Objective: Olfactory identification deficits and their relationship to negative symptoms in patients with schizophrenia were examined in patients with recent-onset psychosis, the majority of whom were neuroleptic naive. Method: Seventy-four inpatients with a first episode of psychosis (27 with schizophrenia or schizophreniform disorder, nine with schizoaffective disorder, 17 with affective psychoses, and 21 with other psychoses), 49 of whom had not received antipsychotic medication, were compared to 38 age- and gender-matched normal subjects. Olfactory identification ability was assessed with the University of Pennsylvania Smell Identification Test. Forty patients and 13 comparison subjects were reassessed at 6 months to examine whether olfactory deficits were specific to schizophrenia or schizophreniform disorder and were stable over time. Results: At baseline, the patients had significant impairment in olfactory identification ability compared to the normal subjects. This difference persisted after controlling for gender, premorbid or current IQ, smoking history, cannabis use, or the effects of medication. Diagnostic subgroups did not differ in olfactory identification ability. The deficits remained stable at 6-month follow-up and were associated with negative symptoms at both time points. No relationship was found between olfactory identification ability and length of either untreated psychosis or illness prodrome. Conclusions: Impairment in olfactory identification ability was apparent from the outset of psychotic illness and was not specific to schizophrenia or schizophreniform disorder. No change in the degree of this deficit was found after patients were stabilized and had responded to medication. The deficit could not be explained by peripheral factors that might contribute to olfactory identification ability, suggesting that it reflects central mechanisms.