The use of melatonin as a hypnotic in the elderly is not new [Fainstein et al. 1997]. Despite this, melatonin has been struggling to find an appropriate therapeutic niche for some time now. Melatonin is a naturally occurring hormone secreted by the pineal gland located in the centre of the brain, between the laterally positioned thalamic bodies. It is biosynthesized from tryptophan via serotonin, and its diverse functions include restoration of circadian rhythmicity (physiological sleep onset and regulation of wake—sleep cycle), cyclic hormone release and regulation of the immune system to name a few [Jansen et al. 2006]. We report a case of its use in an 88-year-old, blind female with dementia. The female concerned was admitted for long-term residential care in June 2006 from a tertiary referral hospital, following a fall and subsequent hip replacement. Comorbidities included end-stage glaucoma, advanced dementia (aetiology unspecified) and hypertension. Over a period of 4 years, the patient gradually decompensated with complete loss of vision, and became fully dependent for activities of daily living. Several admission medications, which included rivastigmine and amisulpride, had been discontinued. Regular medications by September 2010 included quetiapine for behavioural and psychological symptoms associated with her dementia (an unlicensed indication), zopiclone and aspirin. Nurses described overt disturbance of her wake—sleep pattern associated with restlessness, wakefulness and nighttime vocalization, resulting in sleep disruption to other residents. This night-time behaviour resulted in frequent and prolonged sleeps during the afternoon, for several hours at a time. Despite regular use of night-time zopiclone, frequently prescribed in combination with ‘as required’ alprazolam, amelioration of the nocturnal symptoms was not achieved. Postulated reasons for her altered sleep patterns included blindness and tolerance to benzodiazepine therapy. At a medication review meeting, the introduction of melatonin was proposed in an attempt to synchronize her wake—sleep cycle. Subsequently, melatonin in a controlled release formulation was commenced at a dose of 2 mg nightly. The immediate and sustained effects on the patient have been remarkable. Significant improvements in daytime somnolence and a reduction in night-time awakening and calling have been achieved, with consequent benefits to other residents. Medication requirements in terms of ‘as required’ alprazolam have been profoundly reduced and zopiclone has been discontinued. The beneficial effects on sundowning have been maintained 6 months post initiation of therapy. This is despite a license restriction to limit use to 3 weeks of therapy in patients aged 55 years of older, for the treatment of primary insomnia. At future reviews, further dose reduction of quetiapine therapy will be considered. The role of melatonin in controlling circadian rhythm is necessary for a normal wake—sleep pattern. Factors contributing to decreases in melatonin are diverse. The decrease in secretion of endogenous melatonin with aging is well documented [Olde Rikkert and Rigaud, 2001], and more profound reductions are reported in populations with dementia [Cardinali et al. 2006]. Benzodiazepines, which are widely used in the elderly population for the initiation of sleep, as in this patient, have also been reported to reduce melatonin production [Garfinkel et al. 1997]. A recent Cochrane review concluded, however, that there was insufficient evidence to support the effectiveness of melatonin in the management of cognitive and noncognitive sequelae of dementia [Jansen et al. 2006]. In the blind population due to the absence of light cues, disturbances of circadian rhythms are common. These disturbances can result in delays in circadian cycle timing by as much as 60–70 minutes per day [Sack et al. 2000]. Even if they try to sleep at regular times, they typically sleep well only a few days a month, when their internal clocks fall in synchronization with preferred daily schedules. These chaotic free-running circadian rhythms have been successfully entrained with administration of exogenous melatonin resulting in appropriate phase shifts in sleep patterns [Sack et al. 2000]. The decision to commence melatonin in our patient was based primarily on the temporal relationship between blindness and wake—sleep dysrhythmias, but results seem to indicate a beneficial effect on more than just her sleep pattern. Since melatonin is a naturally occurring substance, it is not considered a drug in many countries; however, it does require a medical prescription within the European Union. Interest has been expressed in the effects of high melatonin milk (milk sourced from animals milked at night) on sleep and activity levels in the elderly, where even ultra-low doses of melatonin are suggested to be of benefit [Valtonen et al. 2005]. The use of melatonin in the visually impaired with dementia might represent a ‘novel’ hypnotic treatment in this specific population, but more formal studies would be required to inform any assertions.