UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Agonistic behaviour in finishing pigs (Sus scrofa) following mixing : its effect on productivity 1988

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1988_A6_7 T36.pdf [ 3.07MB ]
Metadata
JSON: 1.0097837.json
JSON-LD: 1.0097837+ld.json
RDF/XML (Pretty): 1.0097837.xml
RDF/JSON: 1.0097837+rdf.json
Turtle: 1.0097837+rdf-turtle.txt
N-Triples: 1.0097837+rdf-ntriples.txt
Citation
1.0097837.ris

Full Text

AGONISTIC BEHAVIOUR IN FINISHING PIGS (SUS SCROFA) FOLLOWING MIXING: ITS EFFECTS ON PRODUCTIVITY by SHENTON S.L. T A N B.Sc. (Agr.), The University of British Columbia, 1983 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Animal Science) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January 1988 ® Shenton S.L. Tan, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Animal Science The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date February 2 9 . 1988 DE-6G/81) A B S T R A C T The e f f e c t of mix ing unfamil iar f in ishing pigs (Sus scrofa) on agonist ic behaviour and product iv i ty over a 3 week per iod w a s invest igated. Nine groups of six pigs were a l located to one of three treatments and eight groups of six pigs to the fourth treatment. In the f irst treatment (unmixed) l ittermate groups were m o v e d into a new pen and in the s e c o n d treatment (3:3 mixed) 3 pigs f r o m one l ittermate group were mixed with 3 pigs f r o m a s e c o n d l i ttermate group. The third treatment (Stresn i l - t reated) w a s s imi lar to the s e c o n d treatment but pigs were injected with the tranquil izer Stresni l (azaperone) prior to mix ing. In the fourth treatment (5:1 mixed) groups of f i ve pigs were introduced into a pen already occup ied by either a s ingle re lat ive ly light weight pig or a re lat ive ly heavy weight p ig . Intense f ight ing w a s d i sp layed by the regrouped pigs immediate ly f o l l o w i n g mix ing, whi le unmixed and S t r e s n i l - t r e a t e d pigs general ly went to s leep . During feed ing per iods , initiated aggress ion w a s the m o s t c o m m o n agonis t i c behaviour exceeding aggress ive r e s p o n s e s and s u b m i s s i v e r e s p o n s e s by a factor of up to 14. In mixed groups initiated aggress ion w a s s ign i f i cant ly higher than in unmixed groups. A d m i n i s t r a t i o n of Stresni l appeared to disrupt the an imals ' behavioural repetoire by de lay ing a g g r e s s i o n , retarding soc ia l hierarchy establ ishment and depress ing product iv i ty . Prior o c c u p a n c y of pen space a lso appeared to inf luence aggress ive behaviour . Over the entire three week sample per iod , average dai ly weight gains ( A D G ) of all three mixed treatments were s ign i f i cant ly less than unmixed groups . The d i f fe rences were s igni f icant during the f irst week but not in the s e c o n d or third w e e k s . The mixed groups were a lso ii poorer conver ters of feed during the f irst week and over the three week per iod . S t resn i l - t reated p igs , on average, exhibited the poorest product iv i ty of the mixed treatments. The e c o n o m i c c o s t s of rais ing mixed groups f r o m an initial weight of 76 kg to a standard final weight of 95 kg as a result of their reduced weight gain and f e e d e f f i c i e n c y , w a s substant ia l : $2.92 per pig for S t r e s n i l - t r e a t e d groups; $1.43 per pig for 3:3 mixed groups; and $1.13 per pig for 5:1 mixed groups . A s s u m i n g that growth rates remain the same, extrapolat ion of the data to a market weight of 102 kg resulted in overal l c o s t s of $3.50 per pig for S t resn i l - t reated groups , $1.94 per pig for 3:3 mixed groups and $1.54 per pig for 5:1 mixed groups . iii TABLE OF CONTENTS A b s t r a c t ii LIST OF T A B L E S v LIST OF FIGURES vi A c k n o w l e d g e m e n t s vi i Introduction 1 M e t h o d s 6 Exper imental P rocedures 6 T r e a t m e n t s : 8 O b s e r v a t i o n s : 10 A . A g o n i s t i c Behaviour Immediate ly F o l l o w i n g Mix ing 10 B. A g o n i s t i c Behaviour During Feeding 10 A s s e s s i n g soc ia l status 13 Results 15 A g o n i s t i c behaviour Immediate ly F o l l o w i n g Mix ing 15 A g o n i s t i c Behaviour During Feed ing 15 E f f e c t s of Mix ing on S o c i a l Hierarchies 22 A . Structure of hierarchies 22 B. S o c i a l Status and Weight Ga in 22 E f f e c t s of Mix ing on Weight Ga in 24 E f f e c t s of Mix ing on Feed E f f i c i e n c y 27 E c o n o m i c Impl icat ions of Mix ing 29 D i s c u s s i o n 32 E f f e c t s of Mix ing on Behaviour 32 A g o n i s t i c Behaviour Immediate ly F o l l o w i n g M i x i n g 33 A g o n i s t i c Behaviour During Feed ing 35 Influence of Sex on Behaviour 42 S o c i a l Structure and Status „ 43 S o c i a l Structure 44 S o c i a l Rank and Weight Ga in 45 E f f e c t s of Mix ing on Product iv i ty 46 C o n c l u s i o n s 49 R e c o m m e n d a t i o n s 50 Future Research Needs 54 Literature C i ted 56 A p p e n d i x I 62 iv LIST OF TABLES Table 1; Experimental treatment combinat ions and repl icates 9 Table 2; Behaviours of unmixed, 3:3 mixed , S t resn i l - t reated and 5:1 mixed groups 16 Table 4: A v e r a g e dai ly gain ( A D G ) of unmixed, 3:3 mixed , S t resn i l - t reated , and 5:1 mixed pigs 26 Table 5: Feed per gain rat ios (F/G) of unmixed, 3:3 mixed , S t resn i l - t reated , and 5:1 mixed p igs 28 Table 6: Linear regress ions of weight over t ime 30 v LIST OF FIGURES Figure 1; Des ign and d i m e n s i o n s of g rower pen 7 Tab le 3: Rat ios of response behaviours to init iated behav iours of unmixed , 3:3 mixed , S t r e s n i l - t r e a t e d and 5:1 mixed groups 17 Figure 2: Initiated aggress ion of unmixed , 3:3 m i x e d , S t r e s n i l - t r e a t e d , and 5:1 mixed pigs 19 Figure 3; A g g r e s s i v e r e s p o n s e s of unmixed , 3:3 mixed , S t r e s n i l - t r e a t e d , and 5:1 mixed pigs 19 Figure 4: S u b m i s s i v e r e s p o n s e s o f unmixed , 3:3 m i x e d , S t r e s n i l - t r e a t e d , and 5:1 mixed pigs .. 21 Figure 5: B ites o f unmixed, 3:3 mixed , S t r e s n i l - t r e a t e d , and 5:1 mixed pigs 21 Figure 6: Or ientat ion of feed ing p i g s ' b o d i e s to b lock a c c e s s to feeder ...23 Figure 7: A v e r a g e dai ly gain ( A D G ) of unmixed , 3:3 m i x e d , S t r e s n i l - t r e a t e d , and 5:1 mixed p igs 25 Figure 8: E s t i m a t e d days to market based on linear r e g r e s s i o n s of weight gains 31 vi A C K N O W L E D G E M E N T S I w o u l d like to express my heartfelt thanks to Dr. Dav id Shack le ton , m y superv isor , adv isor , and on o c c a s i o n , m y ass istant . His encouragement , support and advice over the years has meant a great deal to me. M a n y thanks to Dr. Richard Beames for a l lowing me to conduct m y research at the Un ivers i ty of British Co lumbia 's swine unit, to m y fr iend Mr. T h o m a s Riek, manager of the swine unit and to Mrs . Maureen Gar land, coord inator of the Cont inuing Educat ion department. Without their cooperat ion none of this w o u l d have been p o s s i b l e . I w o u l d a lso like to thank Drs. Richard B e a m e s , K imber ley Cheng, and Ruth Newberry and Mr. A l f r e d Wahl , Swine Spec ia l i s t , B C M A F for their construct ive c r i t i c i sms of this thes is . M a n y thanks to all the peop le who have supported m y endeavours over the years ; m y family., f r iends and c o m m i t t e e m e m b e r s , to name a f e w . L a s t but certainly not least, I w o u l d like to express m y gratitude to my fr iend and ass istant , Mr . Robert M c C a n n . His hard work , w o r l d l y knowledge and quick wit made m y research task easier and the work days shorter. I w o u l d like to thank both Agr icul ture Canada and the Brit ish Co lumbia Min is t ry of Agr iculture and F isher ies for prov id ing funding for this project . vii INTRODUCTION Mixing unfamiliar pigs is common practice in many swine operations. Producers would like to maintain uniform pig weights within pens during the growing-finishing stages so that all the animals from a single pen can be marketed simultaneously, thereby optimizing efficiency, minimizing disease transmission, and reducing labour costs. Although the majority of producers try to regroup pigs in uniform weight groups during the early stages of growth, when stress is thought to be less (Jensen ex a/. 