The CH5N+2 system has been investigated by ab initio MO calculations at the SDCl/6-31G**//6-31G** level of theory and by mass spectrometric experiments. The calculations confirm earlier experimental observations that the diazapropylium ions CH3NHNH+, 1+, CH3NNH+2, 2+ and CH2NHNH+2, 3+ and the hydrazonium ion CH2NNH+3, 4+, are stable species. Theory predicts 1+ and 2+ to be higher in energy than 3+, by 7–8 kcal mol−1, causing a serious discrepancy with existing experimental values, which indicate that 1+ and 2+ are considerably more stable than 3+. The theoretical values are insensitive to inclusion of electron correlation in the geometry determinations. From a critical evaluation of existing energetic data for N2H+3, CH5N+2 and C2H7N+ ions, and collision experiments on deuterium labelled species, it is concluded that theory is correct and that several reported appearance energy (AE) measurements on hydrazines are probably in error owing to interferences from traces of amines.From AE measurements not affected by these interferences, ΔHf(3+) is proposed to be 204 ± 5 kcal mol−1 from which theory leads us to recommend ΔHf values of 211 ± 5 kcal mol for 1+ and 2+. Ab initio calculated proton affinities for HNNH, CH3NNH and CH3-NNCH3 lead to proposed enthalpies for 1+ and 2+ which are consistent with these values.Theory further predicts the ring-closed form of 3+ to be a remarkably stable species (16.7 kcal mol−1 above 3+) but the hydrogen bridged entity CH2 = NH 2H2+ previously proposed to be responsible for the facile interconversion between 3+ and 4+, is not a minimum on the potential energy surface. In fact, large energy barriers (42–63 kcal mol−1) prohibit interconversion among ions 1+, 2+, 3+ and 4+, via 1,2-H shifts.Metastable CH5 N+2 ions dissociate to HCN + NH+4 and to HCNH+ + NH3 and in agreement with experiment, the reacting configuration for HCN formation is the ion 4+. Formation of HCN from 4+ is exothermic but the reverse barrier is large (84 kcal mol−1) thus accounting for the persistence of 4+ in the gas phase and in neutral solvents. The small kinetic energy release (KER) accompanying this reaction is rationalized in terms of ion/dipole attraction in the dissociating [HCN⋯NH4]+ complex.