Gestation group housing of sows - The Gestating and Lactating Sow - page 45

feeding stalls provide the best protection, followed by troughs with decreasing lengths of
barriers to separate feeding places (Andersen et al. , 1999).

For sequential feeding systems, it is even more important that the animals which are eating
cannot be displaced by older sows. This stealing behaviour is to a large extent 'learned':
every time a return visit or challenge to a feeding sow is rewarded with additional feed,
the behaviour is reinforced. Breaking that learning curve is of crucial importance to all
sequential feeding systems. Gilts should be taught how to use an electronic sow feeder
in the absence of older, more dominant sows. In general, they can be trained to use the
system reliably within 1-2 weeks (Nielsen, 2008; Spoolder, 1998).

Feeding level

Pregnant sows are generally fed a restricted amount of feed, aimed at optimising their
body condition and reproductive performance. Nevertheless, in some systems over-
feeding may occur, e.g. in systems with either ad libitum feeding or systems in which
dominant sows can eat a large share of the distributed feed. Sows that eat more than their
physiological need will gain more weight and more backfat than required, although this
does not seem to affect short term reproductive performance (farrowing rate, litter size in
the ongoing pregnancy and subsequent lactation performance). However, over-feeding in
early pregnancy increases embryo mortality in gilts (reviewed by Foxcroft, 1997).

Van der Peet-Schwering et al. (2004) studied the effects on reproductive performance of
feeding group-housed pregnant sows an ad libitum diet with a high level of fermentable
non-starch polysaccharides (in comparison with a restricted conventional diet). They
followed the animals over 3 reproduction cycles. During pregnancy, the ad libitum
fed sows ate 1.3 kg/d more than the restrictedly fed sows and also gained more body
weight and backfat. Feed intake during lactation and reproductive performance were
not negatively influenced by the ad libitum feeding (Van der Peet-Schwering et al. , 2004).
Stewart et al. (2010) also looked at the effects of high fibre diets on behaviour of sows
and compared a diet containing 15% crude fibre to a control diet containing 5% crude
fibre. Sows on the high fibre treatment spent more time resting in the kennel areas, less
time performing stereotypic behaviours and showed a reduction in some aggressive
behaviours relative to sows fed the control diet. Peltoniemi et al. (2010) investigated the
effects of an ad libitum fed diet (7.7 MJ/kg) with a control diet (9.3 MJ/d) on reproductive
performance (pregnancy rate, weaning to oestrus interval, piglets born alive, stillborn
piglets and progesterone concentrations). They found no effect on any of these variables
but reported that control sows weaned significantly more piglets (9.7 vs. 9.4), whereas
piglets from ad libitum fed sows were heavier at weaning (8.8 vs. 8.0 kg).

3.2.4 Housing design

Functional areas: lying area, feeding area, dunging area

Sow housing facilities should be designed with respect to the needs of the animals for
lying, feeding and dunging. Pigs will try to keep these areas separate themselves, by