1969, cited in Jensen 1971; Hi I Iyer 1972,1976), uniformity of weight at marketing is not often achieved. Currently, the premium paid for pigs of a standard slaughter weight is a monetary incentive that leads to the retention of underweight individuals until the desired market weight is achieved. It is not economical, however, to allow the few retained pigs to remain the sole occupants of a pen after their penmates have been marketed. Conseguently, the producer is faced with either shipping underweight pigs at a reduced profit, or mixing them with other finishing pigs and hoping that they will not suffer setbacks in productivity. Mixing unfamiliar pigs usually results in elevated levels of aggression. The majority of fighting occurs within the first half hour following mixing (Symoens and van den Brande 1969; Moss 1978), but can last anywhere from 1 h (Moss 1978) to 48 h (Ewbank and Meese 1971; Meese and Ewbank 1973). The greatest problem associated with regrouping is not only the increased aggression per se, but also its effect on productivity. Production problems range from a depression in animal performance (Teague and Grifo 1961) in the form of decreased weight gains, decreased feed 1 2 e f f i c i e n c y , decreased f e e d intake, or a c o m b i n a t i o n of these, to injuries and death in extreme c a s e s ( S y m o e n s and Van den Brande 1969; M e e s e and Ewbank 1972; Kel ley et al. 1980). A l though many aspects of p o s t - m i x i n g aggress ion have been studied in pigs weigh ing between 8 kg and 60 kg ( S y m o e n s and van den Brande 1969; Cal lear and van Geste l 1971; Ewbank and M e e s e 1971; M e e s e and Ewbank 1973; Sherritt et al. 1974; Graves et al. 1978; Dantzer and M o r m e d e 1979; A r n o n e and Dantzer 1980; Kel ley et al. 1980; T i n d s l e y and Lean 1984), f e w researchers have examined e f f e c t s on heavier p igs ( M o s s 1978; Hines 1985; M c G l o n e et al. 1986). M o s s (1978) studied aggress ion at the s laughterhouse, Hines (1985) examined mixing at d i f ferent per iods just prior to market ing, and M c G l o n e et al. (1986) studied the use of p h e r o m o n e s as aggress ion modu la tors . For the most part, studies of the relat ionship between aggress ion and animal p e r f o r m a n c e s h o w conf l i c t ing resul ts . S o m e authors ( l e a g u e and Gr i fo 1961; Dantzer 1970, c i ted in Ewbank 1972; Graves et al. 1978) report d e p r e s s i o n s in per fo rmance when indiv iduals are regrouped, whi le others ( M c G l o n e and--Curtis 1981a; Friend et al. 1981) have found no s igni f icant reduct ion in product iv i ty . Var iat ions in the experimental cond i t ions cou ld account for many of these d i sc repanc ies , because mixing is probab ly on ly one factor which a long with other s t r e s s o r s , such as group s ize , space reduct ion, and l imited feed ing , d e p r e s s e s animal per fo rmance (Sherritt et al. 1974; Graves et al. 1978). The re lat ionship between the growth stage of the animal and mixing has a lso been stud ied . J e n s e n and c o - w o r k e r s (1969, c i ted in J e n s e n 1971) 3 felt that age may be a fac tor ; mixing appearing to be less s t ressfu l for young p igs . Others r e c o m m e n d mixing a very large individual with a group of smal ler pigs ( A n o n y m o u s 1974; Stone 1983), assuming that the large individual is not l ikely to attack the smal ler ones , nor be chal lenged by them. T i n d s l e y and Lean (1984) s h o w e d , however , no s igni f icant treatment d i f ference in aggress ion when pigs of unegual weights were mixed as c o m p a r e d wi th those of equal we ights . The concept of soc ia l status and. its relat ionship with aggress ion has interested many researchers (Rasmussen et al. 1962; Ewbank 1969a,1972; Craig 1986; M c G l o n e 1986b). Early researchers such as Rasmussen et al. (1962), descr ibed a stable, linear soc ia l order a m o n g swine ; the general concept of which was in vogue at the t ime. More recent work by Ewbank (1969a) indicates that equal soc ia l status and circular re lat ionships frequently occur , which s e e m s typical in many other spec ies (Richards 1974; Gauthreaux Jr. 1978). A l though the concept of soc ia l status is not wel l understood and the determinat ion of the soc ia l hierarchy is sti l l p lagued with p r o b l e m s (Richards 1974; Craig 1986; M c G l o n e 1986b), high levels of mixing a g g r e s s i o n are suggested to aid its establ ishment (Ewbank 1972) . 'A reduct ion in aggress ion is said to f o l l o w the establ ishment of the soc ia l hierarchy (Beilharz and Cox 1967). The re lat ionship between soc ia l status and product ion traits is a lso unclear, if not contrad ic tory . In s o m e work , no re lat ionships between soc ia l rank and b o d y weight , nor between soc ia l rank and aggress ion have been found (Rasmussen et al. 1962; Ewbank and M e e s e 1971; M e e s e and Ewbank 1973; Fraser 1974), while other reports have shown strong corre lat ions 4 between rank and product iv i ty (McBride et al. 1964; Beilharz and C o x 1967; J a m e s 1967; Ewbank 1972). Hansen (1977, c i ted in Hansen and Hage lso 1980) found that in p igs , growth w a s p o s i t i v e l y related to rank, prov ided the soc ia l hierarchy w a s stable. A s yet , no practical and e f f e c t i v e so lut ion to the p r o b l e m s a s s o c i a t e d with the mixing of pigs has been f o u n d . A wide var iety of methods to reduce aggress ion have been invest igated including odour mask ing c o m p o u n d s (Ewbank and M e e s e 1971; M c G l o n e and Curtis 1981b; M c G l o n e et al. 1981; Friend et al. 1981; M c G l o n e et al. 1986; M e e s e and Baldwin 1975,1977); tranquil izers such as azaperone ( S y m o e n s and Van den Brande 1969; Cal lear and Van Geste l 1971) and other a g g r e s s i o n - r e d u c i n g chemica l addi t ives (Dantzer and M o r m e d e , I979; A r n o n e and Dantzer 1980); phys ica l barriers ( M c G l o n e and Curt is 1985; Fraser 1974); " t o y s " (Ashf ie ld 1984); and mixing during darkness (Stone 1983). These have met with on ly l imited s u c c e s s , and in format ion on techniques for mixing f in ish ing pigs is part icularly lacking. The object ives of this study were to : 1. determine if product iv i ty w a s adverse ly a f f e c t e d by regrouping unfamil iar f in ishing p igs , and if s o , to what extent, 2. examine the e f f e c t s of group c o m p o s i t i o n (sex, weight and di f ferent rat ios of famil iar and unfamil iar p igs) on regrouping, and 3. examine the e f f e c t s of the c o m m e r c i a l swine tranquil izer "Stresn i l " (azaperone) on regrouping. The purpose of this research w a s to prov ide useful behavioural guidel ines for producers when mixing f in ishing pigs and to increase the level of knowledge about agonist ic behaviour f o l l o w i n g mix ing. S ince pigs are not managed as individuals , the f o c u s of this study w a s to examine group rather than individual behaviour. METHODS EXPERIMENTAL PROCEDURES This research w a s conducted at the S p e c i f i c Pathogen Free (S.P.F.) Swine Unit, Department of A n i m a l S c i e n c e , Faculty of Agricultural S c i e n c e s , Un ivers i ty of. Brit ish C o l u m b i a . C r o s s b r e d Yorkshire x Landrace pigs of approx imate ly 76 kg (initial weight) were u s e d . T h e y were housed in groups of six in pens of un i form size and const ruct ion , having concrete block wa l l s and part ial ly s latted, concrete f l o o r s (Figure 1). The pigs were l i m i t - f e d twice dai ly (maximium 3.1 kg per pig per day) with a c o m m e r c i a l grower type ration of a nutrient concentrat ion within the standard dev iat ion boundaries set by the National Research Counc i l for full f e d pigs of 51 kg to 100 kg. Each pen was individual ly f e d , the amount of f e e d recorded , and each p i g - w a s w e i g h e d (to the nearest 0.1 kg) every 3 to 4 days throughout a trial. The light regime w a s standardized automat ica l ly to 11 h light and 13 h dark, with a Tork T i m e S w i t c h (Model 7100). The onset of the light c y c l e at 0730 hours was f o l l o w e d by the morning feed ing at 0745 hours and the a f ternoon feeding at 1430 hours. Short ly