The reactivity perturbation method has been investigated as a possible technique for the assay of spent fuel rods from a 233UO2-ThO2-fueled core. A hard interrogating spectrum was provided at the center of the ARMF-1 core by two B4C filters of different thickness. Rods up to 267 cm (2.67 m) long were pulled through the core at speeds up to 25 cm/min (4.17 × 10−2 m/s), and the time-integrated reactivity worth was measured. The fuel response of both filters was found to be linear over a wide fuel density range, with good fuel sensitivity. Fission product sensitivities for the two filters, obtained both experimentally and calculationally, were very low and in good agreement with one another. Single-measurement uncertainty was 0.6 g at the 25 cm/min (4.17 × 10−3 m/s) pulling speed. Total estimated assay precision, including both systematic and random errors, for a hypothetical assay of 500 rods was ∼0.5%.