Effects of social "stressors" on belly-nosing behaviour in early-weaned piglets: is belly-nosing an indicator of stress?

Belly-nosing is a behaviour pattern commonly observed in early-weaned pigle ts, Although, belly-nosing has been referred to as an indicator of stress, the factors involved in its causation are unknown. The objective of this st udy was to determine if the differences in stress at weaning influence the development of belly-nosing behaviour in piglets weaned at 12-14 days of ag e. To test this we attempted to create varying degrees of stress at weaning by subjecting piglets to different combinations of social stressors. The e xperimental design was a 2 x 2 factorial with group composition (mixed litt ers versus littermates) and density (0.15 m(2) per pig (HD) versus 0.4 m(2) per pig (LD)) as the main factors. Six replicates per treatment, of six pi glets/replicate were used in three, 3-week trials (n = 144). Behaviour was recorded every 5 min during one 4 h period on day 1 and during two 4 h peri ods on days 3, 7, 10, 14, 17 and 21 post-weaning. Feed intake, water intake and growth rate were determined weekly. Blood was collected, from half of the piglets, on day 2 pre-weaning and days 3 and 10 post-weaning for neutro phil: lymphocyte ratios (N:L) and plasma cortisol concentrations. During th e first week post-weaning, feed intake was higher for piglets housed at HD (P < 0.05) but no other differences in performance were observed. Contrary to what was expected, aggression was greater for piglets housed at LD than those housed at HD overall, and on days 10 (P < 0.05) and 17 (P < 0.05) pos t-weaning. On day 3 post-weaning, variation in N:L was greater for piglets weaned with Littermates (P < 0.05) and piglets housed at LD (P < 0.05). On day 10 post-weaning, both mean plasma cortisol (P < 0.05) and variation in N:L (P < 0.05) were higher for piglets housed at LD. Although plasma cortis ol, N:L and aggression were all suggestive of a greater stress response in piglets housed at LD, there were no differences in belly-nosing for any tre atment (P > 0.10). Belly-nosing did develop across all treatments by day 7 post-weaning. Belly-nosing does not appear to be a general behavioural indi cator of stress. (C) 2001 Elsevier Science B.V. All rights reserved.