after birth, tails were d o c k e d , canine teeth were c l ipped , ears were notched , and males were castrated . Prior to mix ing , each pig w a s individual ly marked on various parts of the body with either number appl icators and black tattoo ink, or with felt pens . Due to the ease and s p e e d of appl icat ion very little s t ress w a s p laced on the an imals . Numbers remained legible for a per iod of 4 to 7 days and per iod ic checks were made by c r o s s referenc ing the painted number with each pig's permanent 6  8 ear notch number. Treatments: T o determine the e f f e c t s of mixing unfamil iar pigs on behaviour and product iv i ty , the experiment c o n s i s t e d of 4 treatments (Table 1 ) . F o l l o w i n g weigh ing and marking, each group in Treatments 1 and 2 w a s held in the weighing r o o m for approximate ly 5 min before being introduced into a c lean, unfamil iar pen. In Treatment 3, each pig w a s injected subcutaneous ly with the manufacturer 's r e c o m m e n d e d d o s e (1 ml per 18.2 kg l iveweight) of 'Stresni l ' (azaperone) prior to mixing ( P i t m a n - M o o r e , M T C Pharmaceut ica ls) . A f t e r a per iod of 15 to 20 min the pigs were s imul taneous ly introduced into a new pen. In Treatment 4, t w o groups of 6 l i t termates each were w e i g h e d , marked, then returned to their respect ive pens. F ive pigs f r o m each litter were then m o v e d to the holding area of the weighing r o o m . A f t e r 5 min these groups of f ive were introduced into the pen contain ing either a s ingle heavy unfamil iar occupant (on average 5.4 kg heavier than the heaviest pig of the introduced group) or a s ingle light pig (on average 5.2 kg lighter than the lightest member of the introduced group. Due to the l imited avai labi l i ty of e m p t y pens , pigs in Treatment 4 cou ld not be mixed into a new pen. Based on results attained by Meese & Ewbank (1973), Hansen et -al. (1982), M c G l o n e (1985) and results f r o m Treatments 1 - 3, it w a s cons idered unnecessary to contro l the sex c o m p o s i t i o n of the groups in Treatment 4. Table 1. Experimental treatment combinations and replicates. Treatment no. Replicates Treatment Combination 1 (unmixed) 3 2 (3:3 mixed) 3 3 (3:3 mixed) 3 (with) (Stresnil) 6 male littermates 6 female littermates 3 male / 3 female littermates 3 male: 3 male mixed group 3 female: 3 female mixed group 3 male; 3 female mixed group 3 male; 3 male mixed group treated with Stresnil 3 female: 3 female mixed group treated with Stresnil 3 male: 3 female mixed group treated with Stresnil 4 (5:1 mixed) 4 - 5 mixed with 1 relatively heavy pig - 5 mixed with 1 relatively light pig 10 Observations; A g o n i s t i c interact ions within unmixed and within mixed treatments were o b s e r v e d and quant i f ied A ) in the f irst 2 h immediate ly f o l l o w i n g mov ing/mix ing and B) during feed ing per iods . A. Agonistic Behaviour Immediately Following Mixing W h e n groups were f i rst m ixed , a cont inuous 2 h v i d e o f i lm w a s made to accurately record the frequent, rapid agonist ic interact ions . The identit ies of the interactors and the occurrences and the duration of the f o l l o w i n g agonist ic interact ions were then measured by rev iewing the v ideotape : i. F ights - a cont inuous aggress ive interaction invo lv ing t w o or more indiv iduals . A fight w a s terminated when the interactors m o v e d more than 1 m apart and both were fac ing away f r o m each other for at least 10 s or when the interactors were spat ia l ly separated by one or more other p igs . i i . Pursuit/Retreat - a sequence invo lv ing aggress ion by at least one individual and retreat by at least one other interactor. Due to equipment p r o b l e m s v i d e o recordings of 9 of the unmixed groups , 4 of the 3:3 mixed groups , and 2 of the 5:1 mixed groups were unobtainable . B. Agonistic Behaviour During Feeding L i m i t - f e e d i n g created a c o m p e t i t i v e s ituation a m o n g the pigs . C o m m e n c i n g the first morning after mix ing, all occurrences of agonist ic behaviour d i s p l a y e d by the group w a s conducted (focal group sampl ing) 11 (Altmann 1974) for the f irst 30 min after f e e d w a s introduced, four t imes a week (Treatments 1 - 3) and twice a week (Treatment 4), for 3 w e e k s . Data co l l ec t ion w a s conf ined to morning feed ing s e s s i o n s to increase internal va l id i ty . The f o l l o w i n g data were co l l ec ted by direct observat ion , recorded with a S o n y T C - 1 1 0 B tape recorder and subsequent ly t ranscr ibed onto computer sheets : the identities of the initiator and the recipient of any agonist ic behaviour pattern (a retal iat ion w a s s c o r e d for any aggress ive behaviour that occurred within 2 s of the prev ious aggress ive act ion); the type of agonist ic behaviour pattern; the t ime at which the agonist ic interact ion occurred; and the o u t c o m e of any compet i t ion for resources (successfu l or unsucessfu l) . Due to the d i f f i cu l ty of a s s e s s i n g the prec ise onset and terminat ion of an interact ion, behaviour patterns rather than interactions were s c o r e d . The f o l l o w i n g agonist ic behaviour patterns were recorded: A g g r e s s i v e Patterns Bites - opening and c l o s i n g the mouth near, or on , part of another's body (Kelley et al. 1980); each individual bite w a s recorded. Tusk ings - a s i d e w a y s , upwards s w e e p i n g b l o w with the snout against the head or the b o d y of the receiver . Equivalent to head knocks (Jensen 1984). Threats - voca l i za t ions , postures , facial or b o d y m o v e m e n t s that either signal aggress ion (McGlone 1986b) or produce a s u b m i s s i v e response in an interactor. N o n - c o n t a c t aggress ive patterns could be d ist inguished f r o m threats by the c o m p l e t i o n of the aggress ive act ion in the former and not in the latter. Submissive Patterns Retreats - movement away from an interactor. Displacements - a lateral shift away, or a departure from, the feed trough in response to aggression from an interactor. Submissive posture - a stance, displayed in response to aggression from an interactor, in which the back is arched, and the tail and ears are lowered (Meese & Ewbank 1973). These specific agonistic patterns were grouped into three functional behavioural categories: Aggressive Initiations - the sum of initiated bites, tuskings and threats. Aggressive Responses - the sum of retaliatory bites, tuskings and threats. Submissive Responses - the sum of retreats, displacements, and submissive postures. In spite of the virtues of using broad categories in describing feeding behaviour, biting behaviour, alone, can be of some value. Biting behaviour can serve as a good indicator of changes in aggression over time because 1) it commonly occurs among unfamiliar animals, and is usually absent among acquainted animals (Fraser 1974); 2) it correlates well with both total aggression and dominance outcomes (McGlone 1986b); 3) it has low variability; and 4) it is easily recognized and therefore repeatable among observers (Kelley et al. 1980). Bites are also "complete" aggressive patterns, and as such are less ambiguous than threats. Analysis of variance (ANOVA) was used, but when not appropriate (eg. due to departures from normality or homoscedasticity), nonparametric tests 13 were appl ied . A probabi l i ty level of a =0.05 w a s se lected a prior/ for hypothes is test ing and c = 0 . 