Stereotypic behaviour is generally associated with animals maintained in restrictive environments, and has rarely been described in wild or free-ranging animals. The difference between captive and wild populations may be due to their genetic predisposition or to experience of environmental factors. To investigate genetic and environmental factors, we compared the behaviour of 12 wild caught voles and their 9 pups with that of 12 laboratory reared voles and 14 laboratory bred pups. All voles were observed twice. Adults were observed after 10 days housing in a cage, containing food, water, sawdust and hay, and again after 60 days. Pups were observed in the same cages 10 and 60 days after weaning. For each observation, the voles' behaviour was recorded both undisturbed in this cage, and following introduction to an unfamiliar cage. Locomotor stereotypies were observed in laboratory adults, but not in wild caught voles, which spent less time on all locomotor activities and more time under cover than laboratory voles. There was no difference in mortality or fecundity of laboratory and wild caught voles, so there appeared to be no selective advantage to stereotyping. There was no difference in the behaviour of wild and laboratory pups, so early environmental experience of the cage environment, rather than parental background, was an important factor in the development of locomotor stereotypies in this species. Pups that developed stereotypies by 60 days spent less time under cover and more time walking and climbing after 10 days than voles that did not develop stereo typic behaviour. Stereotypic behaviour may therefore have been derived from persistence of locomotor behaviour. Wild caught voles may have failed to develop locomotor stereotypies, either because they did not perform a locomotor response to captivity or because older voles are less prone to develop novel responses to external cues. It would, therefore, be dangerous to use the absence of stereotypic behaviour as a reliable indicator of welfare without taking into account the animal's prior experience.