J. Anim Sci. 2006. 84:2847-2855. doi:10.2527/jas.2005-606
© 2006 American Society of Animal Science
Socializing piglets before weaning: Effects on behavior of lactating sows, pre- and postweaning behavior, and performance of piglets
E. F. Hessel*,1,
K. Reiners
and
H. F A. Van den Weghe
* Institute of Agricultural Engineering, University of Goettingen, D-37075 Goettingen, Germany;
and
Research Center for Animal Production and Technology, University of Goettingen, D-49377 Vechta, Germany
 |
Abstract
|
|---|
This study evaluated how socializing piglets before weaning affects behavior of lactating sows and the pre- and postweaning behavior and performance of piglets. Two farrowing rooms, each with 6 pens, and 1 nursery with 4 pens were used. In total, data were obtained from 24 sows and their litters. In each farrowing room, the solid barriers between 3 farrowing pens were removed on d 12 after farrowing, and the sows remained confined in their crates (experimental group). In the other 3 farrowing pens of each farrowing room, sows and their litters were kept under conventional conditions until weaning (control group). All piglets were weaned 28 d after birth. After weaning, piglets from each group remained together in 1 pen of the nursery. The behavior of sows (lying, standing, sitting, nursing) and piglets (lying, active, suckling) in the farrowing rooms was observed for 24 h before and for 48 h after removal of the barriers between the pens. In addition, behavior (active, lying, feeding, agonistic behavior) of piglets was observed in the nursery during the initial 48-h period after weaning. Each piglet was weighed on d 5, 12, and 28 after birth and thereafter weekly until the fifth week of rearing. In the farrowing room, mixing of litters did not influence behavior of piglets and sows. Preweaning weight gain of the piglets did not differ (P = 0.60) between the treatments. In the initial 48 h after weaning, less agonistic behavior (P < 0.001) was observed in piglets belonging to the experimental group. During 5 wk of rearing, piglets in the experimental group gained more weight compared with the control group (P = 0.05). The advantage shown by the experimental group became especially conspicuous in the first week after weaning (P = 0.05). By socializing unfamiliar piglets before weaning, stress due to mixing could at least be distanced in time from the other burdens of weaning, thereby improving performance.
Key Words: performance • piglet • social behavior • weaning
|
INTRODUCTION
|
|---|
Weaning piglets is one of the most critical phases in modern pig production. The most important factors causing stress in this phase include separation from the sow, transport, changing nutrition, different pathogenic pressures from the new environment, as well as increased aggression as new dominance hierarchies are established following commingling of previously unfamiliar piglets (Held and Mendl, 2001
; Jensen, 2002; King and Pluske, 2003). Keeling and Jensen (2002) reported that mixing piglets from different litters is the main cause of aggression after weaning, which occurs during establishment of new hierarchies. Aggressive behavior changes with age of piglets. Jensen (1994) and Pitts et al. (2000) found that dominance fights are shorter in duration in younger piglets compared with older ones. According to Newberry et al. (2000), piglets that were familiarized with one another during the nursing period did not exhibit any dominance aggression postcommingling. Also, D’Eath (2005) reported that piglets that were commingled once during the nursing period established a hierarchy more quickly when they were remixed with new piglets after weaning.
Weight gain in piglets immediately after weaning may influence their growth until the end of the fattening period (Pluske et al., 2003). In order to optimize feed intake and growth immediately after weaning and thereby ensure high weight gain in the first few days of rearing, weaning stress should be reduced as much as possible (Brooks and Tsourgiannis, 2003).
With this in mind, this study focused on the effect of socializing litters during the nursing period on the behavior and performance of piglets before and after weaning.
|
MATERIALS AND METHODS
|
|---|
Experimental Design
For this investigation, 2 identical farrowing rooms, each with 6 pens, and 1 nursery with 4 pens at the University of Goettingen’s experimental station were used. The local animal care and use committee approved the design of the study. Rooms were mechanically ventilated, and pigs were housed without straw. Animals were moved into the rooms according to the all in/all out principle.
