The reaction of atomic oxygen with ethylene is a fundamental oxidation step in combustion and is prototypical of reactions in which oxygen adds to double bonds. For O-3 + C2H4 and for this class of reactions gen-erally, decomposition of the initial adduct via spin-allowed reaction channels on the triplet surface competes with intersystem crossing (ISC) and a set of spin-forbidden reaction channels on the ground-state singlet surface. The two surfaces share some bimolecular products but feature different intermediates, pathways, and transition states. The overall product branching is therefore a sensitive function of the ISC rate. The O-3 + C2H4 reaction has been extensively studied, but previous experimental work has not provided detailed branching information at elevated temperatures, while previous theoretical studies have employed empirical treatments of ISC. Here we predict the kinetics of O-3 + C2H4 using an ab initio transition state theory based master equation (AITSTME) approach that includes an a priori description of ISC. Specifically, the ISC rate is calculated using Landau-Zener statistical theory, consideration of the four lowest-energy electronic states, and a direct classical trajectory study of the product branching immediately after ISC. The present theoretical results are largely in good agreement with existing low-temperature experimental kinetics and molecular beam studies. Good agreement is also found with past theoretical work, with the notable exception of the predicted product branching at elevated temperatures. Above similar to 1000 K, we predict CH2CHO + H and CH2 + CH2O as the major products, which differs from the room temperature preference for CH3 + HCO (which is assumed to remain at higher temperatures in some models) and from the prediction of a previous detailed master equation study. (C) 2016 by The Combustion Institute. Published by Elsevier Inc.