| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL PRODUCTION |
* Department of Animal and Poultry Science and
and
Department of Population Medicine, University of Guelph, Guelph, ON, Canada N1G 2W1; and
and
College of Veterinary Medicine and
and
Department of Animal Science, Michigan State University, East Lansing 48824
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: aggression • boar presence • group housing • sow • stress response
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Mount and Seabrook, 1993; Arey, 1999). Consequently, sows may suffer from injury and stress, which compromises their welfare and productivity (Marchant et al., 1995; Barnett et al., 2001). The presence of a boar at mixing could potentially be used as a management tool for reducing aggression among sows, as this practice has been shown to reduce fighting among mixed slaughter-weight pigs (Grandin and Bruning, 1992), ovariectomized pigs (Barnett et al., 1993), and sows mixed immediately after weaning (Docking et al., 2001). Application of boar pheromonal androgens has also been shown to reduce aggression among immature pigs (McGlone et al., 1986; McGlone and Morrow, 1988). Any effect of boar presence on reducing aggression and its associated consequences among groups of newly mixed bred or gestating sows has yet to be studied. Furthermore, there is no research documenting the impact of boar removal on the behavior of group-housed sows. The objective of this work was to determine how different levels of boar presence influence agonistic behavior, shoulder scratches, and acute stress response of bred sows during the first few days after mixing. Moreover, we hoped to determine the effect of boar removal, after hierarchy formation, on these measurements.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
All procedures in this study were reviewed and approved by the University of Guelph Animal Care Committee in accordance with the Canadian Council on Animal Care. Two hundred twenty-five bred Yorkshire sows (parity = 2.7 ± 0.12; BW = 203.5 ± 2.0 kg) were used in this study. To facilitate onset and detection of estrus, sows were weaned into groups of 3 or 4 and housed adjacent to a mature boar. Sows were artificially inseminated with approximately 3 x 109 sperm at estrous detection and again at 24-h intervals while they were still in standing estrus. After breeding, sows were housed in individual gestation stalls until they were assigned to groups. This research was conducted concurrently with a larger study investigating the effects of mixing sows at various times postbreeding on their reproductive performance. Therefore, sow groups were composed of 15 sows, of which approximately 3 sows were from each of the 5 postbreeding times (2, 7, 14, 21, or 28 d postbreeding). These postbreeding times were evenly distributed within and across groups so that group composition with respect to postbreeding time was balanced across treatments.
Each group of 15 sows was exposed to 1 of 3 levels of boar presence (n = 5 groups per level): physical (boar in pen with sows; 2.15 m2 per sow), fenceline (boar housed adjacent to sows; 2.3 m2 per sow), or control (no boar in the room; 2.3 m2 per sow). Because of constraints on sow availability, only one group of sows was mixed within a given week, and only one boar was present in the room at any time; no boars were present in the room during mixing of control groups. Five intact mature boars (~2 yr of age) were used in this study, and each boar was balanced across the 2 boar treatments. Boars were chosen from a boar pool based on age, BW, and body condition. All boars were heavier than the average sow (boar BW = 301.0 ± 10.1 kg).
The experiment was performed in 2 phases. Phase 1 was the mixing phase, and phase 2 was the boar removal phase. During phase 1, sows were introduced into the group pen at approximately 1030, thus ensuring that they had eaten their daily ration. A boar was introduced into the sow pen or adjacent to the sow pen, depending on treatment, 15 to 30 min before the sows were mixed (approximately 1000). Behavioral measurements were taken for 48 h after mixing, as described subsequently. During phase 2, boars were removed at approximately 1000 on the seventh day after mixing, and behavioral measurements were taken for 24 h before and for 48 h after boar removal. Before a new group was introduced, all pens housing sows, boars, or both were power-washed with a commercial washer and disinfected to remove boar odors. All pigs were obtained and housed at the University of Guelph Arkell Swine Research Station.
Facilities and Management of the Sows
Pens measured approximately 3.0 m wide x 11.5 m long and were divided widthwise into 3 sections, 2 feeding and/or lying sections and 1 dunging area, by 2 partial walls made of concrete block (0.2 m wide x 1.5 m long x 1.1 m high). The dunging area (3.0 m wide x 5 m long) had a slatted floor where hanging drinkers were located. Feeding and/or lying areas (3.0 m wide x 3.3 m long) had a solid concrete floor and 2 separate drop feeders suspended from the ceiling. Pens were separated lengthwise on one side by spindle penning and by 2 partial walls adjacent to the feeding and/or lying areas and on the other side by solid concrete walls. Space allowance could not be adjusted between treatments because of the permanent nature of the walls. Boars used in the fenceline treatment were housed in the pen adjacent to the sow pen and were separated by spindle penning. The dimensions and layout of the pen in which the boar was housed were identical to the sow pen.