Within each farrowing room, 3 farrowing pens were located on both sides of a central feeding passage. The farrowing pens were equipped with fully slatted floors; each pen had a water-heated piglet resting area, and individual pens were separated by solid partitions that prevented intermingling of piglets. Approximately 1 wk before the calculated farrowing date, sows were moved into the farrowing crates, where they remained until weaning. On d 12 after the calculated farrowing date, the solid barriers between the 3 farrowing pens on 1 side of the room were removed in both rooms, and the sows remained confined in their crates (experimental group). Before removing the barriers, piglets from each litter were marked on their back with a commercial livestock-marking pen. These markings were renewed every second day. Piglets from the first litter were marked with a longitudinal bar, piglets from the second litter were marked with a transverse bar, and piglets from the third litter were unmarked. Sows and their litters from the other 3 farrowing pens of each room were kept under conventional conditions until weaning (control group). Weaning of all piglets took place at 4 wk after birth.
After weaning, piglets were moved into the nursery, and the 3 litters from each group remained together. The pens (2.9 x 4.0 m) of the nursery were identically equipped with a total floor space of 11.6 m2 split into a fully slatted activity area (8.4 m2) and a solid-floored, covered resting area (3.2 m2) that was equipped with zonal heating. All pens contained 1 wet and 1 dry feeder that were placed in the activity area, providing 11 feeding spaces in total. Both feeders were replenished daily at 0700.
Two replicates were carried out simultaneously, and the experiment was repeated twice, resulting in n = 4 replicates per treatment. In order to balance for any effects of position of the pens in the nursery on behavior and performance, pens in which piglets belonging to the experimental group were reared in the first rearing period were used for piglets belonging to the control group in the second rearing period.
Animals.
Twenty-four sows from 3 genotypes [German Large White x German Landrace; (German Large White x German Landrace) x Pietrain; German Landrace] that were all serviced by Pietrain boars were used in this study. Sows were assigned to the 2 treatments according to genotype and age. Sows in the experimental group farrowed on average 1.6 d (± 1.1) and sows in the control group 1.7 d (± 1.4) after the calculated farrowing date. Three days after the calculated farrowing date, some of the piglets were crossfostered to obtain balanced litter sizes, so that group sizes would be similar in the nursery period. In total, 10 piglets were crossfostered 1.3 (± 1.2) d after the actual farrowing in the experimental group and 12 piglets 1.0 (± 1.3) d after the actual farrowing in the control group. Thereafter, the litter size per sow averaged 10.2 piglets in the experimental group and 10.0 piglets in the control group.
Feeding.
In the farrowing room, in addition to sows’ milk, solid feed (14.4 MJ of ME/kg of DM, 17% CP) was offered to the piglets from a dry feeder from the seventh day after the calculated farrowing date. Piglets were fed this prestarter feed for an additional14 d after weaning, after which they received a feed with 13.4 MJ of ME/ kg of DM and 17.4% CP. During the experiment, the feed was available ad libitum.
Data Collection
Behavior.
A single observer performed all behavioral observations. The specific behaviors that were observed and analyzed are presented in Table 1. In the farrowing room, behavior of sows and piglets was continuously recorded on videotape and analyzed from 24 h before to 48 h after the removal of the barriers between the pens. The behavior of the piglets was scored on a group, rather than an individual, basis. A behavior was recorded as occurring when more than 75% of the piglets were engaged in that particular behavior. Four-minute scan samples of the videotapes were used to extract the behavioral data from the videotapes at a later date. Sow and piglet behavior was also recorded using 4-min scan samples by an observer in the farrowing room to evaluate whether commingling of piglets resulted in any cross-suckling behaviors. These observations occurred immediately after the removal of the barriers and 9 d later, 2 d prior to weaning, with each observation period lasting 1 h. Restlessness in the sows was recorded during this period also. The behavior of piglets in the nursery was recorded on videotape during the initial 48 h after weaning. Again, these data were extracted using 4-min scan samples, where the number of piglets performing active, lying, feeding, and agonistic behaviors in the activity area was noted.
Performance.
Each piglet was weighed individually on d 5, 12 (when the barriers were removed), and 28 (weaning) after the calculated farrowing date and at weekly intervals thereafter, until the fifth week after weaning.
Statistics
Statistical analysis was performed using GLM and MIXED procedures of SAS (SAS Inst. Inc., Cary, NC). Least squares means were evaluated using PDIFF of SAS. Results are presented as least squares means with SE (t-test). For consistency, all transformed least squares means are reported as back transformed least square means. A Wilcoxon 2-sample test using the NPAR1WAY procedure was performed for variables, which did not show a Gaussian distribution and could not be transformed into a Gaussian distribution. These results are presented as means (SEM).
