Pulsed-neutron experiments were performed on the 330-MW Fort St. Vrain high-temperature gas-cooled reactor (HTGR) to determine the reactivity of the core for various control rod configurations while the reactor was still subcritical. For all configurations the reactivity was inferred from the in-hour equation using the measured decay constant and a calculated generation time. For the configurations near critical, both the reactivity and generation time were determined using the extrapolated area-ratio method. The originally calculated (i.e., predicted) reactivities agreed poorly with those inferred from the experiments. However, by adding 5 ppm of boron to the reflector calculational model, the calculated generation time was significantly reduced. This brought the inferred reactivity into good agreement with that calculated for all control rod configurations. This emphasizes the dependence of the interpretation of pulsed-neutron experiments on calculations and the importance of the reflector in a large HTGR. Novel aspects of these experiments included the following: extensive two-dimensional computer simulations were performed prior to the experiments to determine the optimum source and detector locations; the neutron generation time was measured near critical by pulsing two different control rod configurations; all the data were fit by least squares to a sum of exponentials corresponding to one or two prompt modes and six delayed sub-modes; and an objective procedure using “tornado plots ” was developed to determine the starting channel for the least-squares analysis.