Sows and boars were each fed approximately 2.5 kg of a standard pelleted breeding diet once daily between 0730 and 0830. Feed was distributed onto 2 separate areas of the floor (each 1 m in diameter) by a drop-feeding system. Boars housed adjacent to a sow pen were fed by hand. The experiment was conducted over 8 mo (August to March); average daily temperatures were 24.1 ± 0.2°C in the room.
Data Collection
Figure 1(a and b) summarizes the data collection schedule.
|
Behavior. A digital recording system (i31808WM, Kodicom, Toronto, ON, Canada) was used to continuously record (30 frames/s) the behavior of the sow groups from the time they were introduced into the pen until 48 h postmixing. The control treatment followed the same video recording schedule. Three color cameras (WV-CP240, Panasonic, Mississauga, ON, Canada) were positioned on the ceiling above each pen. Sows were individually identified with numbers using livestock spray to facilitate data collection. Three rows of fluorescent lights illuminated the room. All lights were on from 0730 to 1600. At 1600, the lights were turned off, except for the row of lights above the dunging area. Individual sow behavior could be discerned even with the low light level.
Behavioral data were collected from the digital recordings. Digital recordings were observed in real time during periods with little activity and then slowed down by 50% to observe behavior during active periods (e.g., mixing and feeding). Groups were observed continuously for 2 periods, 0 to 24 and 25 to 48 h postmixing, to determine the frequencies of intersow agonistic behavior and the frequencies and durations of fights. Interactions involving the boar were also recorded but were not included in the analyses. Five mutually exclusive agonistic behavior patterns were observed (Table 1). Bites, head knocks, and body knocks represent short, independent aggressive interactions that resulted in the immediate retreat of the recipient sow. Frequencies of bites, head knocks, and body knocks were summed to obtain the frequency of independent "aggressive contact." Fights included all of the 3 behavioral components of aggressive contact (bite, head knock, and body knock), but individual frequencies of these behavior patterns within a fighting bout were not recorded. Groups were not observed from 0600 to 0700 or from 1600 to 1700 on all days, as experimenters were in the pen. The control treatment followed the same behavior sampling schedule. Behavioral data were separated into feeding and nonfeeding times. Feeding periods (49.6 ± 1.4 min) were defined as beginning when the feed was dropped on the floor by the feeders and ending when >50% of the sows were no longer engaged in oral and/or nasal activity directed toward the floor.
|
). The scoring system for the number of skin scratches was developed following methods described by Hodgkiss et al. (1998) and de Koning (1984). A score of 0 was assigned when no scratches were evident on a shoulder; 1 was assigned when there were <5 scratches observed on a shoulder; 2 was assigned when 5 to 10 scratches were visible on a shoulder; and a score of 3 indicated >10 scratches per shoulder. The sum of both shoulder scores was calculated, and the resulting values were recategorized as none (score = 0), mild (score = 1 or 2), moderate (score = 3 or 4), or multiple (score = 5 or 6). Qualitative features of the scratches were not documented.
Saliva Collection and Hormone Assay.
Saliva was collected to determine the effects of boar presence on the stress response of the sows. Saliva samples were collected in the morning (0600 to 0700) and afternoon (1600 to 1700) on the day before and for 2 d after mixing (Figure 1a
). Pig saliva was collected from 2 cotton dental buds attached to 1 m of dental floss tied at the center of the buds. While housed in breeding stalls, the sows were trained to chew on the cotton buds until they were thoroughly moistened. To reduce the fearfulness of the sows toward the collector and to reinforce cooperation during the sampling period, the sows were given a food reward after successful collections. The cotton buds were placed in 15-mL conical tubes (Fisher, Mississauga, ON, Canada) and were kept on ice until they could be centrifuged at 1,000 x g at 4°C for 15 min to remove the saliva. Usually, 1 to 2 mL was retrieved, which was then stored at –20°C before cortisol analysis.
