Hypothalamic vasopressin neurons fire action potentials (spikes) in alternating periods of activity (phasic bursts) and silence to optimize the release of vasopressin (the anti-diuretic hormone) from the posterior pituitary gland. During phasic bursts, spikes are superimposed on plateau potentials, generated by summation of post-spike depolarizing after-potentials (DAPs). Using extracellular single unit recording in vivo, we have shown that kappa-opioid receptor antagonist administration increases the activity of vasopressin neurons by increasing the firing rate during, and the duration of, the active periods (1). However, kappa-opioid receptor antagonists do not alter the firing rate of vasopressin neurons at the beginning of bursts, but rather the increases in activity emerge as bursts progress, indicating that there must be a progressive endogenous activation of kappa-opioid receptor mechanisms during bursts (2). By contrast to the effects of kappa-opioid receptor blockade, V1 vasopressin receptor antagonist administration increases firing rate throughout bursts in anaesthetised rats, indicating that the actions of endogenous vasopressin are generally inhibitory and tonically present (2). The kappa-opioid peptide, dynorphin, is co-packaged in vasopressin neurosecretory vesicles and exocytosed from vasopressin cell dendrites during periods of activity; our data indicate that this dendritically released dynorphin feeds back to inhibit vasopressin neurons, terminating activity. Using intracellular sharp electrode recording in vitro, we have demonstrated that DAPs are subject to activity dependent inhibition and that this inhibition is prevented by neurosecretory vesicle depletion or kappa-opioid receptor antagonist administration, but not by oxytocin/vasopressin receptor antagonist administration or mu-opioid receptor antagonist administration, indicating that endogenous activation of kappa-opioid receptors is responsible for activity dependent inhibition of DAPs (3). We have also demonstrated that kappa-opioid receptor antagonist administration enhances plateau potential amplitude during spontaneous firing to increase post-spike excitability and firing rate (4). Overall, these data are consistent with hypothesis that dendritic dynorphin terminates firing in vasopressin neurons through activity dependent plateau potential inhibition (5). During dehydration, vasopressin cells display a higher frequency of activity than normal, which should decrease the duration of active periods due to the feedback inhibition defined above. However, the active periods are actually prolonged during dehydration, indicating that dehydration might induce adaptations in vasopressin neurons that reduce, and/or overcome, the effects of activity dependent feedback; we are currently investigating whether kappa-opioid receptor antagonism is less effective at increasing vasopressin neuron activity in dehydrated rats.