We consider the problem of providing a rigorous model for programming wireless sensor networks. Assuming that collisions, packet losses, and errors are dealt with at the lower layers of the protocol stack, we propose a Calculus for Sensor Networks (CSN) that captures the main abstractions for programming applications for this class of devices. Besides providing the syntax and semantics for the calculus, we show its expressiveness by providing implementations for several examples of typical operations on sensor networks. Also included is a detailed discussion of possible extensions to CSN that enable the modeling of other important features of these networks such as sensor state, sampling strategies, and network security.