Concentrations of cortisol were measured in duplicate saliva samples of 500 µL using a commercial, direct solid-phase 125I-RIA kit (Coat-A-Count Cortisol TKCO Diagnostic Products, Los Angeles, CA) modified for pig salivary cortisol. Samples were thawed and centrifuged for 15 min at 1,300 x g at 4°C to precipitate mucins. The manufacturer’s procedures for assaying cortisol in saliva were used with the following modifications. The human serum-based calibrators were diluted to final cortisol concentrations of 0, 0.005, 0.01, 0.05, 0.1, 0.25, 0.5, 1.0, 5.0, and 10.0 µg/dL. The supernatant was decanted from each tube after a 3-h incubation at room temperature. The remaining radioactivity was measured for 2 min using a Packard Cobra II Auto Gamma Counter (Packard Instrument Co., Meriden, CT). Intraassay and interassay coefficients of variation were determined from 15 assays and were 5.1 and 9.0%, respectively.
Phase 2: Boar Removal
Behavior. Behavior of groups was recorded following the same method previously described but beginning on d 6 after mixing for a 24-h period before and the 48-h period after removal of the boar. Groups were observed continuously for 3 periods, 0 to 24 h preremoval and 0 to 24 and 25 to 48 h postremoval, to determine the frequencies of agonistic behavior and the frequencies and durations of fights. Data were summarized according to feeding and nonfeeding periods as in phase 1. Control groups were observed at the same times relative to boar removal.
Shoulder Scratch Scores.
Superficial skin scratches were assessed following the same method as described previously for both the pre- and postboar removal periods (Figure 1b). Scratch scores were taken between 1000 and 1100. Control groups were scored at the same times of day relative to boar removal in the other groups.
Saliva Collection. Saliva was collected to determine the effects of boar removal on the stress response of sows. Individual sow samples were collected following the same method as previously described. Control groups followed the same saliva collection protocol.
Statistical Analyses
The experimental design incorporated an incomplete randomized block design, accounting for 3 treatments and 5 blocks; group was the experimental unit. Blocks consisted of one replicate of each treatment randomized within a 1-mo period, thus accounting for seasonal effects. Individual boars and blocks were included in the model as random effects. All statistical analyses were conducted using SAS (SAS Inst. Inc., Cary, NC). For all variables, ANOVA using Proc Mixed were completed separately for the 2 phases [phase 1 (mixing) and phase 2 (boar removal); Kuehl, 1994]. The univariate procedure in SAS was used to assess data for normality, and Brown and Forysthe’s modified Levine test was performed to test homogeneity of variance across treatment groups. Data that did not meet the requirements for the ANOVA before analysis were transformed. All data were normally distributed after transformation. When significant main effects were found, Tukey’s W procedure was used to compare treatment means. Untransformed means and SE are presented in all tables and figures.
Behavior.
Frequencies of intersow agonistic behavior patterns and the total duration of fights were calculated as the number per group per h and seconds per group per h, respectively; the average fighting bout duration was calculated in seconds per bout. All behavior variables for the feeding and nonfeeding periods were analyzed within a phase. Feeding and nonfeeding data were also combined and analyzed (data not shown). Frequency and total duration of fights as well as average fighting bout duration were not statistically analyzed during phase 2, as they were rare events and the data sets consisted mainly of zeros. For frequency data, the loge(x + 1) and reciprocal transformations were performed. Square root transformations were performed on the total duration of fights and on the average fighting bout duration. To determine any change within a phase, paired Student’s t-tests were performed on the frequency of fighting behavior between days (Kuehl, 1994). The Pearson Product Moment Correlation was used to test any relationships between the frequency of fights, total duration of fights, average fighting bout duration, and shoulder scratch scores (Kuehl, 1994).
Shoulder Scratch Scores.
Mean scratch scores for groups were analyzed to determine any difference between treatments during the different phases (Snedecor and Cochran, 1980). For the mixing phase, the premixing value was used as a covariate for the postmixing value. Similarly, the preremoval value was used as a covariate for the postremoval value during the boar removal phase. For statistical purposes, the moderate and multiple values were combined to create a moderate/multiple score. Separate analyses were performed on the percentage of sows in a group assigned moderate/multiple scratch scores, as these represented sows that received a great number of shoulder scratches. All shoulder scratch data were normally distributed.
Salivary Cortisol.
Because salivary cortisol follows a distinct diurnal pattern that results in greater morning vs. afternoon concentrations (Zanella et al., 1998), separate analyses were performed on the morning and afternoon values. The premixing morning or afternoon value was used as a covariate when analyzing the corresponding postmixing levels. Similarly, either the preremoval morning or afternoon value was used as a covariate when analyzing the corresponding postremoval levels. All cortisol data were normally distributed.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Behavior.
