The high-spin → low-spin relaxation in spin-crossover compounds can be described as non-adiabatic multi-phonon process in the strong coupling limit, in which the low-temperature tunnelling rate increases exponentially with the zero-point energy difference between the two states. Based on the hypothesis that the experimental bond length difference between the high-spin and the low-spin state of not, vert, similar0.2 Å is also valid for low-spin iron(II) complexes, extrapolation of the single configurational coordinate model allows an estimate of the zero-point energy difference for low-spin complexes from kinetic data. DFT calculations on low-spin [Fe(bpy)3]2+ support the structural assumption. However, for low-spin [Fe(terpy)2]2+ the relaxation rate constant shows an anomalous behaviour in so far as it is more in line with spin-crossover systems. This is attributed to very anisotropic bond length changes associated with the spin state change, and the subsequent breakdown of the single mode model.