The frequency of stereotypies, the nature of social interactions and whether sows or piglets initiate and terminate successful sucklings were compared. Sows on a commercial farm were housed in either crates or pens and studied for a year. Our observations indicated that sows in crates developed significantly more stereotypies (14.9 per sow-hour) than sows in pens (0.3 per sow-hour). Vacuum chewing was the most prevalent stereotypy observed (69%). Seventy-seven percent of the stereotypies were recorded before feedings. Frequency scanning resulted in fewer stereotypies per sow-hour than direct continuous sampling, with the exception of dog-sitting position. This finding suggests that frequency scanning is a superior method for sampling stereotyped behaviour than direct continuous sampling. Sows in crates had significantly more aggressive interactions (0.8 per sow-hour) than sows in pens (0.3 per sow-hour). Housing design did not influence the proportion of successful sucklings. Sows initiated significantly more often (77%) successful sucklings than piglets (23%). Our observations confirm that providing straw and space for exercise reduced the frequency of stereotypies (14.9 per sow-hour), but did not completely suppress them (0.3 per sow-hour). The results obtained confirmed previous reports indicating that housing design influences the frequency of stereotypies and the nature of social interactions between sows.
Bibliographical noteFunding Information:
The authors wish to thank Ann Langer, Leilani Hotaling and Maria Rosario Guerrero for their assistance in collecting data, and Dr Vickie King for her assistance in the statistical analysis. This investigation was made possible by grants from the United States Department of Agriculture, Low Input Sustainable Program (LNC 88-20) and by Consejo Nacional de Ciencia y Tecnolo-gia, M6xico.