No effect of the level of boar presence on the frequency of threats postmixing during either the feeding or nonfeeding periods was observed (Table 2). Control groups had fewer aggressive interactions than fenceline groups during the 25- to 48-h postmixing feeding period (P = 0.01). The number of fights occurring in the 25- to 48-h postmixing feeding period was lower for physical sows than for fenceline sows (P = 0.023); the control sows had an intermediate number of fights. The data for total duration of fighting during feeding corresponded with frequency of fights, although fight duration did not differ between treatments. Average duration of fighting bouts was also unaffected by treatment postmixing during both feeding and nonfeeding periods. Summing the 0- to 24-h and 25- to 48-h postmixing results for feeding and nonfeeding periods did not reflect any treatment differences in frequency of aggressive contacts, threats, or fights; total duration of fights; or average duration of a fighting bout (data not shown). For all treatments combined, frequency of fighting was lower during the 25- to 48-h postmixing period than during the 0- to 24-h postmixing period for both feeding (P < 0.01) and nonfeeding (P < 0.001) periods.
|
. Groups of sows exposed to the physical treatment had lower mean scratch scores postmixing when compared with control sows (P = 0.048). Mean percentages of sows within a group assigned moderate/multiple scores premixing by treatment were control = 2.7 ± 1.6%, fence-line = 4.0 ± 1.6%, and physical = 6.7 ± 3.0%. The percentage of sows within a group exhibiting the moderate/multiple scratch score tended to be lower postmixing (P = 0.06) for physical sows vs. control sows, although neither group was different from the fenceline sows (control = 42.7 ± 9.6%; fenceline = 36.0 ± 8.1%; physical = 28.0 ± 6.8%).
|
Salivary Cortisol.
Mean salivary cortisol concentrations postmixing are presented in Figure 3a. The physical groups had a greater mean salivary cortisol concentration than the fenceline groups during the morning (P = 0.08) and afternoon, at 6 h (P < 0.01) and at 32 h postmixing (P = 0.08). Mean morning and afternoon salivary cortisol concentrations premixing, for all treatment groups combined, were 2.0 ± 0.3 and 1.8 ± 0.1 ng/mL, respectively.
|
Behavior.
No effect of level of boar presence was observed on frequencies of aggressive contacts or threats pre- and postboar removal during either feeding or non-feeding periods (Table 3). Frequency of fighting was low during both feeding and nonfeeding periods, and values remained constant from pre- to postboar removal (P > 0.10).
|
. Mean scratch score was lower for the physical group than for the fenceline (P = 0.046) and control (P = 0.027) groups preboar removal. Mean scores assigned 24 h later, postboar removal, only differed between the physical and fenceline groups (P = 0.047). The mean percentage of sows within a group assigned moderate/multiple scores preremoval was unaffected by treatment (control = 45.3 ± 12.2%; fenceline = 41.3 ± 6.8%; physical = 26.7 ± 5.6%; P > 0.10). Mean percentage of sows within a group exhibiting the moderate/multiple scratch score was lower postremoval (P = 0.05) for physical sows than for fenceline sows, although neither were different from the control sows (control = 40.0 ± 12.1%; fenceline = 42.7 ± 7.0%; physical = 22.7 ± 5.4%).
Salivary Cortisol.
Mean salivary cortisol concentrations relative to boar removal are presented in Figure 3b. No effect of treatment was observed (P > 0.10) on any of the values. The mean morning and afternoon salivary cortisol concentrations preboar removal, for all treatment groups combined, were 3.1 ± 0.2 and 2.0 ± 0.1 ng/mL, respectively.
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Leeb et al., 2001). However, our results demonstrate that the relationship between boar presence at the time of mixing and the occurrence of agonistic behavior, scratches, and stress response among bred sows is not straightforward. During the mixing phase, the behavioral data do not provide sufficient evidence that boar presence, whether fenceline or physical, reduced intersow aggression. The only measure reduced by boar treatment was the frequency of fights occurring at feeding during the 25- to 48-h postmixing period, and this difference was only found between groups of physical and fenceline sows. In fact, frequency of aggressive contacts was greater in fenceline groups than in the control group during the same period. Despite the few behavioral differences between treatments, physical sows had fewer scratches postmixing than did control sows, and scratch scores were numerically, but not statistically, lower for physical sows compared with fenceline sows. Salivary cortisol concentrations did not correspond with either the behavioral or scratch data, as values were markedly greater in groups of sows in physical contact with a boar.
