Over the past few years, various new feeding systems for fattening pigs have been developed. One of these systems is sensor-controlled feeding, whereby the animals are fed with liquid feed. Liquid feeding systems are different from dry feeding systems in many respects. Larger volumes of feed are consumed and a feeling of satiation may therefore be reached sooner in the individual pig. On the other hand, satiation probably lasts for a shorter time and feed must be provided several times during the day. As the water content in the feed is high, it takes the pigs longer to consume the same amount of digestible energy in the form of liquid feed compared to dry feed. With dry feeding systems for fatteners, it has been observed that a restricted a animal/feeding-place ratio (AFR) may lead to competition for feed, and consequently to increased aggression at the feeding trough, reduced duration of feeding and lower weight gain with more variation in weight between individuals (Hansen et al., 1982; Botermans and Georgsson, 2001; Kircher, 2001; Turner et al., 2002). As a result of increased competition at feedings, some pigs, especially the smaller ones, may have difficulty gaining access to the trough, and thus feed less (Botermans and Georgsson, 2001). These individuals may tend to be of lower social rank. In other studies on pigs, it has been found that there is a correlation between weight and social rank within a group (Rushen, 1988; Hicks et al., 1998; Andersen et al., 2000; O’Conell et al., 2004). The aim of the present study was to quantify the effect of three different AFRs (4:1, 7:1 and 13:1) with nine, six and three feeding places per group, respectively, on the behaviour and performance of fattening pigs (25–100 kg) housed in groups of 40 individuals. Each AFR was tested with seven groups and with each group, four pigs were selected from each weight class (lightweight, middleweight and heavyweight) as focal animals. The behaviour in the feeding area was observed by means of 24 h-video-recording for the 12 focal pigs at an age of 14 and 17 weeks. Daily weight gain was calculated for all animals. As the AFR increased, feeding bout duration at the trough declined significantly, whilst the duration of waiting increased. The AFR had no significant effect on the frequency of aggressive displacements at the feeding trough. On the other hand, the fatteners were more frequently pushed away from the trough without aggressive interactions as the AFR increased. Besides the AFR, the age and weight class of the focal pigs had a significant effect on behaviour. Lightweight animals waited longer for access to the trough and were more often displaced from the trough as a consequence of aggressive interactions. Weight gain was lower as AFR increased, an effect that was more pronounced in lightweight focal animals. In conclusion, the AFR has a decisive influence on the behaviour and performance of fattening pigs fed by sensor feeding. Lightweight pigs, most likely the subordinate group members, are more affected by an increased AFR, and hence most at risk of impaired welfare. An AFR of 13:1 has marked negative effects on the pigs’ performance and behaviour and is inadvisable to be used for this feeding system.