Behavior of the Lactating Sow.
A total of 6,830 data points were available, which were aggregated into sums for each sow and each observation day (–1, 1, and 2 d relative to barrier removal). Data for lying sternal, lying lateral, and nursing were normally distributed. Sitting and standing, which were not normally distributed, were logarithmically transformed. Data were analyzed using the MIXED procedure, including fixed effects of observation day and a random sow effect. Restlessness was analyzed using the NPAR1WAY procedure.
Behavior of Suckling Piglets.
The 3 nursing litters, which were commingled, were considered the experimental unit for the behavior performed. For analyzing the behavioral data of suckling piglets, hourly means were calculated for each observation d (–1, 1, and 2 d relative to barrier removal) and each experimental unit. These means were normally distributed; the GLM procedure was used. The statistical model included the fixed effect of observation day and replicate. No interaction between observation day and replicate was found.
Behavior of Weaned Piglets.
Each pen in the nursery was considered the experimental unit for the behavior performed by weaned piglets. Hourly means were calculated for the behavior displayed by weaned piglets for each observation day and each treatment (experimental group and control group) within the replicates. The data for piglets in view, standing, and feeding were logarithmically transformed into a normal distribution and analyzed using the GLM procedure. The model used included the effect of treatment and observation day. No interaction between treatment and observation day was found. Moreover, no effect of replicate was established. The scatter of lying and aggressive behaviors in the activity area was markedly unequal, requiring a nonparametric test for analysis. Thus, statistical analyses were carried out using the NPAR1WAY procedure (Wilcoxontest) to evaluate the influence of treatment, observation day, and time block (a period of 4 consecutive hours).
Performance.
Birth weight of piglets has a great influence on subsequent growth performance (Wolter et al., 2002). Also, in this study, a high correlation (CORR) was found between the 5-d BW and weight gain during the nursing and rearing period. Genotype was also considered in the analysis. Therefore, an equation for predicting the BW gain of each piglet as a function of 5-d BW and genotype was developed, using the GLM procedure. Residual values from the prediction equation were pooled by pen and replicate for each weighing period and analyzed using the GLM procedure. The pens in the nursery were considered the experimental units. The model used for analysis of BW gain included the effect of treatment. No effect of replicate was found.
|
RESULTS
|
|---|
Behavior of Sows
After barriers were removed, sow behavior changed slightly (Table 2). Compared with their behavior on the before removal of the barriers, sows spent more (P = 0.05) time lying laterally and less (P = 0.02) time sitting on the day after removal of the barriers, and less (P = 0.04) time standing on the second day after the barriers were removed. Removal of the barriers did not affect the time sows allowed most (>75%) of their piglets to suckle. Also, removal of the barriers did not influence the frequency or duration of suckling bouts. On average, 29.6 suckling bouts/sow were observed daily with each suckling bout lasting on average 5.6 min (Table 3).
View this table:
[in this window]
[in a new window]
|
Table 2. Time budgets (h, least squares means) of the behaviors performed by sows in the experimental group1 24 h before and 48 h after removal of the barriers between the farrowing pens2
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Duration and frequency of suckling bout of sows in the experimental group1 24 h before and 48 h after removal of the barriers between the farrowing pens2
|
|
Removal of the barriers tended to result in increased (P = 0.08) restlessness in the experimental sows compared with the control group, 15.0% (3.9) vs. 4.4% (1.3; Wilcoxon test). This difference was no longer present 9 d later when no restlessness was observed in the sows.
Behavior of Piglets
Suckling Piglets.
Removal of the barriers did not influence the active and lying behaviors in piglets that were commingled in the farrowing room 12 d after birth (Table 4). Before removal of the barriers, piglets spent 76.1% of observations lying and showed 12.0% of observations in active behavior. After removal of the barriers, the proportion of lying lasted 75.9% (d 1 after removal) and 74.0% (d 2 after removal), respectively. After removal of the barriers, the proportions of observations of active behavior were 9.1% (d 1 after removal) and 11.0% (d 2 after removal). Removal of the barriers influenced the suckling behavior in the piglets (P < 0.02). After removal of barriers, the proportion of observations piglets suckled lasted 2.4% (d 1 after removal) and 3.1% (d 2 after removal) longer, respectively, compared with their suckling behavior before commingling. No differences in the behavior of piglets were found between the first and the second day following removal of the barriers.
