Genetic variation of sitting frequency and duration in pigs

McGlone, J.J., Akins, C.K. and Green, R.D., 1991. Genetic variation of sitting frequency and duration in pigs. Appl. Anita. Behav. Sci., 30:319-322. Prior to weaning, many piglets die from crushing, chilling, starvation, diarrhea and weakness. Piglets often die from crushing by the sow when she lies down after sitting. The objective of this study was to determine the amount of heritable genetic variation of sitting behavior. A total of 312 pigs from 62 litters were given ad libitum feed and water. Frequency and occurrence of sitting were summarized from 24-h video records. Heritability estimates (h 2) were calculated for these traits using a full-sib analysis. The h 2 for the occurrence of sitting was 0.41 + 0.14 and for frequency of sitting was 0.43 + 0.14. Genetic correlation between frequency and occurrence of sitting was 0.84 and the phenotypic correlation was 0.68 (P< 0.01 ). Sitting behavior was moderately to highly heritable indicating possible favorable selection against occurrence of this trait. Genetic selection against sow sitting could potentially decrease mortality rates among piglets, increase production and decrease costs for pork producers. I N T R O D U C T I O N No doubt the causes of piglet crushing are many and complex. Many pigs die before weaning from crushing, chilling, starvation, diarrhea and weakness. While behavioral studies will probably not completely solve the problem of piglet crushing, there is a need for greater attention to be given to sow behaviors which may impact rate of piglet crushing. Edwards et al. ( 1986 ) found that 46% of piglet deaths were due to crushing. In addition, 30% of these crushing deaths occurred when the sow lay down after sitting, 22% of the pigs were crushed when the sow lay down after standing and only 4% were crushed by the sow rolling over. McGlone and MorrowTesch (1990) , while researching behavior of sows in farrowing environments, observed that the time sows spent in a sitting position was positively correlated with the number of piglets crushed (r=0.54; P = 0 . 0 1 ). Furthermore, the rate of piglet crushing was 1.0_+0.36 piglets crushed per sow for 0168-1591/91/$03.50 © 1 9 9 1 Elsevier Science Publishers B.V. 320 J.J. MCGLONE ET AL. sows that showed sitting behavior and was only 0.29_+ 0.12 piglets crushed per sow for sows that showed no sitting behavior. In addition, Hansen and Curtis ( 1981 ) found that those sows which had crushed piglets spent more time sitting during the first 48-h period before the first piglet was born than sows which did not crush piglets. These studies have attributed sitting behavior to the type of environment the sows were in and no attempts have been made to determine what role genetics may play in pig-sitting behavior. Therefore, the objective of this study was to determine the level of heritable variation of sitting in the domestic pig. ANIMALS, MATERIALS AND METHODS Finishing pigs were housed in a mechanically ventilated building with 9 or 10 pigs per pen. Pigs were fed and watered ad libitum from a three-hole feeder and a nipple waterer. Diets were nutritionally balanced sorghum-soybean meal containing 14% crude protein. Dim lights remained on for 24 h day1 (which is normal practice in this building). The pigs were derived from a 4-breed cross of Duroc, Landrace, Yorkshire and Hampshire or terminal offspring from a commercially produced crossbred line. Pigs were weighed between 40 and I00 kg and were between 4 and 6 months of age during the period of data collection. Pigs were videotaped (3 frames s-1 recording speed) for sitting over a 24-h period. A total of 7488 pig-hours of video tape were viewed. Frequency of sitting (number of times each pig sat ) and the occurrence of sitting (scored as 0 or 1 ) were recorded from replayed tapes. STATISTICAL ANALYSIS Data represented 312 pigs from 30 pens, 62 sows, 15 sires and 8 blocks (over time ). The statistical model was as follows: Yijklm = It"lB~ + S / B i j -.b D / g / B i j k -b Sexl + %ktm where: # = overall mean; B~ = the fixed effect of the ith block; S/B~j= the random effect of the j th sire within the ith block; D/S/Bljk= the random effect of the kth dam of the j th sire within the ith block; Sext= the fixed effect of the /th sex (male or female) of pig; eijklm = residual error. Heritability estimates (h 2) were calculated for sitting frequency and occurrence of sitting, using a full-sib analysis (Harvey's ( 1987 ) least squares analysis) as follows 2 ( g ~ + a ~ ) + + where gs = sire variance, ~D = dam variance and, gz = error variance. GENETIC VARIATION OF SITTING IN PIGS