Group and individual sow behavior is altered in early gestation by space allowance in the days immediately following grouping

Aggression between domestic sows is greatest when sows are first introduced to each other and hierarchies form. The aim of this study was to determine the effect of a spacious "mixing pen" on sow aggression and stress. Sows were mixed into groups of 6 and allowed 2 (LOW; 8 groups and 48 sows), 4 (MED; 7 groups and 42 sows), or 6 m(2)/sow (HIGH; 7 groups and 42 sows) for 4 d after mixing, at which point all pens were equalized to 2 m(2)/sow. Salivary cortisol concentration and injury counts were measured on d -1, 0, 1, 3, and 4 relative to mixing, and behavior was also recorded on each of these days following mixing. Reproductive performance was assessed at farrowing. A linear mixed model was applied to the data. Data are presented as least squares means and standard error of the mean. Where transformations occurred, nontransformed adjusted means are presented in parentheses following the presentation of transformed data. In the primary analyses where measures were considered at the pen level, there were no effect of space allowance on fight number per sow, duration of fights, percentage of total time spent fighting, displacements, bites, knocks, and lunges (P > 0.05). These measures were higher on d 0 (i.e., fight number 1.0 +/- 0.1 [13.8]) compared with d 1 (0.4 +/- 0.1 [4.2]), 3 (0.7 +/- 0.1 [5.3]), and 4 (0.7 +/- 0.1 [5.5]; P < 0.05), with no increase in aggression on d 4 when pen sizes were standardized (P > 0.05). There was increased percentage of time spent active (1.5 +/- 0.02 [33.7] for LOW, 1.5 +/- 0.02 [36.5] for MED, and 1.6 +/- 0.02 [43.4] for HIGH) and time spent exploring (1.8 +/- 0.1 [3.5] for LOW, 2.0 +/- 0.1 [4.0] for MED, and 2.3 +/- 0.1 [5.7] for HIGH) and number of nonaggressive sow-sow contacts (0.3 +/- 0.09 [2.2] for LOW, 0.4 +/- 0.07 [3.2] for MED, and 0.5 +/- 0.07 [4.5] for HIGH) in HIGH compared with LOW (P < 0.05). Farrowing rate and total piglets born were not affected by treatment (P > 0.05). A secondary analysis was conducted that examin ed individual sow behavior within each pen, and this identified increased injury number in the lowest ranked sows (involved in no fights on d 0 and no displacements on d0 to d4) in LOW (9.3 +/- 1.2 [107.9] for LOW, 6.2 +/- 0.8 [53.0] for MED, and 5.1 +/- 0.8 [28.1] for HIGH) and also decreased fight number and duration in HIGH compared with LOW on d 0 and 1 (P < 0.05). Our primary data analysis demonstrates positive exploratory and social behaviors with increased space and suggests that a reduction in space following hierarchy formation is not a significant stressor. Additionally, there is some evidence at an individual sow level that increased space at mixing benefits sow welfare parameters, especially for lowranked sows.