The utility of vertical propagation by the Green's function to test response linearity has been explored for magnetic force microscopy (MFM) data from current-carrying wires, by comparing the measured signal at various tip heights to the corresponding propagated MFM signals. Application of a one-dimensional Green's function was found to be sufficient to predict signal height variation for sample regions of high to moderate field symmetry. For regions of high field asymmetry, the two-dimensional Green's function was required to obtain good prediction of the height variation. Agreement between the measured and propagated signals was generally within 5%, except at the tails where the signal is not well behaved. The quality of agreement deteriorates gradually with the size of the height propagation. The good agreement spanning a decade of tip and sample separation suggests that the MFM signal is not significantly affected by nonlinearities and can thus be interpreted in terms of classical electromagnetic relations governing current flow. © 2002 American Institute of Physics.