Behavioral Genetics in Pigs and Relations to Welfare - Genetics and the Behavior of Domestic Animals - page 412

Half of the crushed piglets are probably healthy piglets, but to identify
the cause of death accurately on a large scale, e.g., in nucleus herds, is very
difficult. The heritability for percentage of crushed piglets was estimated at
0.06 on data from a research herd ( Grandinson et al., 2002 ) and at 0.04 on
data from nucleus herds ( Grandinson et al., 2003 ). The sow's reaction to the
sound of a screaming piglet has been studied by several researchers. The
postural reaction of the sow is heritable ( Grandinson et al., 2003 ). Using a
questionnaire to the farmers, Vangen et al. (2005) estimated the heritability
of sow carefulness in two populations at 0.1 and 0.2.
The sow reaction to a screaming piglet is genetically correlated to piglet
mortality; stronger reaction 0.2; Grandinson et al.,

2003 ). But Hellbr¨gge et al. (2008) found a stronger correlation between pig-
let mortality and sows' reaction to pop music (r g 52

lower mortality (r g 52

0.3) than to piglet
scream. Our interpretation is that a test of this kind reveals the sow's mater-
nal ability, regardless of whether it believes the sound comes from its own
piglets or not.
The relationship between a sow and its piglets seem to rely upon piglets'
bond to their teats rather than strong sow piglet bonds. Even so, there is a

variation in sows' reactions to being separated from their piglets. Hellbr ¨ gge
et al. (2008) estimated the heritability for this trait at 0.1 both at farrowing
and three weeks later. The reaction was stronger at three weeks and the
genetic correlation to piglet mortality was rather high (r g 52 0.4). G¨de

et al. (2008) evaluated maternal ability 3 5 days after farrowing on a scale

from 1 to 5 by combining several aspects: behavior during farrowing,
behavior during lying down (carefulness), nursing behavior (position during
nursing), reaction to screaming piglets, and number of crushed piglets. They
estimated the heritability for this combined trait at 0.05. Canario et al.
(2007) have observed breed differences in nursing behavior between Large
White and Meishan sows, and Vangen et al. (2005) found some genetic vari-
ance in nervousness during suckling and willingness to expose the udder.
Many sows are kept in crates during farrowing and lactation, but in some
European countries farrowing crates are forbidden due to welfare reasons.
Furthermore, sows in outdoor production are never in crates. A frequently
asked question is whether there are important G E for maternal behavior in

systems with crates, farrowing pens without crates, or huts outdoors. Baxter
et al. (2011) compared two indoor sow lines, one selected for high survival
from birth to weaning, and a control line selected for average piglet survival.
All studied animals were born and reared on outdoor commercial units. They
were later compared in two environments, indoors in pens without crates and
outdoors. In both environments, the control gilts crushed more piglets than
the high-survival gilts. Thus, selection for improved piglet survival limits the
crushing behavior regardless of environment. Indoors, the high-survival gilts
did, however, show more aggression towards the piglets and savaged more
of them as compared to the control gilts. No such difference was observed

3