In this paper we present a semantics for distributed object, extending Abadi and Cardelli's object calculus and Gordon and Hankin's concurrent object calculus. This paper uses techniques form distributed calculi such as Sewell's distributed \pi, Hennessey and Riely's D\pi and Yoshida and Hennessy's D\lambda\pi. We show the importance of serializable types and object registries in examples of object code. The main results are subject reduction results, showing that the resulting language is type-safe even in the presence of intruders.