A series of experiments will be performed on the Large Plasma Device (LAPD) at Basic Plasma Science Facility (BaPSF) at UCLA to understand the propagation of Alfven waves in inhomogeneous plasmas. The current work will focus on two classes of inhomogeneity: gradients in the Alfven speed parallel to the ambient magnetic field and perpendicular to the field. Theory predicts that parallel gradients in the Alfven speed reflect Alfven waves. But, previous experiments conducted on LAPD by the PI’s group observed no such reflection, although wave power transmission through the gradient was greatly reduced. This lack of reflection contradicts the basic theory. Therefore, new experiments are proposed on LAPD to explain this surprising lack of reflection. It will be determined whether enhanced Landau damping in the parallel gradient rapidly damps the Alfven waves. A second explanation for the lack of reflection is that it takes time and energy to set up the boundary layer currents in the plasma that generate the reflected waves. Initially, wave energy goes into driving these currents and then reflection occurs later. This possibility will be tested by exciting long wave trains and observing whether the reflectance changes over time. In perpendicular gradients, Alfven waves on neighboring field lines travel at different phase velocities and the wave fronts become distorted. This is known as phase mixing and the distortion is characterized by wave power spreading to larger perpendicular wavenumbers (k-perp), or equivalently to smaller perpendicular wavelengths. Some phase mixing experiments were also performed on LAPD. These results show that phase mixing does drive energy to larger k-perp, that the spreading of the k-perp spectrum grows with distance from the antenna, and that this evolution is faster for larger gradients. It will also be determined whether the initial k-perp spectrum influences the phase mixing rate in the same way as the sharpness of the transverse Alfven speed gradient, as is predicted by theory.