Providing an integrated access to multiple heterogeneous sourcesis a challenging issue in global information systems for cooperation and interoperability. In the past, companies haveequipped themselves with data storing systems building upinformative systems containing data that are related one another,but which are often redundant, not homogeneous and not alwayssemantically consistent. Moreover, to meet the requirements ofglobal, Internet-based information systems, it is important thatthe tools developed for supporting these activities aresemi-automatic and scalable as much as possible.To face the issues related to scalability in the large-scale, in this paper we propose the exploitation of mobile agents in the information integration area, and, in particular, their integration in the Momis infrastructure. MOMIS (Mediator EnvirOnment for Multiple Information Sources) is a system that has been conceived as a pool of tools to provide an integrated access to heterogeneous information stored in traditional databases (for example relational, object oriented databases) or in file systems, as well as in semi-structured data sources (XML-file).This proposal has been implemented within the MIKS (Mediator agent for Integration of Knowledge Sources) system and it is completely described in this paper.