To demonstrate time dependence of diffusivity and kurtosis parameters in gray and white matter of rat spinal cord.The interaction of water molecules with microstructural impediments in biological tissue leads to time dependence of diffusion parameters, yet experimental evidence has been inconclusive[1, 2]. Although time dependencies were observed in q-space imaging[3, 4], studies investigating time dependence of kurtosis parameters[5] in neural tissue are lacking, presumably due to time consuming acquisitions. Nevertheless, the functional form of these time dependencies is expected to bear significant microstructural information. Here we demonstrate significant time dependence of DTI parameters in white and gray matter of ex-vivo rat spinal cord, and, for the first time in neural tissue, time dependence of mean kurtosis, using a recently proposed fast kurtosis imaging protocol[6, 7].Long-Evans rats, 14-16 week old, were perfused intracardially using 4% PFA, and the cervical region of the spinal cord was isolated (n=3), washed in PBS, aligned in a 5mm NMR tube, and scanned using an ultrahigh-field 16.4T Bruker Aeon Ascend magnet (700MHz). To keep a constant echo time, a Stimulated-Echo based Echo-Planar-Imaging DTI sequence was executed using 15 directions in a single b=1.2 ms/μm2 shell, with isotropic in-plane resolution of 34x34 μm2 and slice thickness of 900 μm, TR/TE = 4000/28.5 ms, and the diffusion times (Δ) were varied between 7 and 150 ms. The 1-9-9 scheme[7] was executed using identical imaging parameters, and the 9 directions in two b-value shells b=1.0 and 3.2 ms/μm2, with Δ between 4 and 130 ms. White matter ROIs were selected in order of increasing axonal diameters as in refs. [8, 9], see Fig. 1, which also shows 6 gray matter ROIs used.The time dependence of the parallel diffusivity $$$D_\parallel$$$ and perpendicular diffusivity $$$D_\perp$$$ are shown in Fig. 2. Within the experimental uncertainty, $$$D_\parallel$$$ is decreasing roughly linearly with time in all ROIs, whereas $$$D_\perp$$$ appears more nonlinear. Although there is a tendency for areas with larger axons to have higher $$$D_\perp$$$, the order of the curves reveals no clear-cut relation to the axonal radii, indicating that the diffusion coefficients are modulated also by other factors, e.g. the extracellular space (such as packing). The long time behavior suggests that the tortuosity regime is not reached in the parallel direction, in contrast to the perpendicular direction, where tortuosities can be estimated roughly by $$$D(t=0)/D(t\to \infty )$$$, using the first time point for the numerator and the average of the last three for the denominator, to yield 1.83, 1.77, 2.50, 1.98, 1.80, 1.67, and 1.92, for regions (A-G) averaged over animals. Note that the estimation of these values neglects the intra-axonal water fraction contribution to the diffusivity, argued to be subdominant in ref. [10]. Fractional anisotropy is generally lower in the ROIs with larger diameters, and increases strongly with time initially with a tendency to level out somewhat above ~70 ms. In Fig. 3, the output from the 1-9-9 protocol is shown. The behavior of mean diffusivity in white matter is consistent with Fig. 2, and we note that even in gray matter, mean diffusivity depends on time, with a roughly linear behavior. The mean kurtosis tensor $$$\bar{W}$$$ was only reliably determined for 4 diffusion times between 4 ms and 30 ms, due to SNR limitations and gradient cross terms effects, which can be a confound in the 1-9-9 scheme. In gray matter, $$$\bar{W}$$$ decreases smoothly with time, consistent with diffusion becoming more Gaussian at long diffusion times. In contrast, $$$\bar{W}$$$ increases initially in white matter (robust across animals), perhaps indicating that water has not fully sampled the intra-axonal space in the transverse direction when the diffusion time is below 10 ms. However, directionally resolved kurtosis values (underway), which cannot be estimated with the fast protocols, are necessary for a deeper understanding of the form of the kurtosis time dependence in white matter. We demonstrated time dependence of parallel and perpendicular diffusivity in white and gray matter of rat spinal cord, robust across three animals. We presented the first demonstrations of time dependent kurtosis in neural tissue using the 1-9-9 fast kurtosis protocol, and mean kurtosis was found to depend on time in both white and gray spinal cord matter, with a nontrivial time dependence in white matter. These may have significant implications both for analyzing and planning kurtosis experiments as well as for probing the neural tissues using DKI.