The applicability of the quasi-steady assumption for two staggered circular cylinders of equal diameter in cross-flow has been tested for ten geometrical arrangements, for Reynolds number (based on cylinder diameter and free-stream velocity) 4 × 104 ≤ Re ≤ 2 × 105, and reduced flow velocity 15 ≤ U/fD ≤ 300. With one cylinder stationary, the other one is forced to oscillate transversely to the flow with a half peak-to-peak amplitude / cylinder diameter ratio of 0.043, 0.047 or 0.17. The dynamic forces are deduced either from measured pressure distributions, or from direct force measurements on the cylinder, and these measurements are post-processed in the time and frequency domains, respectively. The applicability of the quasi-steady assumption is found to be strongly dependent on the relative positions of the two cylinders, in addition to U/fD. When the cylinder is in a sensitive flow region, where the spatial gradients of the static force coefficients are high, the quasi-steady assumption may not be applicable even for U/fD = 250. Away from the sensitive flow regions, the quasi-steady assumption is usually applicable for U/fD ≥ 50. In general, the fluidelastic forces depart rapidly from predictions based on the quasi-steady assumption with decreasing U/fD below the critical value.