An experiment was performed with a mock-up of the core of the Fast Flux Test Facility (FFTF) reactor to evaluate three reactivity measurement methods for application to liquid-metal fast breeder reactors (LMFBR): modified source multiplication measurements with the low-level flux monitor for refueling (∼35 dollars subcritical) of FFTF, noise analysis to 35 dollars subcritical, and inverse kinetics rod drop to∼12 dollars subcritical. To investigate the spatial dependence of these measurement methods and to resolve discrepancies previously reported, detectors were placed in the core, reflector, and radial shield, and experimental data were collected with the reactivity at near delayed criticality to∼35 dollars subcritical. Conclusions from this experiment are the following. Low-level flux monitors in the shield of the FFTF will be adequate for reactivity surveillance during refueling, using the modified source multiplication method calibrated near critical by an inverse kinetics rod drop measurement. The break frequency noise analysis method to−35 dollars with in-core detectors, the modified source multiplication method to−35 dollars, and inverse kinetics rod drop method to -12 dollars with detectors at all locations (corrected for changes in nuclear parameters), yielded the same reactivities within<5%. From reactor physics considerations, breakfrequency noise analysis with in-core detectors is best for monitoring reactivity down to full shutdown since it requires only a simple correction with reactivity that depends on global parameters of the system rather than a correction that depends on the value of the flux at a point or on the inherent source intensity, such as are required for the modified source multiplication method. However, for simple point kinetics interpretation of the results, the measurements must be made only with in-core detectors.