single vital capacity inhalation or 5% CO₂ inhaled for up to 20 min. We have used a 20-min inhalation of 7.5% CO₂ in volunteers to produce symptoms of anxiety and fear, and the associated physiological arousal (Bailey et al. 2005). We have also examined the effects of a benzodiazepine agonist, lorazepam (LZP) on these measures. The local research ethics committee approved the study protocols. In the first study twenty (6 female) healthy volunteers (mean age 25.2 ± 5.02 years, mean ± S.D.) who met our inclusion criteria underwent the procedure. They inhaled medical air and 7.5% CO₂ for 20 min with a 15-min interval between gases. During each inhalation, blood pressure (BP) and heart rate (HR) were measured continuously using the Finapres. Subjective state was assessed by visual analog ratings. The second study examined the effects of 2mg LZP in 12 (4 female) healthy volunteers (mean 27.4 ± 2.5 years) in a double-blind, placebo-controlled design. BP and HR values recorded beat-by-beat during the inhalation period were expressed as a mean for the total duration. During CO₂ inhalation, systolic BP and HR were significantly increased compared with air. Data are shown in Table 1. At peak effects of gas 7.5% CO₂ increased subjective feelings of anxiety, fear, feel like leaving, nervous tense and worried and decreased feelings of being relaxed and happy. In the lorazepam study the effects of CO₂ on BP and HR were similar. However, in both gas conditions LZP increased HR further (air + placebo 69 ± 3.0, mean ± S.E.M.), air + LZP 77 ± 2.8, CO₂ + placebo 74 ± 2.9, CO₂ + LZP 84 ± 3.7). The subjective effects of CO₂ were significantly attenuated by LZP (e.g. subjective anxiety at peak CO₂: placebo 31.3 ± 6.8), LZP 17.9 ± 2.7). These experiments demonstrate measurable physiological effects of inhaling 7.5% CO₂ in healthy, non-anxious volunteers. These effects are of fairly rapid onset and are sustained for the duration of the inhalation. This cardiovascular response to 7.5% CO₂ differs to that produced by 35% CO₂ where BP is increased but a bradycardia is seen (Argyropoulos et al. 2002). LZP reduced the subjective effects of 7.5% CO₂, but did not change the cardiovascular effects. The finding of increased heart rate was unexpected and contradicts the belief that CO₂-induced anxiety can partially be explained by a cognitive misinterpretation of physical symptoms.