The Nottingham Onset Schedule (NOS) is a short, guided interview and rating schedule to measure onset in psychosis. Onset is defined as the time between the first reported/observed change in mental state/behaviour to the development of psychotic symptoms. Onset is conceptualised as comprising of (i) a prodrome of two parts: a period of 'unease' followed by 'non-diagnostic' symptoms; (ii) appearance of psychotic symptoms; and (iii) a build-up of diagnostic symptoms leading to a definite diagnosis. Twenty consecutive cases of first-episode psychosis were administered the NOS schedule to determine its psychometric properties including inter-rater and test-retest reliability. Its clinical and research potential as a reliable measure of duration of untreated psychosis (DUP) was assessed in a cohort of 99 cases of first-episode psychosis (56 schizophrenia, 43 affective psychoses). NOS identified all prodromal symptoms previously reported in other studies. There was high degree of inter-rater and test-retest reliability for all components of NOS. Duration of untreated psychosis was significantly longer (p<0.05) in schizophrenia (mean 179 days, S.D. 344; median 52 days) than in affective psychosis (mean 15 days, S.D. 116; median 12 days) but there were no gender differences between lengths of prodrome or treatment delays. The NOS provides a standardised and reliable way of recording early changes in psychosis and identifying relatively precise time points for measuring several durations in emerging psychosis. The scale is easy to use and is not time-consuming or labour intensive. Onset, as measured by NOS, is significantly longer in schizophrenic disorders than in affective psychosis. A small proportion of schizophrenia cases have very long DUP. Some cases with schizophrenia receive anti-psychotics in the prodromal phase, prior to the emergence of frank psychotic symptoms.