Mobile ad hoc networks are noncentralised, multihop, wireless networks that lack a common infrastructure and hence require self-organisation. Their infrastructureless and dynamic nature entails the implementation of a new set of networking technologies in order to provide efficient end-to-end communication according to the principles of the standard TCP/IP suite. Routing, IP address autoconfiguration and Web service discovery are among the most challenging tasks in the ad hoc network domain. Swarm intelligence is a relatively new approach to problem solving that takes inspiration from the social behaviours of insects, such as ants and bees. Self-organization, decentralization, adaptivity, robustness, and scalability make swarm intelligence a successful design paradigm for the above-mentioned problems. In this paper we propose BeeAdHocServiceDiscovery, a new service discovery algorithm based on the bee metaphor, which also takes into account quality metrics estimates. The protocol has been specifically designed to work in mobile ad hoc network scenarios operating with Beeadhoc, a well-known routing algorithm inspired by nature. We present both the protocol strategy and the formal evaluation of the discovery overhead and route optimality metrics showing that BeeAdHocServiceDiscovery guarantees valuable performances even in large scale ad hoc wireless networks. Eventually, future research suggestions are sketched.