Since their covariant description by Israel (clarifying earlier treatments by e.g. Lanczos and Darmois) [1], thin shells have been intensively used as a model for matter in general relativity. However a question is rarely asked when considering a four dimensional spacetime, to wit whether or not matter on the shell obeys the three dimensional Einstein equations, as there is no (experimental) reason why it should. The situation changed recently with the increasing interest in gravity theories within spacetimes with large extra dimensions and the advent of the idea that our universe may be a four dimensional singular hypersurface, or “brane”, in a five dimensional spacetime, or “bulk” [2]. Indeed, in this new context, it becomes crucial to recover Einstein gravity for realistic matter on the brane, at least to some approximation compatible with the present experiments.