1 0 for S c h e f f e ' s range test because of its conservat ive nature (Winer 1962). ASSESSING SOCIAL STATUS S u b m i s s i v e behaviours and o u t c o m e s of c o m p e t i t i o n for f o o d and water were used to a s s e s s relative soc ia l status. Intrusions (aggress ive and n o n - a g g r e s s i v e attempts to obtain resources) were s c o r e d and recorded as either s u c c e s s f u l or unsuccess fu l depending on the o u t c o m e . D i s p l a c e m e n t s , retreats, s u c c e s s f u l and unsuccess fu l intrusions, and s u b m i s s i v e postures , were used to construct a s o c i o m e t r i c matrix of "winners" and " losers" (A l tmann 1974). Under the contro l led cond i t ions of this s tudy, every animal had an equal probabi l i ty of being o b s e r v e d and therefore the requirements o f the s o c i o m e t r i c matrix out l ined by A l t m a n n (1974) were met . A winner w a s cons idered to be a pig that e l ic i ted a s u b m i s s i v e response f r o m another p ig , or one which s u c c e s s f u l l y intruded between one or more p igs . The o u t c o m e of intrusion attempts were , for the m o s t part, unambiguous; result ing in a clear winner and loser(s). The one except ion w a s the case in which a pig w a s unsuccess fu l in its intrusion attempt between t w o feeding p igs . In this case it could not be determined if the intruding pig w a s unsuccess fu l because of the presence of on ly one , the other, or both of the feed ing p igs . On ly encounters with unambiguous o u t c o m e s were used to calculate the relat ive soc ia l status of each pig within the group. "Dominance Va lues" (DV's), using the arcsine square root t rans fo rmed average proport ion of wins of each pig (0.0 to 1.0) (Beilharz and Cox 1967) were then ca lcu lated f r o m the s o c i o m e t r i c matrix. These DV's were ranked within groups to est imate relat ive soc ia l status. Landau's Index of Linearity (h) w a s calculated f r o m the DV's , and if h>0.9, the soc ia l hierarchy was a s s u m e d to be linear (Lehner 1979). RESULTS AGONISTIC BEHAVIOUR IMMEDIATELY FOLLOWING MIXING Fighting a lmost a l w a y s occurred within a f e w minutes after p igs were introduced to their new pen. Overa l l , 3:3 mixed pigs d i s p l a y e d the highest rate and duration of agonist ic interactions (fighting and pursuit/retreat) per group per minute of observat ion t ime (0.441 /min; 24.8s/min; n = 5 ) f o l l o w e d by the 5:1 mixed groups (0.822/min; 9.3s/min; n = 7 ) , and the St resn i l - t reated groups (0.15/min; 1.9s/min; n = 9 ) with the unmixed l i t termates showing the lowest levels (0.002/min; 0.016s/min; n = 7 ) . AGONISTIC BEHAVIOUR DURING FEEDING No s ign i f i cant d i f fe rences were found in any of the agonis t i c categor ies between the di f ferent sex c o m b i n a t i o n s examined (Aggress ive in it iat ions, H = 0 . 1 , d f = 2 ; A g g r e s s i v e r e s p o n s e s , H = 5 . 2 , d f = 2 ; S u b m i s s i v e r e s p o n s e s , H = 1.3,df=2; B i tes , H = .5,df=2). A g o n i s t i c behaviour d i f fered markedly a m o n g the treatments and w a s quite variable over the 3 weeks f o l l o w i n g mixing (Table 2). A g g r e s s i v e init iations were , by far, the most c o m m o n behaviour patterns; the probabi l i ty of a response behaviour reaching a high of 0.81 in unmixed groups and ranging f r o m a low of 0.07 to a high of 0.35 in mixed groups (Table 3). A g g r e s s i v e init iations did not d i f fer s ign i f i cant ly a m o n g the treatments in the f irst morning f o l l o w i n g mixing (Kruskal -Wal l i s :H =3.3,df =3), but by the end of the f irst week, s igni f icant d i f fe rences were apparent occurred (H = 13.7,df =3) and cont inued throughout the s e c o n d and third w e e k s (H=9.0 ,d f=3, H = 8 . 4 , d f = 3 15 Table 2. Behaviours of unmixed, 3:3 mixed, Stresnil-treated and 5:1 mixed groups. 1 6 Mean number of behaviours per group per 30 mm sample during feeding 8ehavlour unmixed groups (n-9) 3:3 mixed groups (n-9) StresnlI - treated groups (n-9) 3:1 m1xed groups (n-8) Aggressive 1m t fat 1ons Day 1 a 45. 2 63. 6 74. 9 89. 5 week 1 39. 9 b 78. 0= 89. 9= 77. 8° Week 2 46. ,b 72. 9= 76. .1 = 61 . 6B- C Week 3 44 . 9 b 69. 3b.c 70. 4 = 69. 6 B- c Overa11 43 . 6 b 73. 4 = 78. .8= 69. 7 b-. c ;ive responses Day l a 26 3 b a 6 C 15 .6 b- C 13. 4 b' . c Week t 20. .5 12. .3 16. , 1 16. .3 week 2 15 . 4 17 .6 13 .5 14, . 7 Week 3 13. . 6 19. . 7 12. .3 14 , . 1 Overal1 16 .5 16 .6 14 .0 15. .0 i we responses Day t a 9 .8 23 . 1 17 .3 8 .5 Week 1 a .6 b 26 . 1 = 13 8 b- = 8 .5° week 2 9 .7 b - C 25 .6° 7 .0= 5 .3 = Week 3 8 .3 B - C 24 .1 B 7 .7°- = 6 .9= OveraI I 8 .8°- C 25 .2= 9 .5°- = 6 .9 b Aggressive resoonses/AggressWe i n i t i a t i o n s Oay 1 3 .S6 b Week 1 .64 b Week 2 .38° week 3 .31 Overa11 b.c Submissive resDonses/AggressWe Day 1 a Week 1 Week 2 Week 3 Overa11 . 44 1nl11 a11ons .22 .22° .19 B . 14 . 17 .26 .33' .29 b.c b .24" .19= .19= .19= . 18 .23 .33 .29' .28 . 14 . 13 .09 . 1 1 b.c b.c b.c .30 . 11 Oay 1° week 1 Week 2 week 3 Overa1I 8.7' 7.61 9 6.9 7.9' 37.6" 38. 3 C 23.6° 25. 1 C 29. 0 = 32. 4" 35. S C 21.7 C 15.9= 24 . 4 C a F i r s t morning a f t e r day of mlx1ng 6 ' c * d Means with d i f f e r e n t s u p e r s c r i p t s , w i t h i n each row are s i g n i f i c a n t l y d i f f e r e n t (P< 0.05) Tab le 3. Rat ios o f r e spon se behav iou r s to in i t ia ted behav iou r s of unm ixed , 3:3 m i xed , S t r e s n i l - t r e a t e d and 5:1 m ixed g roups . A G G R E S S I V E R E S P O N S E S : A G G R E S S I V E INIT IATIONS S amp l e Unm i xed 3:3 m i xed S t r e s n i l - 5;1 m ixed groups groups t r ea ted groups g roups 1 0.56 0.14 0.24 0.18 2 0.77 0.18 0.14 3 0.81 0.12 023 0.28 4 0.44 0.24 0.18 5 0.38 0.18 0.18 0.28 6 0.38 0.28 0.18 7 0.34 0.24 0.29 0.37 8 0.41 0.32 0.20 9 0.24 0.34 0.20 0.24 10 0.41 0.29 0.15 11 0.29 0.34 0.21 0.33 12 0.31 0.35 0.22 *• Four s amp l e s in each week S U B M I S S I V E R E S P O N S E S : A G G R E S S I V E INITIATIONS S amp l e Unm i xed 3:3 m i xed S t r e s n i l - 5:1 m ixed groups groups t rea ted groups groups 1 0.22 0.32 0.23 0.14 2 0.23 0.28 0.16 3 0.24 0.28 0.18 0.12 4 0.20 0.33 0.17 5 0.16 0.30 0.10 0.09 6 0.22 0.31 0.10 7 0.21 0.34 0.07 0.08 8 0.18 0.30 0.11 9 0.17 0.29 0.13 0.11 10 0.15 0.27 0.09 -. 11 0.19 0.31 0.11 0.10 12 0.17 0.32 0.10 Four s amp l e s in each week 18 r e s p e c t i v e l y ) (Table 2). Over the entire 3 week per iod , the d i f fe rences in aggress ive init iations were s igni f icant (H = 11.0 df =3). P igs in unmixed groups exhibited fewer aggress ive init iations than those in the other treatments throughout the 3 week per iod (Table 2). With the except ion of the f irst day f o l l o w i n g mix ing, pigs treated with Stresni l s h o w e d the highest average w e e k l y and overa l l , levels of initiated aggress ion (Table 2), but there w a s a not iceable dec l ine in the aggress ion they and the other mixed treatments init iated over t ime (Figure 2). In spite of this decl ine by the third week, the aggress ion init iated in the mixed treatments was sti l l higher than that in unmixed groups . The mean number of aggress ive r e s p o n s e s did not change after the s e c o n d or third day p o s t - m i x i n g (Figure 3). No s igni f icant d i f fe rences between the treatment groups were found other than those o b s e r v e d on the f irst day ( H = 9 . 4 , d f = 3 ; Tab le 2). On the f irst day f o l l o w i n g mixing the unmixed groups d i s p l a y e d s ign i f i cant ly more aggress ive r e s p o n s e s per aggress ive init iations than the 3:3 mixed group ( M a n n - W h i t n e y : U =8.5). The 3;3 mixed group exhibited the f e w e s t aggress ive r e s p o n s e s on the f irst day and to the end of the f irst week (Table 2). The rat ios of aggress ive r e s p o n s e s to aggress ive init iations s h o w e d s imi lar trends to those of the mean number of aggress ive r e s p o n s e s in all the treatment groups except for the 5:1 mixed treatment. The ratio of aggress ive r e s p o n s e s to aggress ive init iations of 5:1 mixed treatment s h o w e d an increase through the f irst and s e c o n d w e e k s (Table 2). The treatment groups d i f fe red s ign i f i cant ly in their levels of s u b m i s s i v e behaviour over the 3 week per iod (H =8.3,df = 3 ; Table 2). 120-, Legend • UNMIXED GROUPS (n=9) 3:3 MIXED GROUPS (n=9) STRESNIL GROUPS (n=9) O 5:1 MIXED GROUPS (n=3) DAYS POST MIXING gure 2. Initiated aggress ion o f u n m i x e d , 3:3 mixed , S t r e s n i l - t r e a t e d , and 5:1 m i x e d p i g s . 30-i 25- 20-I 15- 10- 5 - it / / i / 'A* I Legend • 1 UNMIXED GROUPS (n=9) E23 3:3 MIXED GROUPS (n=9) STRESNIL GROUPS (n=9) O 5:1 MIXED GROUPS (n=3) 0 7 14 DAYS POST MIXING lure 3. A g g r e s s i v e r e s p o n s e s of u n m i x e d , 3 : 3 mixed S t r e s n i l - t r e a t e d , and 5:1 m i x e d p i g s . 