Finishing pig performance when using single gender grouping and phase feeding, March 2013

Date published: 01 March 2013

Phase feeding is commonly used throughout Europe. However, the majority of European pig production is based on castrates whilst entire male pigs are used in the UK and Ireland.


The UK and Irish pig industries are hesitant about adopting phase feeding due to the unknown effect on pig performance, especially of entire boars.  However, phase feeding should reduce diet costs and the excretion of nitrogen and ammonia.  This study aimed to compare the performance of gilts and boars when penned in single or mixed gender groups and offered either a single diet throughout the finishing stage or a two-phase dietary regime.  A total of 480 pigs were penned in groups of 10 and six treatments were compared (in a 2 x 3 factorial design).  Pens of pigs represented 1) all boars, 2) all gilts and 3) 50:50 mixture of gilts and boars.  These pens of pigs were offered either a single diet between 12 weeks of age (44 kg) and 120 kg (crude protein (CP) 18%, lysine 0.95%) or a two-phase dietary regime where the aforementioned diet was offered between 44 and 80 kg and a diet containing 16.7% CP and 0.8% lysine was offered between 80 and 120 kg.  The performance of pigs was similar when either of the two dietary regimes were offered (P>0.05).  Pig growth rate averaged 905 g/day and feed conversion efficiency averaged 2.56 between 44 and 120 kg across the two dietary regimes.  

There was no overall effect (P>0.05) of dietary regime on carcass performance either.  The 18% CP, 0.95% lysine diet cost £6 per tonne more than the 16.7% CP, 0.8% lysine diet.  Due to the lack of difference in the performance of pigs but differences in feed cost and nitrogen content of the diets, nitrogen excretion was reduced by approximately 8% and finisher diet costs were approximately £1.00 per pig lower when the two-phase dietary regime was offered compared with when the single diet was offered throughout the finishing period.  With regard to the effect of pig gender, the overall performance of gilts was poorer (ADG 865 g/day, FCR 2.70; P<0.01 and P<0.001 respectively) than that of boars (942 g/day, 2.46) between 44 and 120 kg.  Although the kill out percentage of gilts was better than boars (77.7 vs 76.1%, P<0.01), their backfat depth at P2 (probe) was 0.9 mm greater (12.4 vs 11.5 mm).  The performance of the ‘mixed’ gender pens was intermediate to that of gilts and boars during the finishing period which was expected but the backfat depth of pigs in the mixed gender groups was higher than that of boars or gilts in the single gender groups (13.0 mm).  When examining the effect of grouping on the performance of gilts and boars separately, gilts and boars performed as well in single gender groups as in the mixed gender groups.  However, when the boars were in mixed gender groups their backfat depth (probe) was greater than when they were in single gender groups (13.7 vs 11.7 mm, P<0.001).

In conclusion, the phase feeding regime used in this study did not significantly affect the performance of boars or gilts and as a result nitrogen excretion was reduced by approximately 8% and feed costs by approximately £1 per pig compared to when a single diet was offered throughout finish.  These findings are in agreement with the results of other studies.  In other studies, where differences in the response of gilts and boars were found due to lysine level, the levels used were more extreme than those used in the current study.  The overall performance of boars (average daily gain and feed conversion ratio) was better than gilts so the main advantage of split gender grouping would be the ability to market pigs more efficiently, e.g. the more efficient boars could be kept to heavier slaughter weights with less likelihood of becoming excessively fat.  However, the backfat depth of boars in mixed gender pens was found to be greater than that of boars in single gender pens which could have implications on grading, especially when pigs are taken to heavy slaughter weights.  However, this finding requires further investigation.