The hydrosilylation of alkenes catalyzed by Fe(CO)5 is an intricate process involving formation of alkylsilane, vinylsilane, and alkane. Herein, we represent a computational study of this reaction to determine the underlying reaction mechanism using density functional theory (DFT) techniques. On the basis of an extensive exploration of the potential energy surfaces, the modified Chalk–Harrod mechanism was found to be competitive with the Chalk–Harrod mechanism in the stoichiometric reaction of ethylene and trimethylsilane. The source of product selectivity was predicted to be determined by the relative stability of transition state TS(4b–5b), which is for the ethylene-insertion into FeSi bond from (H)(Me3Si)Fe(CO)3(C2H4) to (H)(C2H4SiMe)Fe(CO)3, and TS(5a–7a1) for the conversion of ethylene hydrometallation product (Me3Si)Fe(CO)3(η2-H)(C2H4) to (C2H5)(Me3Si)Fe(CO)4. The relative free energy difference of 3.27kcal/mol between these two transition states gives a percentage ratio of 79:21 for C2H5SiMe3 to C2H4SiMe3, which is qualitatively in agreement with the experimental observations. The largely excessive ethylene favors not only the formation of Fe(CO)3(C2H4)(H)(SiMe3) 4b but also the release of vinylsilane from Fe(CO)3(H2)(C2H3SiMe3) or Fe(CO)4(C2H3SiMe3) via the addition of another ethylene molecular to the metal center. Alternatively, in the presence of excessive R3SiH, species Fe(CO)4(H)(SiR3) will be dominate over Fe(CO)4(C2H4). In the catalytic process, Fe(CO)4(H)(SiR3) plays the main role of active species. The experimental findings were rationalized in terms of two reaction pathways, the accessibility of which depended on the ratio of silane/alkene.