Agriculture Reference
In-Depth Information
(Bressers
et al.
,
1993; Hunter
et al.
,
1988) and/or lying patterns (Moore
et al.
, 1993). In
contrast, stalled sows have been observed to maintain a degree of aggressiveness towards
their neighbours for prolonged periods of time, possibly because they are physically
prevented from resolving any dispute satisfactorily (Barnett
et al.
,
1987; Vestergaard and
Hansen, 1984).
Rank order fights are initiated when animals do not recognise each other or when they
dispute each other's position in the hierarchy. Stookey and Gonyou (1998) looked at the
patterns of aggression and proposed that recognition among young piglets appears to
be based on familiarity gained when reared or housed together, and does not involve
genetic relatedness. To reduce aggression, be it over social status or over access to a
resource, animals tend to develop a so called 'avoidance order' (Jensen, 1982; Jensen
and Wood-Gush, 1984). This order helps to minimise aggression and is closely linked
to the dominance hierarchy formed. Sows recognise each other after several weeks of
separation. Spoolder
et al.
(1996) looked at pairs of gilts introduced at intervals of two
weeks into a large dynamic group. Pairs from an original subgroup of six recognised each
other without fail, and had significantly less aggression among each other than aggression
towards other gilts introduced in pairs during the same period. The largest separation
time tested in their study was 4 weeks.
It seems that the stability of the hierarchy improves as animals become older. Parent
et
al.
(2012) tested the stability of the hierarchy during the growing phase and during the
pregnancy phase and found that the temporal stability was higher for pregnant sows than
for growing females. They suggested that the hierarchy of females becomes more stable
over time during pregnancy, when compared with that of young gilts during the growing
period (Parent
et al.
, 2012).
Competition over resources
Competition over resources such as space is regulated via lying patterns. Lying patterns
change over time, as pregnancy progresses, and generally involve migration of lying sows
towards more preferred lying areas (relatively quiet and comfortable areas) (Spoolder,
1998). This is particularly noticeable in large groups. At the same time, assessments of
lying patterns show that proximity to other sows changes during pregnancy as well:
the inter-individual distance to unknown animals decreases (Spoolder
et al.
, 1996,
1998). This was more recently confirmed by Krauss and Hoy (2011). They assessed the
preference of sows to lie nearest to group members they were reared with vs. sows they
had not met before being moved into the group-housing system. They calculated the
percent of occasions where the nearest neighbour was a familiar animal on the day after
introduction as well as three weeks later, and found advancing integration of new sows
into the dynamic group (new sows: 1
st
day 94.3%, 21
st
day 46.0%; resident sows: 1
st
day
96.8%, 21
st
day 74.6%). Just as in the study by Spoolder (1998), they could not conclude
whether this preference was an expression of territorial behaviour or whether it showed
a definite preference for an individual pen mate. However, since this effect significantly
declined within 3 weeks after the introduction of new sows without any changes being
