Elements of husbandry procedures, such as handling, may disrupt rodent social behaviour. Such effects may be contingent upon the familiarity between individuals and upon the quality and quantity of the disruption. We investigated this issue using laboratory rats. We placed 36 rats into groups of three. At the point of group formation, and at 24 h, 7 days and two weeks afterwards, individuals received one of three treatments: ‘handling’, exposure to novel conspecific ‘urine’, or ‘control’ (undisturbed), for a duration of either 5 or 15 mins. We used a social recognition test to measure the ability of the rats to recognise the urine of group members of increasing familiarity following the implementation of these treatments. The ‘control’ treatment did not appear to disrupt social recognition. The 5 min ‘urine’ treatment appeared to disrupt recognition only when the rats had received the briefest experience of the ‘familiar’ urine (5 mins). The 5 min ‘handling’ treatment, however, appeared far more disruptive, with an apparent disruption of social recognition even when familiarity with the urine donor was high (eg 7 days of group housing). Both the ‘handling’ and ‘urine’ treatments appeared more disruptive when presented for an increased duration (15 mins). There was also some evidence that increased experience of the handling procedure might reduce its disruptive effect. The results of this study have several implications for the welfare of laboratory-housed rats, and these are discussed.