21 20 A l though there were no s igni f icant d i f fe rences o b s e r v e d on the f irst day f o l l o w i n g mixing (H =4.5,df =3), by the end of the f irst week d i f fe rences were s igni f icant (H =8.2,df =3), and remained so throughout the s e c o n d (H = 10.7,df =3) and third w e e k s (H =7.7,df =3). Wi th the except ion of the S t resn i l - t reated groups, the rat ios of s u b m i s s i v e r e s p o n s e s to initiated aggress ion remained fair ly constant throughout the 3 week per iod (Table 2). The 3;3 mixed group d i s p l a y e d the highest inc idence of s u b m i s s i v e behaviours (Figure 4), and by the end of the third week, their level sti l l had not dec l ined to those of the other groups . The 5;1 mixed groups exhibited the lowest level of s u b m i s s i v e behaviour , although they were not s tat is t ica l ly d i f ferent f r o m those for p igs in the unmixed groups . Bit ing behaviour d i f fered s ign i f i cant ly between treatment groups in every t ime per iod tested (Table 2). In the f irst day, s igni f icant d i f fe rences were evident between the unmixed and 3:3 mixed groups (U = 10.0), and between the unmixed and St resn i l - t reated groups (U = 13.5). These d i f fe rences pers is ted throughout the 3 w e e k s . The leve ls of biting in the 5:1 mixed groups were intermediate. P igs in the 3:3 mixed groups d i s p l a y e d the highest inc idence of bit ing behaviour, but a dist inct peak in the mean number of bites w a s o b s e r v e d in the S t resn i l - t reated groups on the s e c o n d day (Figure 5). L e g e n d • i UNMIXED GROUPS (n =9) ZZ1 3:3 MIXED GROUPS (" =9) STRESNIL GROUPS (n =9) a 5:1 MIXED GROUPS (n= =8) DAYS POST MIXING . S u b m i s s i v e r e s p o n s e s of unmixed 3 - 3 m ixed S t r e s n i l - t r e a t e d , and 5 : 1 m ixed pig's. L e g e n d • • UNMIXED GROUPS (n =9) EZ3 3:3 MIXED GROUPS (n =9) m STRESNIL GROUPS (n =9) O 5:1 MIXED GROUPS (n= 8) DAYS POST MIXING m i x e d ° p i 9 U s n n n i X e d - m i X e d - S M " " ' - < " » « • « . - d 6:1 22 EFFECTS OF MIXING ON SOCIAL HIERARCHIES A. Structure of hierarchies The soc ia l hierarchies of the unmixed and mixed groups were quite di f ferent with respect to structure. In 78% (n=9) of the unmixed groups the hierarchy w a s non- l inear , whi le in the 3:3 mixed and Stresni l t reatments , on ly 22% (n=9) and 56% (n=9) of the groups exhibited non- l inear i ty , r e s p e c t i v e l y . In the 5:1 mixed groups , non- l inear soc ia l hierarchies were found in 88% of the pens (n=8). A complete hierarchy could not be const ructed for one group because s o m e animals did not interact with each other during the o b s e r v a t i o n per iods . The relative soc ia l status of the s ingle original occupant w a s not cons is tent ; the s ingle individuals occup ied either an intermediate pos i t ion (no. 2 no . 5) in the hierarchy (75%; n = 6), or were the m o s t subordinate (25%). B. Social Status and Weight Gain Feeding pigs were of ten o b s e r v e d using their their bod ies to block a c c e s s to f o o d by n o n - f e e d i n g individuals (Figure 6). The n o n - f e e d i n g individuals were c o m m o n l y animals of low soc ia l status, but in spite of the di f ferent ia l abi l ity of pigs to obtain f o o d , on ly the 3:3 mixed groups s h o w e d a s igni f icant corre lat ion between weight gain and relative soc ia l status, and on ly during the f irst week (Kendall:r =-.25,df =52). The magnitude of the d i f f e r e n c e s in weight gain was such that over 3 weeks the corre lat ion w a s sti l l s igni f icant (r =-.21 df =52). 23 Figure 6. Orientation of feeding pigs' bodies to block access to feeder. 24 EFFECTS OF MIXING ON WEIGHT GAIN C o n s i s t e n t l y , unmixed groups had a higher average dai ly weight gain ( A D G ) than any other treatment groups. Over the 3 week s a m p l e per iod d i f fe rences in A D G a m o n g the treatment groups were s igni f icant (F =3.73,df =3) because of the large d i f f e r e n c e s during the f irst 3 days (F=5.56,df =3) and throughout the f irst w e e k (F =3.30 df =3). By the s e c o n d and third w e e k s , these d i f fe rences were no longer s igni f icant (F = 1.57,df=3 and F =0.92,df =3). Neither the initial start ing weights (i.e. using the initial starting weight as a covar iate ; F = 1.12,df = 1), nor the sex c o m b i n a t i o n s used (F=0.82,df =2), had s igni f icant e f f e c t s on A D G . Over the three week per iod there w a s a not iceable change in the pattern of A D G a m o n g treatments . A l t h o u g h the A D G of 3:3 mixed groups were initial ly much lower than that of unmixed groups, by the s e c o n d week they had reached a level comparab le to that of the unmixed groups (Figure 7). The other treatments were intermediate to the unmixed and 3:3 mixed groups and s h o w e d s imi lar t rends. A v e r a g e d over the entire 3 w e e k s , the greatest d i f fe rences occurred betweeen unmixed pigs and S t resn i l - t reated and 5:1 mixed groups , indicating that whi le the 3:3 mixed group had p r o g r e s s i v e l y better A D G over t ime, the S t resn i l - t reated and 5:1 mixed groups did not grow as wel l (Table 4). In the 5:1 mixed group treatment, no s igni f icant d i f fe rence in weight gain w a s found between light weight and heavy weight original occupants in any of the weeks tested (first w e e k , F=0.01,df = 1; s e c o n d week, F=0.03,df = 1; third week, F=0.16,df = 1) nor over the 3 w e e k s (F=0.05,df = 1). When the A D G ' s of the s ingle heavy p igs were c o m p a r e d to the A D G ' s of Legend m UNMIXED GROUPS (n =9) EZ3 3:3 MIXED GROUPS (n =9) S 3 STRESNIL GROUPS (n =9) O 5:1 MIXED GROUPS (n= •8) 10 14 DAYS POST MIXING a g e d a i l y g a i n ( A D G ) o f u n m i x e d , 3:3 m i x e d , l - t r e a t e d , a n d 5:1 m i x e d p i g s . Table 4. Average daily gain (ADG) of unmixed, 3:3 mixed, Stresnil-treated, and 5;1 mixed pigs. x (+ S.E.) Average D a i l y Weight Gain (kgday ) 3:3 mixed S t r e s n l l - 5:1 mixed Time p e r i o d unmixed groups groups treated groups groups (n=9) (n=9) (n=9) (n=8) F i r s t 3 days 0. .72 + 0 37 a 0. . 17 + 0. 35 b 0. 31 + 0. 21 3' .b 0. 55 0. 2 9 a - b Week 1 0. 83 + 0. 14 a 0. 62 + 0. 17 b 0. 63 + 0. 18 3' .b 0. 76 0. 2 0 a ' b Week 2 0. .96 + 0. 14 0. 93 0. 1 1 0. 90 + 0. 14 0. 82 + 0. 15 Week 3 0. 88 + 0. IB 0. 88 + 0. 13 0. 81 0. 14 0. 78 + 0. 13 Overa11 0. 89 + 0. 07 a 0. 81 0. 0 8 a ' b 0. .78 + 0. 08 b 0. 79 + 0. 0 9 b Means with d i f f e r e n t s u p e r s c r i p t s , within each row are s i g n i f i c a n t l y d i f f e r e n t (P< 0.05) 27 the other individuals in the group, no s igni f icant d i f fe rence w a s found (first week, F=2.53,df = 1; s e c o n d w e e k , F=0.53,df = 1; third w e e k , F =0.39,df = 1; over the 3 w e e k s , F=0.30,df = 1). S imi la r ly , the A D G ' s of the original light weight pigs were not s ign i f i cant ly d i f ferent f r o m those of the other group m e m b e r s (first week, F=3.75,df = 1; s e c o n d week, F = 1.41 ,df = 1; third week, F=0.01,df = 1; over the 3 w e e k s , F=2.24,df = 1) EFFECTS OF MIXING ON FEED EFFICIENCY Trends s imi lar to those of A D G ' s were found a m o n g the f e e d per gain ratios (F/G) of treatment groups. S igni f icant d i f fe rences were apparent in the f irst w e e k , but not in the s e c o n d or third w e e k s (F =3.83,df = 3 ; F = 1.45,df=3 and F = 0 . 4 1 , d f = 3 , respect ive ly) , although averaged over the 3 week sample per iod , the F/G's were s igni f icant (F =4.22,df =3). Sex had no s igni f icant e f fec t on F/G (F =3.12,df =2) over the entire 3 w e e k s . Throughout the exper iment , p igs in unmixed groups were general ly the m o s t e f f i c ient in convert ing f o o d to weight gain, and S t resn i l - t reated pigs were the m o s t ineff ic ient converters (Table 5). Within 5:1 mixed groups no s igni f icant d i f fe rence in the F/G w a s found between groups with s ingle unfamil iar heavy individuals and those with s ingle unfamil iar light weight pigs (F=0.26,df = 1). Table 5. Feed per Gain Ratios (F/G) of unmixed, 3:3 mixed, Stresnil-treated, and 5:1 mixed pigs. x (± S.E.) Feed per Galnlkg kg ) 3:3 mixed S t r e s n i l - 5:1 mixed Time p e r i o d unmixed groups groups treated groups groups (n=9) (n=9) (n=9) (n*8) Week 1 3.47 ± 0.51 a 4.63 + 1.09 a' b 4.74 ± 1 . 3S b 3.72 ± 0.70 a , t > Week 2 3.15 ± 0.43 3.25 + 0.38 3.39 ± 0.52 3.64 ± 0.68 Week 3 3.69 + 0.95 3.69 ± 0.65 4.03 + 0.74 3.90 ± 0.79 Overall 3.35 ± 0.28 3 3.74 + 0.36 b 3.85 ± 0.31 b 3.65 ± 0 . 