View this table:
[in this window]
[in a new window]
|
Table 4. Time budgets (percentage of observations, least squares means) for the behaviors performed by piglets in the experimental group1 24 h before and 48 h after removal of the barriers between the farrowing pens2
|
|
Direct observations in the farrowing room were performed immediately after and for 9 d following the removal of the barriers, as well as 2 d before weaning. No incidence of crosssuckling was seen to occur during the direct observations in the farrowing room.
Weaned Piglets.
After weaning, clear differences in behavior between piglets of the experimental and control groups occurred. Considering the whole 48-h observation period, no differences were found for the proportion of piglets in the activity area (Table 5). On average, only 12.9% (experimental group) and 15.4% (control group), respectively, of the piglets stayed in the activity area. Moreover, the proportion of piglets exhibiting active and feeding behavior in the activity area did not differ between the treatments. However, a greater (P < 0.001) proportion of piglets performing lying and agonistic behaviors in the activity area was observed in the control group compared with the experimental group. Immediately after weaning, few piglets in the experimental group lay in the activity area, whereas 10% of the piglets in the control group lay in the activity area in the initial 4 h after weaning (P < 0.001, Table 6). Four to eight hours after weaning, the proportion of piglets lying in the activity area in the control group decreased to 2.4%. Subsequently, the proportion of piglets lying in the activity area remained at a low level. On the second day after weaning, no differences in lying behavior were found between the treatments.
View this table:
[in this window]
[in a new window]
|
Table 5. Proportion of piglets and their behaviors performed in the activity area during the initial 48 h after weaning as a function of the treatment1
|
|
View this table:
[in this window]
[in a new window]
|
Table 6. Means (SEM) of piglets lying in the activity area 24 h after weaning as a function of treatment and time beginning at 11301
|
|
In the case of piglets that were commingled before weaning, few agonistic behaviors were observed. Piglets that were housed conventionally during the nursing period exhibited more (P < 0.001) agonistic behaviors in the initial 4 h after weaning (Figure 1). Subsequently, the rate of agonistic behavior declined at nighttime and increased once more the next morning, when feeding started. Twenty-four hours after weaning, the proportion of piglets performing agonistic behavior remained at a low level in both groups. However, the rate of agonistic behavior of piglets in the experimental group was still lower 24 to 28 h (P = 0.008) and 44 to 48 h (P = 0.04) h postweaning.
View larger version (19K):
[in this window]
[in a new window]
|
Figure 1. Proportion of piglets performing agonistic behavior as a function of time after weaning (scan sampling) beginning at 1130. Item was analyzed using the SAS (SAS Inst. Inc., Cary, NC) NPAR1WAY procedure (Wilcoxon test); results are presented as means. P < 0.10; *P < 0.05; **P < 0.01; and ***P < 0.001; and NS = not significant.
|
|
Performance of Piglets
Three days after farrowing, 122 piglets belonging to the experimental group and 120 piglets from the control group were nursed. Of these, 9.8% of the experimental group and 7.2% of the control group died during the nursing and rearing period. Only piglets that survived are included in the following account.
On the first weighing, 5 d after the calculated farrowing date, piglets belonging to the experimental group weighed 1.89 ± 0.44 kg and piglets from the control group were 2.03 ± 0.53 kg. Correlations between 5-d BW, weaning weight, and BW gain during the nursing, rearing, and total trial period are summarized in Table 7. The BW gain of piglets was highly correlated (P < 0.001) with their 5-d and weaning weight.
During the nursing period, no differences in BW gain were found between the 2 treatment groups. A greater BW gain was observed for the experimental group during the rearing period (P = 0.05). Over the course of the entire study, experimental piglets gained 1.09 kg more (P = 0.05) in BW compared with their controls (Table 8).
View this table:
[in this window]
[in a new window]
|
Table 8. Means and 5-d BW and genotype-adjusted residuals (least squares means) of BW gain (kg) of piglets during the nursing and rearing periods, and in total, as a function of the treatment
|
|
The BW gains during the first week after weaning were very low in the experimental and the control group (Table 9). However, piglets that were commingled during the nursing period tended to gain more (P = 0.05) BW compared with those that were conventionally housed. In wk 1 after weaning, piglets in the experimental group gained 0.29 kg more weight than those in the control group.
