The reaction O(P-3) + N2O is important to models of NOx pollutant and propellant chemistry and to the understanding of the thermal decomposition of N2O, which has historically played a key role in the development of unimolecular reaction theory. The reaction has two important product channels: O + N2O --> NO + NO (Delta H-0 = -36 kcal/mol) (R1); O + N2O --> O-2 + Nz (Delta H-0 = -79 kcal/mol) (R2). Rate coefficients of these reactions have been the subject of several reviews. However, clear reasons why many of the evaluated, nonretained data differ from recommendations have not previously been known. There has been a great deal of controversy over the rate coefficients, particularly for reaction R2. Here, the relevant data are critically evaluated using detailed chemical modeling as an important tool. The results explain many of the discrepancies. Some of the data of central importance in earlier evaluations are shown to be incorrect. Additionally, some important features of the global behavior of the mixtures studied, which had previously not been understood, are explained, and the possible effects of hypothetical H2O contamination on N2O shock tube studies was quantitatively investigated. It is shown that the bulk of the rate coefficient results remaining after the evaluations can be combined with the intermediate temperature results for k(tot) = k(1) + k(2) from FGFAM (Fontijn, A.; Goumri, A.; Fernandez, A.; Anderson, W. R.; Meagher, N. E. J. Phys. Chem., preceding paper in this issue) to obtain fitted recommendations: k(1) = 1.52 x 10(-10) exp(-13 930/T) cm(3) molecule(-1) s(-1) (1370-4080 K); k(2) = 6.13 x 10(-12) exp(-8,020/T) cm(3) molecule(-1) s(-1) (1075-3340 K). Until recently, it was believed rate coefficients of the two product channels were approximately equal over a very wide temperature range. In contrast, the present study has led to the conclusion that reaction R2 dominates below, and reaction R1 above, 1840 K.