1) To evaluate a novel theoretical model for in vivo axonal Mn(2+) transport with MRI data from the rat optic nerve (ON); and 2) to compare predictions from the new model with previously reported experimental data. Time-resolved in vivo T(1)-weighted magnetic resonance imaging (MRI) of adult female Sprague-Dawley rat (n = 9) ON was obtained at different timepoints after intravitreal MnCl(2) injection. A concentration-dependent and a rate-dependent function for the Mn(2+) retinal ganglion cell (RGC) axon entrance was convolved with three different transport functions and each model system was optimized to fit the ON data. The rate-limited input function gave a better fit to the data than the concentration-limited input. Simulations showed that the rate-limited input leads to a semilogarithmic relationship between injected dose and Mn(2+) concentration in the ON, which is in agreement with previously reported in vivo experiments. A random walk transport model and an anterograde predominant slow model gave a similar fit to the data, both better than an anterograde predominant fast model. The results indicate that Mn(2+) input into RGC axons is limited by a maximum entrance rate into the axons. Also, a wide range of apparent Mn(2+) transport rates seems to be involved, different from synaptic vesicle transport rates, meaning that manganese does not depict synaptic vesicle transport rates directly.