The electroretinogram (ERG) represents the summed electrical response of the retina to light stimuli. It can be recorded from human subjects non-invasively in vivo using electrodes placed around the eye. In response to flashes of light of a range of strengths, an initial negative-going wave is observed (the a-wave), followed by a positive-going wave (the b-wave). Stimulus protocols can be employed to estimate dark-adapted rod-driven and cone-driven response components. Application of mathematical models to the a-wave allow parameters relating to retinal current flows to be extracted from the human eye. This can be applied to probe retinal function in health and disease. Examples will be given in this talk. The ERG can also be used to track kinetics of retinal light and dark adaptation, including by monitoring the b-wave elicited by dim flashes delivered during and following exposure to different backgrounds (Cameron et al., 2006). We used a similar protocol to track dark adaptation in healthy volunteers and patients with selected diseases. Participants’ pupils were pharmacologically dilated. ERGs were recorded using conductive fibre electrodes in response to the dim white flashes (c.0.02 scotopic cd m-2 s) delivered following the extinction of a white background (30 photopic and c.85 scotopic cd m-2), after steady state exposure to this background. Flash series were repeated every 2 min for 20-60 min, and b-wave amplitudes (normalised to their final level) were plotted against time in the dark. Participants gave informed consent; the study had local ethics committee approval and complied with the tenets of the Declaration of Helsinki. The following participants underwent recordings: 7 healthy volunteers (aged 21-82 years); a patient (aged 50) with early-onset widespread retinal drusen; two patients (ages 42, 49), with molecularly confirmed Sorsby Fundus Dystrophy (SFD); a 70 year old with Vitamin A deficiency (VAD), both after treatment and at varying levels of deficiency. Responses were initially of low amplitude but then recovered gradually in the dark. In healthy participants (and the participant with early-onset drusen), recovery kinetics were similar: amplitudes reached the final level at c.20 min (and half the final level by c.10 min). The VAD patient showed normal recovery following treatment, but no recovery when markedly deficient, and a slowed recovery when mildly deficient. The SFD patients displayed recoveries similar to that found in the VAD patient when mildly deficient. Although abnormalities in SFD on clinical examination are largely in the macula, these findings confirm delayed recovery affecting the retina as a whole, consistent with “ocular vitamin A deficiency” in SFD. The protocol can be useful in investigating conditions in which photoreceptor dark adaptation might be impaired.