View this table:
[in this window]
[in a new window]
|
Table 9. Means and 5-d BW and genotype-adjusted residuals (least squares means) of BW gain (kg) of piglets in each weighing period as a function of the treatment
|
|
|
DISCUSSION
|
|---|
Group housing of nursing sows, in which piglets are commingled with nonlittermates, has been the subject of discussion for many years, but widespread adoption of this practice has not occurred. A number of arguments support this decision, including more piglets being crushed, poorer performance of piglets due to cross-suckling, as well as the larger amount of work required to manage such systems. In the current study, piglets from 3 litters were grouped at 12 d of age by removing barriers between farrowing pens while sows remained confined. Removal of barriers only influenced the behavior of the sows to a limited extent. Sows in both treatment groups showed some degree of restlessness, which may have resulted in part from the other work (weighing, barrier removal) being carried out in the farrowing room. However, sows in the experimental group showed a greater increase in restlessness compared with controls suggesting that commingling itself did impact sow behavior. One reason for only a small change in sow behavior after commingling may be due to some degree of environmental or genetic induced decrease in maternal responsiveness. Modern genetic selection of breeding stock has emphasized piglet growth characteristics, the number of piglets born alive, and the number of piglets weaned. Thus, maternal behavior varies widely among sows (Lay et al., 2002). On the other hand, there is the possibility that sow behavior was only influenced to a limited extent because piglets under seminatural conditions also mix with one another at around the 10th day of life. Pedersen et al. (1998) noted that grouping litters during the nursing period leads to a significant decrease in nursing time. Similar results were also demonstrated by Weary et al. (1999) who reported a decrease in nursing following commingling litters at 11 d of age within the farrowing environment. These results were not confirmed in the current study. After commingling piglets, neither nursing time per day nor frequency of suckling bout per sow were changed compared with the day before removal of the barriers. In contrast, observation of piglets in the farrowing rooms revealed that the time piglets spent suckling increased after the piglets were commingled. Behavioral analyses of the nursing behavior of sows using continuous observation and the suckling behavior of piglets using scan sampling observation were always carried out on the same observation day. Continuous observation of the sows appears to be more precise than the scan sampling method with a 4-min interval. In the current study, single nursing bouts lasted about 5 to 6 min. It could be assumed that the time piglets suckled was overestimated by the scan sampling method, due to the 4-min interval.
Wattanakul et al. (1997a) studied the behavior of sows in a multisuckling system (group housing of sows and piglets from 2 wk after farrowing until weaning) and reported that piglet suckling behavior was severely disrupted during the initial 48 h after grouping. Furthermore, the litters also showed a high incidence of crosssuckling throughout lactation.
In contrast to alternative multisuckling systems, sows in this study remained confined during the entire nursing period. It is possible that continued confinement of sows after commingling enables piglets to more easily identify their mothers. This hypothesis is substantiated by the findings of Wattanakul et al. (1997b), who demonstrated that when sows remained confined after piglets were commingled crosssuckling occurred at a very low level. In this study, no crosssuckling was observed to occur, which seems due to the fact that sows remained confined. However, crosssuckling may have occurred but may have been missed because of the short direct observation period of 1 h on each observation day.
Jensen (2002) recommended mixing piglets when they are very young, preferably when they are 2 to 3 wk old. In the current study, piglets were commingled 12 d after the calculated farrowing date. At this age, piglets housed under natural conditions would normally begin to interact with unfamiliar littermates with low levels of aggression (Jensen and Redbo, 1987; Pedersen et al., 1998). Several studies demonstrated this experimentally. For instance, Petersen et al. (1989) reported that preweaning mixing produces nonaggressive and often playful social interactions between piglets. Weary et al. (1999) observed little aggression when litters were allowed to mingle at 11 d of age. According to Jensen (1994), the age of piglets at commingling determines the length and course of fighting. In his study, 1-wk-old piglets fought for a shorter time and in a more offensive way compared with 5- and 9-wk-old piglets. Similarly, Pitts et al. (2000) reported that fights between younger piglets had a shorter duration and resulted in fewer injuries than those between older ones. They suggested that mixing litters at a younger age has some potential welfare advantages compared with mixing at weaning.