2 8 3 - b ** a ' b Means with d i f f e r e n t s u p e r s c r i p t s , w i t h i n each row are s i g n i f i c a n t l y d i f f e r e n t (P< 0.05) 29 ECONOMIC IMPLICATIONS OF MIXING Based on linear regress ions of weight gain (Table 6), the est imated days to a c o m m o n final weight of 95 kg f r o m an initial l iveweight of 76 kg were calculated as f o l l o w s ; unmixed treatment, 21.08 d a y s ; 3:3 mixed, 22.92 d a y s ; S t resn i l - t rea ted , 23.63 d a y s ; and 5:1 mixed, 23.68 d a y s (Figure 8). Extrapolat ing further to a market weight of 102 kg, the es t imated addit ional days to market were 7.62 days for the unmixed treatment, 7.66 days for the 3:3 mixed , 8.14 days for S t resn i l - t rea ted , and 8.59 days for 5:1 mixed pigs (Figure 8). Based on these project ions and 1987 est imates of f e e d c o s t s , pen s p a c e , and drug c o s t , the added c o s t s a s s o c i a t e d with mixing were $1.43/pig (to 95 kg), $1.94/pig (to 102 kg) for the 3:3 mixed; $2.92/pig (to 95 kg), $3.50/pig (to 102 kg) for Stresni l t reated; and $1.13/pig (to 95 kg), $1.54/pig (to 102 "kg) for 5:1 mixed (see A p p e n d i x 1 for ca lculat ions) . S ince f e e d cost w a s the largest variable cos t it had a much greater inf luence on total cost than days to a c o m m o n final weight . Table 6. Linear regressions of weight (W;kg) over time (t:days). Treatment Regression equation Unmixed groups W = 0.91823 t + 75.645 3:3 mixed groups W = 0.91414 t + 74.050 Stresnil-treated groups W = 0.86018 t + 74.674 5:1 mixed groups W = 0.81500 t + 75.698 31 Figure 8. Estimated days to market based on linear regressions of weight gains (Regression lines have been adjusted to common initial and final weights). DISCUSSION Mixing pigs is a management procedure that has not been wel l s tudied, yet has pro found e f f e c t s on behaviour and product iv i ty . A l though behaviour and product iv i ty may appear to be unrelated, they are, in fact , c l o s e l y l inked; changes in behaviour o f ten being re f lec ted by d e p r e s s i o n s in product iv i ty . A n understanding of the .effects of mixing on behaviour is therefore essent ia l to opt imiz ing product iv i ty . EFFECTS OF MIXING ON BEHAVIOUR M o s t authors recognize two di f ferent phases of aggress ion when pigs are mixed; 1) immediate p o s t - m i x i n g a g g r e s s i o n , and 2) aggress ion d i s p l a y e d in c o m p e t i t i o n for l imit ing resources such as f o o d (Fraser 1974; A r n o n e 1979, c i ted in A r n o n e and Dantzer 1980; Dantzer and M o r m e d e 1979; A r n o n e and Dantzer 1980). Initially, the lack of fami l iar i ty between the animals (Fraser 1974) and the disrupt ion of previous soc ia l hierarchies inf luence a g g r e s s i o n , but after fami l iar i ty has been ach ieved , l imited a c c e s s to the resource is the dominant factor . These two phases of aggress ion are by no means independent, the o u t c o m e s of immediate p o s t - m i x i n g agonist ic interact ions probably greatly inf luencing an individual 's abi l i ty to obtain f o o d resources later. 32 33 Agonistic Behaviour Immediately Following Mixing When unfamil iar animals were grouped together, a s igni f icant increase in aggress ion occurred . In contrast to the agonist ic interactions during f e e d i n g which rarely invo lved more than one aggress ive pattern, interactions during the initial mixing per iod resembled those descr ibed for w i ld pigs (Fradrich 1974; J e n s e n and W o o d - G u s h 1984; Barrette, 1986); p ro longed with many patterns p e r f o r m e d . It has been suggested that the increase in aggress ion is necessary for the fo rmat ion of the new soc ia l hierachy ( S y m o e n s and van den Brande 1969; Ewbank and M e e s e 1971; M e e s e and Ewbank 1973; M c G l o n e 1986a). Whi le durations of f ight ing have been reported to range f r o m 8 h ( S y m o e n s and van den Brande 1969) to 48 h (Ewbank and M e e s e 1971; M e e s e and Ewbank 1973; M c G l o n e 1986a), m o s t of the intensive f ighting in this study c e a s e d within 24 h. On ly o c c a s i o n a l l y w a s f ight ing o b s e r v e d in the morning f o l l o w i n g mix ing. Not surpr is ingly the greatest amount of immediate p o s t - m i x i n g aggress ion w a s o b s e r v e d in the 3:3 mixed groups, probably because the largest number of unfamil iar animals were present in this treatment. Within each 3:3 group, there was a total of nine poss ib le dyadic interact ions between unfamil iar an imals . This is in contrast to the 5:1 mixed treatment which d i s p l a y e d the s e c o n d highest level of immediate p o s t - m i x i n g aggress ion and where there were only f ive p o s s i b l e dyadic interact ions between unfamil iar an imals . A m o n g the unmixed groups, no f ight ing occurred and although single aggress ive behaviour patterns were exchanged, they were u n c o m m o n . A f t e r a per iod of intense invest igatory act iv i ty of their new pen, the unmixed animals general ly lay down and slept. There are two 34 o b v i o u s c o n c l u s i o n s f r o m these f ind ings . First, unfami l iar i ty between group m e m b e r s p r o m o t e s agonist ic interact ions and s e c o n d , greater numbers of unfamil iar animals result in increased f ight ing. A n i m a l s sedated with the tranquil izer Stresni l exhibited the least amount of aggress ion of all the mixed groups during the immediate p o s t - m i x i n g per iod , most l ikely because of the propert ies of the drug (retarding coord inat ion and inducing s leep) . In the S t resn i l - t reated groups however , sporad ic f ighting w a s o b s e r v e d during the f irst and s e c o n d morning's feed ing per iods whereas f ight ing had ceased in m o s t of the other treatment groups by these t imes . Reports of S t resn i l - t reated animals f ight ing f o l l o w i n g their r e c o v e r y f r o m the sedat ive e f f e c t s of the drug (often as much as undrugged an imals) have a lso been made by other researchers ( M a r s b o o m 1969; S y m o e n s and van den Brande 1969). G i v e n the fast act ing nature of the drug, the sedated pigs w o u l d be expected to d isp lay even less aggress ion than the ful ly c o n s c i o u s unmixed groups , but there were probably two reasons w h y this did not occur . F irst , Stresni l appeared to a f fect pigs d i f ferent ia l ly , producing deep sedat ion in s o m e pigs whi le on ly d r o w s i n e s s in others . S e c o n d , the sedated animals had to be awakened in order to m o v e them into the new pen. The or iginal pen occupant in the 5:1 mixed treatment d i s p l a y e d very interesting behaviour . A l though o b v i o u s l y outnumbered by unfamil iar group m e m b e r s , the original occupant initiated the f irst bouts of aggress ion in every trial . W i l d p igs are not known to be territorial (Matschke and Hardister 1966; Kurz and Marchinton 1972; W o o d and Brenneman 1980; Singer et al. 1981; T i sde l l 1982), but the behaviour of the original pen occupant w o u l d s e e m to suggest that under conf ined c o n d i t i o n s , prior ownersh ip of space may inf luence a g g r e s s i o n , or individual d is tance may increase under s t ress . C o n v e r s e l y , aggress ive tendenc ies of the introduced pigs cou ld have been inhibited by a c o m b i n a t i o n of the s t ress of m o v i n g to a new pen, and being f a c e d with an occupant . Th is inhibit ion of aggress ive tendenc ies is unl ikely , however , because in other mixed treatments new surroundings and new group mates were a s s o c i a t e d with an increase in agonist ic interact ions. A l though pr ior i ty of ownersh ip may encourage original pen occupants to d i sp lay more aggress ion than they otherwise might, the s ingle pigs were not a l w a y s s u c c e s s f u l in their agonist ic encounters with introduced group m e m b e r s . Agonist ic Behaviour During Feeding A g o n i s t i c behaviour at the feeder w a s probably the most important aspect of this study because of its direct re lat ionship to product iv i ty . The measurement of this type of behaviour is not without its d i f f i cu l t ies , however . In the past , many researchers have fa i led to adequately def ine their te rmino logy when descr ib ing behaviours (Fraser and Rushen 