Infrared emission from vibrationally excited ozone was monitored as a function of time following pulsed laser photolysis of O-3/O-2 mixtures With total pressures from 300 to 1800 Torr at 295 K. The emission data obtained at 9.6 mu m were analyzed by nonlinear least squares and by constructing chi(2) surfaces. The results are entirely consistent with a conventional mechanism that includes the following reactions: (1a) O-3 + hv --> O(D-1) + O-2(a(1)Delta); (1b) O-3 + hv --> O(P-3) + O-2; (2a) O(D-1) + O-2 - O(P-3) + O-2(1 Sigma(g)(+)); (2b) O(D-1) + O-2 O(P-3) + O-2; (3) O(P-3) + O-2 + O-2 --> O-3(V) + O-2; (4) O-3(v) + O-2 --> O-3 + O-2; (5a) O-2(1 Sigma(g)(+)) + O-3 --> O + O-2 + O-2; (5b) O-2(1 Sigma(g)(+)) + O-3 --> O-3(v) + O-2; (6) O-2(1 Sigma(g)(+)) + O-2 --> O-2 + O-2. There is no evidence for participation by ozone excited electronic states, but the reaction time scales are not well separated, leading to complexities in the analysis. The measured rate constants k(3) (+/-sigma) = (6.0 +/- 1.1) x 10(-34) cm(6) s(-1) and k(5) (+/-sigma) = (2.26 +/- 0.15) x 10(-11) cm(3) s(-1) are in good agreement with literature values. The phenomenological rate constant k(4) (+/-sigma) = (1.2 +/- 0.2) x 10(-11) cm(3) s(-1) is consistent with a model for vibrational deactivation. The measured value for the ratio k(1a)k(2a)/(k(1)k(2)) = 0.86 +/- 0.13 is combined with a literature value for k(1a)/k(1) to give an improved estimate for k(2a)/k(2) = 0.95 (+0.05/-0.13).