After weaning, piglets in the experimental group showed few agonistic behaviors, whereas piglets that were members of the control group exhibited such behaviors, especially in the initial 4-h period after weaning. Piglets that were grouped 12 d after the calculated farrowing date had already established a hierarchy during the nursing period, so no new dominance conflicts occurred following weaning in contrast to the control groups in which unfamiliar piglets were commingled for the first time. Keeling and Jensen (2002) reported that the main reason for agonistic behavior being shown by piglets after weaning results from the commingling of unfamiliar individuals. Piglets had to fight in order to establish a new hierarchy. Wattanakul et al. (1997b) and Weary et al. (1999) assumed that the main effect of preweaning mixing appears to be a decrease in fighting after weaning.
In this study, immediately after weaning, few piglets in the experimental group lay in the activity area, whereas noticeably more piglets from the control group lay in the activity area. If most piglets stayed under the covered resting area, it was assumed that they were resting, due to the fact that this area occupied a space of only 3.2 m2 and thus there would not be enough space for them to perform a full repertoire of activity-related behavior. Therefore, it seems that piglets from the control group avoided close contact with others after weaning, until the new hierarchy had been established. As agonistic behavior in the control group decreased over time, their behavior patterns became more similar to those of the experimental piglets.
Previous studies evaluated the impact of birth and weaning weight on subsequent performance. The feed intake and the rate of weight gain in the nursery period are positively correlated with BW at weaning (Mahan and Lepine, 1991; Wolter and Ellis, 2001). Wolter et al. (2002) suggested that the birth weight of piglets has a great impact on postweaning growth. In the current study, the 5-d BW was highly correlated with the weaning weight and had an influence on the BW gain during the nursery period.
However, Wattanakul et al. (1997a) reported that during the nursing period, BW gain of piglets that were housed with their mother in group housing systems for lactating sows was lower in comparison with those of conventionally housed suckling piglets. This study confirmed the results found by Weary et al. (1999) where no differences were observed in BW gain between piglets that were socialized at d 11 of age compared with those that were not grouped with other litters.
Previously, Wattanakul et al. (1997a) also reported that piglets that were already grouped during the nursing period gained more BW after weaning, similar to results found in this study, particularly in the first week of the nursery phase. The findings of Keeling and Jensen (2002) further support these data because they also demonstrated that grouping unacquainted piglets is the main reason for agonistic behavior, which results in growth depression. According to Pajor et al. (1991) and King and Pluske (2003), piglets usually suffer a post-weaning growth lag in the 7- to 14-d period following weaning, which is thought to be associated with abrupt and premature weaning. This is characterized by low and variable feed intake and a poor and variable growth rate. It appears that piglets in this study that were commingled before weaning were better equipped to overcome the usually growth lag after weaning. In practice, commingling piglets prior to weaning is relatively easy to carry out by removing the barriers between adjacent farrowing pens. By allowing unfamiliar piglets to mix with each other at the time they normally would under natural conditions (around d 10 of life), it is possible to reduce weaning stress by removing a lot of the negative agonistic behaviors that contribute to poorer performance.
Mixing of litters during the nursing period had no negative effects on behavior or performance of piglets. Socializing piglets early in life by mixing litters before weaning had a positive effect on behavior and performance after weaning. By commingling litters at a younger age, the problem of exposing piglets to multiple stressors at weaning can be reduced. In order to establish whether the positive effects of socializing are also evident in larger groups, further research should be carried out on commingling more than 3 litters at any one time.
1 Corresponding author: earkena{at}gwdg.de
Received for publication October 20, 2005.
Accepted for publication May 15, 2006.
|
LITERATURE CITED
|
|---|
Brooks, P. H., and C. A. Tsourgiannis. 2003. Factors affecting the voluntary feed intake of the weaned pig. Pages 81–115 in Weaning the Pig—Concepts and Consequences. J. R. Pluske, J. Le Dividich, and M. W. A. Verstegen, ed. Wageningen Academic Publishers, Wageningen, the Netherlands.
D’Eath, R. B. 2005. Socialising piglets before weaning improves social hierarchy formation when pigs are co-mingled post-weaning. Appl. Anim. Beh. Sci. 93:199–211.[CrossRef]
Held, S., and M. Mendl. 2001. Behaviour of the young weaned pig. Pages 273–297 in The Weaner Pig—Nutrition and Management. M. A. Varley, and J. Wiseman, ed. CAB Int., Oxon, UK.
Jensen, P. 1994. Fighting between unacquainted pigs—Effects of age and of individual reaction pattern. Appl. Anim. Behav. Sci. 41:37–52.[CrossRef]
Jensen, P. 2002. Behaviour of pigs. Pages 159–172 in The Ethology of Domestic Animals—An Introductory Text. P. Jensen, ed. CAB Int., Oxon. UK.
