If hydride transfer from a donor, A j H, to an acceptor, A i + , involves tunneling, phenomenological theory suggests a characteristic functional relation between rate constants and the equilibrium constant, i.e. between ln (k ij, H /k ij, D ) and In K ij . Tunneling occurs at a heavy atom separation larger than that of the transition state (saddle point). This gives the critical configuration (point of no-return) for H-transfer a larger heavy atom separation than that for D-transfer and leads to the pattern described through a corollary of Marcus theory of atom transfer. The theory is called corner-cutting because it shortens the path which represents the reaction in multi-dimensional potential energy space. If the reactions are moderately spontaneous and the variation in structure which generates the variation in k ij, H , k ij, D and K ij is in the acceptor, the slope of the plot of ln (k ij, H /k ij, D ) against ln K ij is expected to be negative; if the structure variation is in the donor it is expected to be positive. Similar results are expected for proton and hydrogen atom transfer. This behavior has now been observed, for hydride transfer between various nitrogen heterocycles (all of which may be regarded as analogues of the enzymatic cofactor, NAD + ). These reactions have k ij, H /k ij, D ~ 5. The increase of k ij, H /k ij, D , with increasing K ij , reported in this paper, is particularly hard to explain without corner-cutting. The present results, therefore, support the view that some tunneling is involved in most hydrogen transfer reactions, even when no anomalies are obvious.