Eleftheria Pissadaki and Paul Bolam addressed, in the latest issue of Frontiers in Computational Neuroscience, the question why dopamine neurons of the substantia nigra pars compacta are paricularly sensitive to degeneration in Parkinson's disease and its models (further discussion in Movement Disorders). We hypothesize that their massive, unmyelinated axonal arbour that is orders of magnitude larger than other neuronal types, puts them under such a high energy demand that any stressor that perturbs energy production leads to energy demand exceeding supply and subsequent cell death. A prediction of this is that susceptible dopamine neurons will have a higher energy demand than less susceptible dopamine neurons. Our approach to this prediction was theoretical. By creating a biophysical compartmental model of SNc dopamine neurons, we examined the energetic impact of their extensive, unmyelinated axonal arbour. Our main findings show that the energy demand associated with action potential conduction is related, in a supra-linear fashion, to the axonal size and complexity. Our results also suggest that the axons of the SNc dopamine neurons may serve as a third physical site, in addition to the soma and the dendrites, for generating oscillatory behaviour. We showed that synaptic stimulation is necessary to ensure reliable propagation throughout the axonal arbors of neurons with larger axonal arbours and, that besides the considerable energetic cost of calcium currents, their presence augments action potential invasion into the very distal axonal terminals. Our model predicts that action potentials at the axonal endings are metabolically inefficient which may contribute to the loss of striatal terminals occurring in animal models of PD. Our data thus suggest that SNc dopamine neurons, particularly in humans, whose axons we estimate to give rise to more than 1 million synapses and have a total length exceeding 4 m, are at a distinct disadvantage with respect to energy balance which may be a factor in their selective vulnerability in PD.