The temporal resolution for measuring of vasopressin (VP) release from neurohypophysial nerve terminals in response to various challenges is relatively poor. By combining advances in both fluorescent protein (FP) and transgenic rat technology, we describe a fibre optic approach for monitoring vasopressin (VP) release from the neurohypophysis in response to hyperosmotic stimulation and baroreceptor reflex unloading in vivo. Enhanced green fluorescent protein (eGFP) fluorescence is highly pH dependent (Kneen et al. 1998), being low in acidic conditions, but increasing markedly towards neutral pH. We used this property of eGFP to directly monitor eGFP release from the acidic conditions of the secretory granule to the neutral conditions of the extra-cellular space. Transgenic rats expressing a VP-eGFP fusion gene (Ueta et al. 2005) were anaesthetised with ketamine:medetomidine (60:250 mg kg^-1 i.m.). A femoral artery and vein was catherterised for blood pressure measurement and intravenous administration, respectively. Using our recently developed technology (Bradley et al. 2003), a dual ‘optrode’ consisting of two optic fibres was constructed; one fibre coupled to a 488nm Ar laser was used for excitation and the other fibre attached to a photomultiplier tube was used for signal detection. The optrode was stereotaxically placed immediately dorsal to the neurohypophysis during bolus intravenous injections of hypertonic saline (HS; 0.1ml, 3M NaCl) and sodium nitroprusside infusion (SNP; 100μM; 100μl/min for 5 min) to unload baroreceptors. HS induced a rapid increase in fluorescence (40.2±3.2%; n=5, P<0.05) signal lasting approximately 40 s. This produced a significant increase in plasma omolality from 296±4 to 336±6 mosmol kg^-1. The evoked change in fluorescence was not related temporally to the evoked changes in arterial blood pressure. However, infusion of SNP to induce a sustained fall in arterial pressure was associated with a stimulus-locked increase in fluorescence (19.5±1.5%; n=3). Neither bolus injections nor infusions of saline (0.9%) caused a change in fluorescence. Post-hoc histological analysis confirmed that in those animals where the optrode was placed immediately dorsal to the neurohypophysis was a change in fluorescent signal was detected to HS (n=5) and SNP (n=3) but no fluorescence signal was detected when optrode placements fell distant to the neurohypophysis (n=7). We conclude that optrode recording permits detection of putative vasopressin release from the neurohypophysis in vivo. This approach can be used for real-time monitoring of fluorescent reporter proteins during homeostatic cardiovascular reflexes evoked by physiological stimuli.