Jensen, P., and I. Redbo. 1987. Behaviour during nest leaving in free-ranging domestic pigs. Appl. Anim. Behav. Sci. 18:355–362.
Keeling, L., and P. Jensen. 2002. Behavioural disturbances, stress and welfare. Pages 79–98 in The Ethology of Domestic Animals—An Introductory Text. P. Jensen, ed. CAB Int., Oxon, UK.
King, R. H., and J. R. Pluske. 2003. Nutritional management of the pig in preparation of weaning. Pages 37–51 in Weaning the Pig—Concepts and Consequences. J. R. Pluske, J. Le Dividich, and M. W. A. Verstegen, ed. Wageningen Academic Publishers, Wageningen, the Netherlands.
Lay, D. C., Jr., R. L. Matteri, J. A. Carroll, T. J. Fangman, and T. J. Safranski. 2002. Preweaning survival in swine. J. Anim. Sci. 80(E. Suppl. 1):E74–E86.[Abstract/Free Full Text]
Mahan, D. C., and A. J. Lepine. 1991. Effect of pig weaning weight and associated nursery feeding programs on subsequent performance to 105 kilograms body weight. J. Anim. Sci. 69:1370–1378.[Abstract]
Newberry, R. C., M. 
pinka, and S. Cloutier. 2000. Early social experience of piglets affects rate of conflict resolution with strangers after weaning. Page 67 in Proc. 34th Int. Congr. ISAE, 17–20 October 2000. A. Ramos, L. C. Pinheiro Machado, M. J. Hötzel, ed. Florianópolis, Brazil.
Pajor, E. A., D. Fraser, and D. L. Kramer. 1991. Consumption of solid food by suckling piglets: Individual variation and relation to weight gain. Appl. Anim. Behav. Sci. 32:139–155.[CrossRef]
Pedersen, L. J., M. Studnitz, K. H. Jensen, and A. M. Giersing. 1998. Suckling behaviour of piglets in relation to accessibility to the sow and the presence of foreign litters. Appl. Anim. Behav. Sci. 58:267–279.[CrossRef]
Petersen, H. V., K. Vestergaard, and P. Jensen. 1989. Integration of piglets into social groups of free-ranging domestic pigs. Appl. Anim. Behav. Sci. 23:223–236.[CrossRef]
Pitts, A. D., D. M. Weary, E. A. Pajor, and D. Fraser. 2000. Mixing at young ages reduces fighting in unacquainted domestic pigs. Appl. Anim. Behav. Sci. 68:191–197.[CrossRef][Medline]
Pluske, J. R., J. Le Dividich, and M. W. A. Verstegen. 2003. Conclusions. Page 421 in Weaning the Pig—Concepts and Consequences. J. R. Pluske, J. Le Dividich, and M. W. A. Verstegen, ed. Wageningen Academic Publishers, Wageningen, the Netherlands.
Wattanakul, W., A. G. Sinclair, A. H. Steward, S. A. Edwards, and P. R. English. 1997a. Performance and behaviour of lactating sows and piglets in crate and multisuckling systems: A study involving European White and Manor Meishan genotypes. Anim. Sci. 64:339–349.
Wattanakul, W., A. H. Stewart, S. A. Edwards, and P. R. English. 1997b. Effects of grouping piglets and changing sow location on suckling behaviour and performance. Appl. Anim. Behav. Sci. 55:21–35.
Weary, D. M., E. A. Pajor, M. Bonenfant, S. K. Kross, D. Fraser, and D. L. Kramer. 1999. Alternative housing for sows and litters: 2. Effects of a communal piglet area on pre- and post-weaning behaviour and performance. Appl. Anim. Behav. Sci. 65:123–135.[CrossRef]
Wolter, B. F., and M. Ellis. 2001. The effect of weaning weight and rate of growth immediately after weaning on subsequent pig growth performance and carcass characteristics. Can. J. Anim. Sci. 81:363–369.
Wolter, B. F., M. Ellis, B. P. Corrigan, and J. M. DeDecker. 2002. The effect of birth weight and feeding of supplemental milk replacer to piglets during lactation on preweaning and postweaning growth performance and carcass characteristics. J. Anim. Sci. 80:301–308.[Abstract/Free Full Text]
Top
Abstract
INTRODUCTION