The general lack of behavioral differences was surprising because several previous studies have demonstrated that a boar effectively reduces aggression among pigs. There may be a number of reasons for the discrepancies among studies. The first may be the time frame during which behavior of newly formed groups is observed. Grandin and Bruning (1992) observed groups of 50 slaughter-weight pigs for 50 min postmixing when they were housed with either 3 mature boars or no boars. Boar presence reduced the number of fights, resulting in fewer pigs being wounded. Barnett et al. (1993) also found that housing a boar with groups of four 7.5-mo-old ovariectomized pigs reduced aggression. However, their results related only to a 90-min period after mixing. Although the majority of fighting generally occurs within this time frame (Luescher et al., 1990; Mount and Seabrook, 1993), it takes a group of newly mixed sows approximately 1 wk to become socially stable (Moore et al., 1993; Arey, 1999). Therefore, concentrating on only the first few hours postmixing may be misleading.
The age, size, or physiological status of the animals studied may be another important determinant of how newly mixed groups respond to the presence of a boar. It might be expected that smaller, younger pigs respond submissively and stop fighting when a boar is present. For example, application of boar pheromonal androgens alone has been shown to reduce aggression among immature pigs (McGlone et al., 1986; McGlone and Morrow, 1988). However, they alone did not reduce aggression among sows (Stansbury and McGlone, 1987). For mature sows, reproductive and, therefore, endocrine state could play a role. The endocrine profiles of sows at various reproductive stages are different and could influence how they respond to boar presence. During lactation, levels of progesterone and estrogen are extremely low (Pond and Mersmann, 2001). At weaning, reproductive hormones rapidly change, and within a few days, sows show receptive behavior to the boar as ovulation approaches (Pond and Mersmann, 2001). These behavior patterns subside once ovulation has occurred. Whereas a few of the sows in the current study were weaned 6 d (2 d bred) before groups were mixed and exhibited behavioral signs of estrus during this study, the majority of the sows in all groups were in the early stages of gestation. The fact that groups consisted of sows from various days postbreeding might have resulted in differential effects within a group that could have influenced the overall effectiveness of the boar. In addition, a sow’s previous experience with a boar could alter her response. However, Docking et al. (2001) found reduced aggression among newly weaned sows that had previous experience with boars.
A difference in results among studies may also be due to the size of group to which the boar was introduced. Both Barnett et al. (1993) and Docking et al. (2001) used group sizes of <7 animals, whereas groups consisted of 15 sows in our study. Therefore, presence of a boar might have more impact when fewer animals are present. Finally, the lack of behavioral differences in the current study might have been due to the way that aggression was measured. Whereas "independent" aggressive contacts outside of fighting were quantified, the numbers and types of aggressive contacts occurring within a fight were not, which might have influenced the outcome of our study. However, the general time patterns for fighting in our study are in agreement with other postmixing studies, and similar methods for data collection were used (Docking et al., 2001). Fighting was greatest on the first day and was lower by d 2 (Luescher et al., 1990; Docking et al., 2001). It was also rarely observed during the second week. Our results also agree with those of others (Csmerly and Wood-Gush, 1987; Barnett et al., 1993; Brouns and Edwards, 1994), who reported that the majority of aggressive encounters occurred during feeding.
Analyzing behavioral acts occurring within a fight (Jensen, 1980, 1982) may be more important than the frequency or duration of a fight. In our study, although the frequency and duration of a fight were recorded, the intensity was not incorporated. For example, if sows were to receive fewer bites or head-to-head knocks, during one fight compared with another of similar duration, it might have resulted in fewer shoulder scratches. Therefore, because we found no correlation between frequencies of fights and scratch scores, as did Barnett et al. (1993), the intensity of a fight might have been the determining factor. Grandin and Bruning (1992) have previously reported that the nature of fights occurring in the presence of a boar seemed to be less intense. Whereas we did not find a treatment difference in duration of fights, our results indicate a relationship between total duration of fighting, average duration of a fighting bout, and scratch scores, suggesting that fewer scratches were found in individual groups of sows that spent less time fighting postmixing. However, variation among groups was large. Because the sows in physical contact with a boar did not fight any less than the other treatments, perhaps the intensity of these fights was different, leading to the lower scratch score. Although statistically significant, the mean differences between treatments with or without a boar physically present were less than one-half of 1 point on a 6-point scale. Even when the boar is physically present, nearly 30% of sows had shoulder scores in the moderate/multiple category, indicating that the effectiveness of the boar was modest at best.
