A numerical model was developed to simulate Ni composition profiles developed around γ (FeNi) precipitates growing during martensite (α2) decomposition in Fe-Ni at low temperatures (300 °C to 400 °C). The model is based on the theory of partial interface reaction control of the precipitate growth process. Experimental Ni composition profiles were measured across γ -α2 interfaces using high spatial resolution analytical electron microscopy. The simulated Ni composition profiles show good agreement with the experimentally measured profiles, indicating that partial interface reaction control of the γ growth is a reasonable assumption. The diffusion coefficients and the interface mobilities were estimated from the simulations. The activation energy for diffusion in the α2 matrix obtained from the computer model is 0.7 eV with an error range from 0.58 to 0.98 eV. This value is similar to the activation energy for diffusion obtained from the calculated γ -α2 interface mobility (0.62 eV with an error range from 0.57 to 0.67 eV). This result is consistent with the observed high dislocation density in the α2 matrix. Both these values of the activation energy are well below that for lattice diffusion (223C;3 eV). Therefore, it is concluded that the prevailing diffusion mechanisms at these temperatures are short circuit (defect) diffusion in the α2 matrix and rapid diffusion across the γ -α2 interface.