1987). A l s o , the absence of a standardized set of behaviour patterns makes c o m p a r i s o n between studies d i f f icult and f ina l ly , many individual behaviour patterns have extreme var iances . The data f r o m this study and that of M e e s e and Ewbank (1972) indicates that levels of individual behaviours o f ten change drast ica l ly f r o m day to day. In light of these p r o b l e m s , the use of broad funct ional categor ies is o f ten more useful in descr ib ing feed ing aggress ion than discrete units of behaviour. For this study, the behavioural categor ies , aggress ive init iat ions, aggress ive r e s p o n s e s , and s u b m i s s i v e r e s p o n s e s were chosen and are, in fact , s y n o n y m o u s with the three p h y s i o l o g i c a l mechan isms of behaviour recogn ized by A d a m s (1979) as o f f e n s i v e attack, de fens ive attack, and s u b m i s s i o n . The inf luence of the m o v i n g p r o c e s s on establ ished groups of p igs has not been c l o s e l y examined in feed ing aggress ion studies , although it is not general ly cons idered to be s t ress fu l . M y data and those of Mardarowicz (1985),. indicate that m o v i n g alone does s t ress the an imals . This stress w a s ref lected in the behaviour of the unmixed pigs f o l l o w i n g their introduct ion into a new p e n . In the f irst week , the unmixed pigs exhibited more aggress ive r e s p o n s e s than in subsequent w e e k s . This w a s m o s t not iceable in the f irst day , and in fact , the unmixed groups exhibited the highest level of aggress ive responses of any treatment during this per iod . Interestingly enough, the p r o c e s s of m o v i n g did not increase an animal 's tendency to initiate a g g r e s s i o n nor to bite. Rather, it appeared to reduce an animal 's tolerance of , or lack of overt react ion to , any initiated aggress ion . S u b m i s s i v e behaviour , in contrast , did not f luctuate greatly over t ime, which may not be surpris ing if, as has been suggested , the soc ia l hierarchy is maintained pr imar i ly through the behaviour of the subordinates (Col l ias 1944; Rowel l 1966; J e n s e n and W o o d - G u s h 1982; M c C o r t and Graves 1982; J e n s e n 1982,1984). P r e s u m a b l y , the soc ia l hierarchy of unmixed groups had been estab l i shed prior to the animals being m o v e d into the new pen. In contrast to the unmixed groups, the p rocess of mixing unfamil iar pigs resulted in several behavioural changes during feed ing . Bit ing behaviour probably best i l lustrates the extreme stress a s s o c i a t e d with mix ing. A s o b s e r v e d in immediate p o s t - m i x i n g a g g r e s s i o n , bit ing behaviour appeared to be related to the number of unfamil iar animals in the treatment. The average number of bites w a s highest in the 3:3 mixed treatment f o l l o w e d by the S t resn i l - t reated and 5:1 mixed treatments . A l though it dec l ined steadi ly , by the end of 3 w e e k s the numbers of bites in all the mixed groups were sti l l more than twice that of unmixed groups. Whi le this suggests that aggress ion d o e s decrease over t ime as fami l iar izat ion occurs , the p r o c e s s is s l o w . The presence of stress in the mixed treatments w a s a lso evident f r o m the e levated levels of init iated aggress ion . The mean number of aggress ive init iations far exceeded the levels of either aggress ive r e s p o n s e s or s u b m i s s i v e r e s p o n s e s ; as much as 14 t imes in s o m e mixed groups . In unmixed groups there w a s a higher f requency of retal iat ions and zero r e s p o n s e s . The except iona l ly low levels of init iated aggress ive behaviour in the 3:3 mixed groups , and to a lesser extent, the St resn i l - t reated groups in the first morning f o l l o w i n g regrouping were interesting and s o m e w h a t unexpected. A l though a cont inuat ion of the high levels of aggress ion d i s p l a y e d in the prev ious day w o u l d have been expected , the apparent lull in aggress ion can be explained a posteriori. There were l ikely t w o fac tors depress ing aggress ion in the first morning f o l l o w i n g mix ing; fat igue, and newly acquired soc ia l status. Many of the pigs appeared to be exhausted f r o m the e f f e c t s of immediate p o s t - m i x i n g f ight ing as only a f e w pigs were present at the trough during the feed ing per iod . In addi t ion , s o m e of- the pigs that attempted to gain a c c e s s to the feeder were aggress ive ly 38 repulsed, and after a f e w attempts lay d o w n again, leaving only a f e w pigs at the feeder . In the 5:1 mixed treatment no decrease in initiated aggress ion w a s o b s e r v e d in the f irst morn ing f o l l o w i n g regrouping. A s there w a s on ly one unfamil iar m e m b e r in the 5:1 treatment and f e w intra-l i tter agonist ic encounters occurred , fatigue w a s not a major factor in this case , at least a m o n g the f i ve l i t termates. The combinat ion of m o v i n g s t ress , more pigs at the feeder , and the intense aggress ion by the s ingle original occupant , a lso probably contr ibuted to the high levels of aggress ion at this t ime. S ince Stresni l is usual ly v o i d e d f r o m the s y s t e m within 24 h ( M a r s b o o m 1969; Porter and S lusser 1984), the behaviour of the drugged animals should have been s imi lar to that of the 3:3 mixed groups f o l l o w i n g the p o s t - m i x i n g per iod , but this w a s not the case . S t resn i l - t reated groups d i sp layed the highest level o f initiated aggress ive behaviour in every t ime per iod tes ted . There w a s a lso an extremely high peak of init iated aggress ion during the s e c o n d morning f o l l o w i n g mixing. During the f i rst and s e c o n d morn ings f o l l o w i n g mixing agonist ic interact ions at the feeder of ten esca lated into severe f ight ing in the middle of the pen . The c o m p e t i t i o n at the feeder therefore served as the catalyst or st imulus for more intense and p r o l o n g e d agonist ic interact ions; interact ions that were more typ ica l l y seen immed ia te ly f o l l o w i n g mix ing. These f ind ings indicate that rather than e l iminat ing f ight ing as c la imed by the manufacturer, M T C Pharmaceut ica ls , Stresni l d e l a y s the onset of aggress ion and may therefore retard the establ ishment of the soc ia l hierarchy. The s u b m i s s i v e r e s p o n s e s of the Stresni l treatment a lso d i f fe red f r o m 39 that of the 3:3 mixed treatment, again indicating that the drug probably has p ro longed e f f e c t s on the p igs ' behavioural repetoire . A f t e r the f irst week , Stresni l treated groups d i sp layed a dramatic decl ine in s u b m i s s i v e r e s p o n s e s that cannot be readi ly expla ined. The initial high levels of s u b m i s s i v e behaviour during the f irst week can probably be attributed to a combinat ion of high leve ls of aggress ive init iations and low levels of aggress ive r e s p o n s e s . With aggress ive init iations and aggress ive r e s p o n s e s on ly dec l in ing s l o w l y , however , no reason can be found for the sudden decrease in s u b m i s s i v e behaviour. The patterns of initiated aggress ion and s u b m i s s i v e r e s p o n s e s in the 5:1 mixed treatment s h o w e d t w o peaks ; the s e c o n d smal ler than the f irst but neverthe less , d ist inct . This is probab ly due to two fac tors ; mixing animals and introducing animals into a pen that is a l ready o c c u p i e d . Initially the s ingle animal probably exhibited aggress ion due to its fami l iar i ty with the surroundings. The introduced group a lso exhibited a g g r e s s i o n , f irst because of the stress a s s o c i a t e d with m o v i n g , and s e c o n d because of the presence and behaviour of an unfamil iar individual . F o l l o w i n g the defeat of the s ingle individual by the members of the l ittermate group, agonist ic behaviour dec l ined . A s fami l iar izat ion among the animals occurred, h o w e v e r , agonist ic behaviour increased once more as the s ingle individual sought to establ ish its pos i t ion within the hierarchy. The extremely low leve ls of s u b m i s s i v e r e s p o n s e s , even in c o m p a r i s o n to the unmixed treatment, indicates that because the