'P50 suppression', an index of sensory gating, has demonstrated utility in schizophrenia research. It is widely reported that P50 suppression is deficient in schizophrenia patients and an endophenotypic marker for the disorder. However, unresolved methodological issues including the unestablished reliability of the measure, unknown effects of time-on-task and long protocol undermine its usefulness. In order to address these methodological issues, twenty healthy participants' P50 suppression was measured in a long P50 paradigm. This enabled the measurement of within-session reliability, temporal course of P50 suppression, and effects of the inter-pair interval parameter. Results indicated good within-session reliability for P50 suppression (ICC = .668); changes in P50 suppression across the session (a 31% increase over the 78 minute recording); and comparable P50 suppression at long (9 s) and short (3 s) inter-pair intervals. It is concluded that given appropriate conditions, P50 suppression can be measured reliably within-session. Further, time-on-task effects need to be taken into account when measuring P50 suppression in a long paradigm or calculating reliability. Lastly, the inter-pair interval can be substantially shortened in studies with healthy participants allowing for an appreciable reduction in P50 suppression recording time.