Invertebrate neuroscience has provided a number of very informative model systems that have been extensively utilized in order to define the neurobiological bases of animal behaviours (Sattelle and Buckingham in Invert Neurosci 6:1-3, 2006). Most eminent among these are a number of molluscs, including Aplysia californica, Lymnaea stagnalis and Helix aspersa, crustacean systems such as the crab stomatogastric ganglion and a wide-range of other arthropods. All of these have been elegantly exploited to shed light on the very important phenomenon of the molecular and cellular basis for synaptic regulation that underpins behavioural plasticity. Key to the successful use of these systems has been the ability to study well-defined, relatively simple neuronal circuits that direct and regulate a quantifiable animal behaviour. Here we describe the pharyngeal system of the nematode C. elegans and its utility as a model for defining the genetic basis of behaviour. The circuitry of the nervous system in this animal is uniquely well-defined. Furthermore, the feeding behaviour of the worm is controlled by the activity of the pharynx and this in turn is regulated in a context-dependent manner by a simple nervous system that integrates external signals, e.g. presence or absence of food, and internal signals, e.g. the nutritional status of the animal to direct an appropriate response. The genetics of C. elegans is being effectively exploited to provide novel insight into genes that function to regulate the neuronal network that controls the pharynx. Here we summarise the progress to date and highlight topics for future research. Two main themes emerge. First, although the anatomy of the pharyngeal system is very well-defined, there is a much poorer understanding of its neurochemistry. Second, it is evident that the neurochemistry is remarkably complex for such a simple circuit/behaviour. This suggests that the pharyngeal activity may be subject to exquisitely precise regulation depending on the animal's environment and status. This therefore provides a very tractable genetic model to investigate neural mechanisms for signal integration and synaptic plasticity in a well-defined neuronal network that directs a quantifiable behaviour, feeding.