In our study, housing sows in physical contact with a boar at mixing resulted in a greater concentration of salivary cortisol compared with sows without physical boar contact. Previous work demonstrated that mixing unfamiliar pigs resulted in elevated cortisol concentrations (Mendl et al., 1992); thus, we expected that the presence of a boar would reduce stress among bred sows if fighting were minimized. Previous work by Pearce and Hughes (1987) showed that introduction of prepubertal gilts to boars to stimulate puberty leads to an increase in plasma concentrations of cortisol. In addition, ovariectomized gilts that were hormonally induced to show behavioral signs of estrus also showed immediate elevations in cortisol levels when introduced into a pen housing a boar and after mating (Barnett et al., 1982). In our study, it is possible that the boar could have sexually aroused some of the sows, as mounting was observed on a few occasions. More specifically, there were a few sows that were showing receptive behavior; thus, they might have had a greater cortisol response caused by sexual stimulation, as indicated by Barnett et al. (1982). However, reanalysis of the cortisol data, removing all individuals bred 2 d before mixing from the analysis, did not alter the results. Therefore, the elevation in salivary cortisol was likely not due to sexual arousal.
In view of this, evidence indicates that introducing nonsexually receptive females to a boar, for detection of estrus, increases their levels of plasma cortisol compared with sows with fenceline exposure to a boar within 5 (Turner et al., 1998) or 15 min (Jongman, 1993 in Turner et al., 1998). When females are not sexually receptive, they will actively try to avoid the courting boar (Barnett and Hemsworth, 1991). Therefore, sows that are not sexually receptive may find the presence of a boar aversive when physical contact is possible; thus, boar presence may be more stressful than fighting itself. To truly determine whether sows perceive the boar as aversive, some other method, such as a preference test, needs to be conducted. Signoret (1970) previously demonstrated that when given a choice, estrous females will consistently choose a boar over another female pig. However, this choice may be different if the individual is not sexually receptive.
In natural settings, mature boars live a relatively solitary life and generally only interact with sow groups during breeding seasons (Mauget, 1981). Boars may assume dominance over sows in natural groups because of their size (Mauget, 1981), although it is not clear whether they actually become a part of the hierarchy. In a captive setting, where males and females are in proximity of one another, European wild boars have been observed to show aggression toward a sow cohort, which was interpreted that the boar was a part of a multisex social hierarchy (Schnebel and Griswold, 1983). In our study, agonistic behavior involving the boar was minor. Out of 4,298 interactions compiled from the 5 groups exposed to the physical treatment during the 0- to 48-h postmixing and 0- to 24-h preboar removal periods, boars initiated or received 407 agonistic interactions or 9.5% of all agonistic behavior. In general, this represents one interaction per boar per h for each day he was in the pen. More specifically, boars were involved in 7 of 155 fights within the first 24 h postmixing, of which 6 occurred in one particular group; furthermore, 2 different sows initiated 2 fights each. In all instances, the sow retreated. Grandin and Bruning (1992) reported that pigs display submissive behavior upon boar approach, perhaps indicative of a boars’ social status. In our study, of 196 recorded fights occurring within the first 48 h, boars interfered in only 5 fights.
We designed the second phase of our experiment to address a second hypothesis that if boar presence reduced fights during the postmixing period, fighting might emerge once the boar was removed if social hierarchy were disrupted. According to our results, removing the boar 1 wk later did not alter intersow interactions. Thus, the lack of differences among measured variables, as well as the fact that over the course of time the frequency of fighting decreased or remained constant, indicates that removal of the boar did not influence the social structure of the group. Observations of the boars throughout the experiment also indicated that the boar was not a part of the sow social structure. Boars housed in the same pen as sow groups rarely slept with the cohort and generally slept in the dunging area. Whereas they ate with the sows, boars did not dominate the feed piles and were on occasion the first to lie down. Boars generally behaved in a solitary manner.
In conclusion, using a boar as a management tool to reduce aggression among bred sows is questionable. Although we found subtle reductions in fighting behavior and scratches, sows elicited a greater stress response when in the presence of the boar. Agonistic behavior and scratch scores postremoval remained unchanged, suggesting that the boar did not suppress fighting.
|
Footnotes |
|---|
2 Corresponding author: twidowsk{at}uoguelph.ca
Received for publication August 22, 2005. Accepted for publication December 19, 2005.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
-androst-16-en-3-one reduced agonistic behavior and temporarily improved performance of growing pigs. J. Anim. Sci. 63:679–684.-androst-16-en-3-one (androstenone) on sow social behavior and post-weaning return to estrus. J. Anim. Sci. 65 (Suppl. 1):234. (Abstr.)
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
