Results obtained using a first-order conditional moment closure (CMC) approach to modeling turbulent, nonpremixed flames of methane are presented. Predictions are based on both k-epsilon and second-moment turbulence closures and use the GRI-Mech 2.11 kinetic scheme involving 279 reactions and 49 species and the GRI-Mech 3.0 scheme applying 325 reactions and 53 species. To provide a comprehensive assessment of the ability of these methods to predict a wide range of flames, comparisons are made with experimental data on three piloted and two unpiloted flames, the majority of which exhibit no local extinction effects. It is concluded that first-order CMC modeling is capable of yielding reliable predictions for flames with little or no extinction effect, with the only exception being NO, which is overpredicted at most locations and stoichiometries. Results derived using the two turbulence closures are in general in close accord, although real space predictions demonstrate the superiority of results derived on the basis of the second-moment turbulence closure. Calculation of flames that exhibit extinction effects demonstrate a gradual deterioration of results with increasing Re, pointing to the requirement for second-order CMC modeling as such effects become more significant. Anomalies are observed in regard to NO, which is significantly overpredicted for the piloted flames, but less so for the unpiloted flames. These results demonstrate a requirement not only to explore second-order CMC modeling approaches for such flames, but also to examine the application of different kinetic schemes, including a detailed investigation of the mechanisms and rates employed for NO chemistry. (C) 2004 The Combustion Institute. Published by Elsevier Inc. All rights reserved.