The development of new interstitial oxide conductors with less restrictive structural constraintsrequires understanding of the mechanisms allowing both incorporation and mobility of the excessoxide. We have recently shown that two-dimensionally connected tetrahedral gallium oxidenetwork in the melilite structure La1+xSr1-xGa3O7+x/2 stabilises oxygen interstitials by localrelaxation around them, affording oxide ion conductivities comparable to apatite systems, and more recently that the Ca and Ba systems do also. The further development of these materials requires a better understanding of the actual local environment of the oxide interstitial charge carrier. In order to do this we plan to exploit the PDF method in conjunction with our ongoing conventional crystallographic studies, to better understand these systems.