Usually, the content of the theory of special relativity is reduced to the description of uniform motion and the corresponding Lorentz Transformations. The pure geometrical interpretation of the theory is attributed to the theory of general relativity with its assumptions concerning general coordinate transformations. That this is not a correct opinion will be shown. Both theories are theories of a space-time world and admit arbitrary permissible coordinate transformations. In a sense, "general relativity" is not a good name for the theory, the main meaning of which is the interpretation of gravitation as a curvature of space-time. The difference between the two theories concerns the space-time structure: the flat, pseudo-Euclidian (topologically simple) manifold of special relativity and the curved, Riemannian (topologically nontrivial) manifold of the general relativity. The gravitation is the curvature and, as a result of its tensor nature, we cannot "turn it off" in some finite region by means of a coordinate transformation. Therefor, the equivalence principle has only a local and heuristic meaning. An observer can always distinguish gravitation from acceleration. It should be emphasized that the acceleration is not allied to the curvature of space-time and, as a result, it can be described in the framework of the special relativity.