Non-offspring nursing (allosuckling) is costly for lactating females. It is reasonable to assume, therefore, that mechanisms have evolved to control the frequency of allosuckling. It is suggested that the synchronization of nursing between females could be such a mechanism. The aim of this study was to investigate whether nursing synchronization is related to the timing of nursing behaviour of paired sows and the behaviour of alien piglets in relation to allosuckling. Specifically, we predicted that: (i) nursing synchronization will increase if there is a real risk of allosuckling compared with a situation where there is none, (ii) nursing synchronization indeed reduces allosuckling and, therefore, those nursings with allosuckling attempts will be less closely synchronized than those without allosuckling attempts and (iii) in non-synchronized nursings, the sows will more often employ another strategy to reduce allosuckling, namely, they will make the nursings non-nutritive by not releasing milk. Ten pairs of sows and their litters were used in this study. For each pair, the two sows were housed individually until their litters were weaned; however, their litters had access to both sows from day 11 post-partum. The behaviour of the 10 pairs of sows with their litters was videotaped for 6 h on days 10, 11, 17 and 24 postpartum. The frequency of allosuckling attempts, non-nutritive nursings and allosuckling during milk ejection was analysed for every pair for every day of observation. The observed frequency of nursing synchronization was high. The median interval between the start of a nursing within one pair of sows was 32 s. The first prediction was supported as synchronization became significantly tighter when piglets were given the possibility to allosuckle on day 11 [Cochran-MantelHaenszel (CMH) statistics, p < 0.0001]. The second prediction was also supported; the nursings were less closely synchronized when there were allosuckling attempts (CMH statistics, p < 0.05). Of all nursings, 36% had an allosuckling attempt; however, only 17% resulted in real allosuckling. Evidence in favour of the third prediction was also found as non-nutritive nursings were more frequent in non-synchronized nursings (CMH statistics, p < 0.05). Finally, real Ethology 111, 652-668 (2005) Ó 2005 Blackwell Verlag, Berlin allosuckling during milk ejection was unrelated to the degree of nursing synchronization, as in less synchronized nursings, the sows responded to allosuckling attempts by not releasing milk. Our results suggest that sows are not able to completely suppress allosuckling; however, they can reduce its frequency through nursing synchronization and non-nutritive nursings.
