Pathways for the reaction of ethene with diazomethane to cyclopropane and dinitrogen catalyzed by Pd(0) complexes have been investigated at the B3LYP level of theory. The computed Gibbs free activation energy of 71.7 U mol(-1) for the most favorable catalytic cycle is by far lower than previously reported computed barriers for Pd(II)-catalyzed pathways of this reaction and is now in the range of experimental expectations. Pd(eta(2)-C2H4)(2) is predicted to be the resting state of the catalyst and the product of a Pd(OAC)(2) precatalyst reduction. The Pd(0) ethene complex is in equilibrium with Pd(eta(2)-C2H4)(kappaC-CH2N2), from which N-2 is eliminated in the rate-determining step. The resulting carbene complex (eta(2)-C2H4)Pd=CH2 reacts without intrinsic barrier with CH2N2 to Pd(eta(2) -C2H4)(2) and N-2 and with ethene to the palladacyclobutane (eta(2) -C2H4)Pd-II[kappaC(1),kappaC(3)-(CH2)(3)]. The N-2 elimination from Pd(eta(2)-C2H4)(2)(kappaC-CH2N2) to (eta(2) -C2H4)(2)Pd=CH2 leads to an overall Gibbs free activation energy of 84.2 U mol(-1). The intramolecular rearrangement of (eta(2)-C2H4)(2)Pd=CH2 to the palladacyclobutane (eta(2)-C2H4)Pd-II[kappaC(1),kappaC(3)-(CH2)(3)] and the subsequent reductive elimination of cyclopropane are facile. At the BP86 level of theory, Pd(0) preferentially coordinates three ligands. Pd(eta(2)-C2H4)(3) is predicted to be the resting state, and the N-2 elimination from the model complex Pd(eta(2) -C2H4)(2)(kappaC-CH2N2) is the rate-determining transition state leading to an overall Gibbs free activation energy of 69.4 kJ mol(-1).