A detailed kinetic network for homogeneous CH 4 -O 2 -NO x reactions is used to estimate maximum attainable formaldehyde (and methanol) yields and to identify elementary steps that lead to the observed enhancement effects of NO x on CH 4 oxidation rates, to HCHO yield limits, and to NO x losses to unreactive N-compounds. NO x was shown previously to increase CH 4 oxidation rates and HCHO yields in CH 4 -O 2 reactions, but maximum yields were low (<10%) and intrinsic kinetic limits were not rigorously examined. We show here that the CH 4 oxidation rate increases because NO 2 reacts with CH 4 during an initial induction period. NO and NO 2 lead to similar effects, except that residence times required for a given yield are higher for NO feeds because NO-NO 2 interconversion must first occur. CH 4 leads to supra-equilibrium NO 2 concentrations because HO 2 formed during HCHO oxidation reacts with NO to form OH and NO 2 faster than NO 2 can decompose to NO. Oxygenate selectivities decrease with increasing CH 4 conversion, because weaker C-H bonds in HCHO and CH 3 OH relative to CH 4 lead to their fast sequential oxidation to CO and CO 2 . Rate-of-formation analyses show that NO x molecules introduce more effective elementary steps for the formation of CH 3 O intermediates and for its conversion to HCHO, but H-abstraction from CH 4 and HCHO remains the predominant step in controlling rates and selectivities in the presence or absence of NO x . Without NO x , OH radicals account for all H-abstraction reactions from CH 4 , while HCHO reacts with OH but also with less reactive H and HO 2 radicals. NO x increases HCHO yields by converting these less reactive H and HO 2 radicals to OH radicals, which become the predominant H-abstractor for both CH 4 and HCHO and which react less selectively with HCHO than do H and HO 2 . Kinetic selectivity, based on C-H bond energy differences between CH 4 and HCHO, becomes weaker with increasing radical reactivity and increasing reaction temperature. Maximum HCHO yields of 37% are theoretically possible for radicals that abstract H from CH 4 and HCHO at equal rates, but radical species prevalent during CH 4 -O 2 -NO x reactions lead to maximum HCHO yields below 10% under all conditions. Higher yields appear unlikely with more reactive radicals, because their reactivity would lead to cascade reactions that form species with greater kinetic sensitivity to C-H bond energies. Maximum C 1 -oxygenate yields increase with increasing O 2 pressure, suggesting that the O 2 distribution along a reactor will not improve HCHO yields but may prove useful to inhibit NO x losses to less reactive N-compounds.