s ingle pig has the advantage of being fami l iar with the surroundings and the group has the advantage of numbers , neither are l ikely to act s u b m i s s i v e l y . Perhaps the most interesting trend d i s p l a y e d w a s that of aggress ive r e s p o n s e s . During the f irst day after mix ing , pigs in 3:3 mixed groups were the least l ikely to respond a g g r e s s i v e l y , probab ly because of the newness of the soc ia l hierarchy and the f ierce c o m p e t i t i o n for the highest soc ia l ranks. Not only were retal iat ions by low ranking m e m b e r s of the group met with overt aggress ion (e.g. bites), but a lso subordinate animals were re lent less ly pursued. A s fami l iar izat ion occurred and pursuit of low ranking m e m b e r s dec l ined , aggress ive r e s p o n s e s increased. This increase is interest ing, because in c o m b i n a t i o n with the decrease in initiated a g g r e s s i o n , it could indicate a general increase in the soc ia l to lerance probably approaching that found in unmixed groups. Increases in soc ia l to lerance w o u l d m o s t probably occur on ly with increasing fami l iar i ty a m o n g group m e m b e r s and stabi l i ty in soc ia l orders (Col l ias 1953). The trend of aggress ive r e s p o n s e s in the Stresni l treatment after the post mixing per iod is s o m e w h a t puzzl ing. A s o p p o s e d to the increase seen in the 3:3 mixed treatment, a gradual decl ine in aggress ive r e s p o n s e s w a s o b s e r v e d in the S t resn i l - t reated pigs over the 3 w e e k s . Relative to init iated a g g r e s s i o n , aggress ive r e s p o n s e s did not change over the 3 w e e k s . The lack of an increase in the trend of aggress ive r e s p o n s e s may have occurred for a number of r e a s o n s , all related to Stresni l having di f ferent ia l e f f e c t s on -the p igs . The f irst is that the drug may have excited s o m e (aggress ive) individuals whi le sedat ing others (less aggress ive animals) . If this were the case , then the aggress ive animals w o u l d be advantaged f r o m the outset ; their e levated aggress ion re inforc ing the learning p r o c e s s and inhibiting the aggress ive r e s p o n s e s of the less aggress ive p igs . The s e c o n d poss ib l i l i t y is that the drug had its greatest e f fect on the less aggress ive an imals . Dominant animals are be l ieved to be more aggress ive than subordinate ones (Ewbank and M e e s e 1971; Hansen et al. 1982), and this attribute cou ld have raised the dominants ' to lerance of the drug to a point where the drug's e f f e c t i v e threshold w a s not reached. Consequent ly , the full or partial behavioural repertoire of the aggress ive animals w o u l d remain intact whi le the subordinates w o u l d be unable to respond a g g r e s s i v e l y . The third poss ib i l i t y is that the drug had a di f ferent ia l e f fec t on the type of aggress ion d i s p l a y e d . O f f e n s i v e aggress ion (init iations) and d e f e n s i v e aggress ion ( responses) are thought to be contro l led by di f ferent neural m e c h a n i s m s (Brown 1970), s o if the tranquil izer acted preferent ia l ly on one pathway, a d isrupt ion of the c o m p l e t e behavioural repertoire cou ld result. Drugs such as lithium carbonate are known to decrease intra-group aggress ion a m o n g feeding pigs whi le being inef fect ive in prevent ing aggress ion in newly regrouped pigs (Dantzer and M o r m e d e 1979). S imi la r ly , amphetamines have been s h o w n to a f fec t dominant squirrel m o n k e y s d i f ferent ly f r o m subordinate m o n k e y s (Miczek and G o l d , 1983; Miczek et al. 1984) and other drugs have di f ferent ia l e f f e c t s on -the behaviour of aggress ive and t imid mice (Krsiak et al. 1984). Clear ly further research is n e c e s s a r y to determine the full e f f e c t s of Stresni l on pig behaviour. The trend of aggress ive r e s p o n s e s in 5:1 mixed groups w a s a lso di f ferent f r o m either the 3:3 mixed and St resn i l - t reated groups. Rather than f o l l o w i n g the b imoda l d istr ibut ions of aggress ive init iations and s u b m i s s i v e r e s p o n s e s , aggress ive r e s p o n s e s in 5:1 mixed groups s l o w l y dec l ined. Whi le at f irst this might suggest increasing fami l iar i ty , this d o e s not necessar i ly 42 have to be the case. Although aggressive responses declined absolutely over the three weeks, relative to aggressive initiations they increased slightly during the second week. A similar trend was observed in the 3:3 mixed groups during the same time period, but in the third week aggressive responses in the 3;3 mixed group continued to increase while they decreased in the 5:1 mixed groups. Initially a single pig is likely to initiate aggression as a means of defending its familiar space at the feeder. However, it is unlikely to continue this behaviour for a prolonged period of. time when confronted by five unfamiliar animals. Influence of Sex on Behaviour In all the treatments in this study, no differences were found in agonistic behaviour between the sexes. While similar observations have been made by some (Meese and Ewbank 1973; Hansen et al. 1982; McGlone 1985b), others have suggested that barrows are more aggressive (McBride et al. 1964; Beilharz and Cox 1967; Gallwey and Tarrant 1978, cited in Stephens 1980). The lack of sex differences was not unexpected since the average starting age of the pigs in this experiment was about 136 days and the onset of puberty in females is normally between 180 - 210 days and in intact males is about 120 - 150 days (Whittemore 1980; Pond and Maner 1984). Further, the use of castrated, rather than intact males reduced any effects of male hormones on aggression. 43 SOCIAL STRUCTURE AND STATUS An understanding of social structure and organization in swine is important to both the researcher and the producer. In this study the assessment of individual social status was based on two measures; general submissive behaviours and competitive outcomes over access to feed and water. Van Kreveld (1970) has defined dominance as "a priority of access to an approach situation or to leaving an avoidance situation" and although there is much debate on the virtues of the different methods of assessment (Richards 1974; Craig 1986; McGlone 1986b), dominance in pigs has generally been evaluated in two ways. The first method involves scoring the frequency and direction of aggressive interactions within groups (e.g. Rasmussen et al. 1962; Ewbank and Bryant 1972; Meese and Ewbank 1972, 1973; Fraser 1974) and the second examines outcomes of competition for resources (e.g. Scheel et al. 1977; Craig 1986). Most of the aggression is thought to be displayed by dominant individuals (Ewbank and Meese 1971; Hansen et al. 1982), but it was felt that the direction of aggression was not always a good indicator of dominance. In this study and those of McBride et al. (1964) and Ewbank and Bryant (1972), some feeding subordinates were observed to initiate aggressive behaviour when higher ranking individuals intruded at the feed trough. Under these circumstances, often little or no aggression was displayed by the intruding dominant individuals. Further, low ranking pigs were observed, at times, to retaliate when displaced from the feeder. For these reasons the exhibition of submissive behaviour, rather than aggressive behaviour was used in this study to determine social status. Rather than 44 using a s ingle compet i t i ve measure such as rep lacements (d isp lacements) at the feeder (Meese and Ewbank 1973), both d i sp lacements and intrusion attempts were recorded to gain a better picture of the soc ia l re lat ionships. S o c i a l S t r u c t u r e The soc ia l hierarchy structure var ied between the t reatments , as di f ferent structures s e e m e d to ref lect the d i f fer ing degrees of fami l iar i ty a m o n g the animals in each treatment. In the 3:3 mixed groups , s t rongly linear hierarchies p redominated , presumably the result of the high leve ls of unfami l iar i ty a m o n g the an imals . In this treatment overt aggress ive interact ions quickly estab l i shed a c lear ly def ined hierarchy with no undisputed soc ia l p o s i t i o n s . In the Stresni l treatment, which should have had the same levels of unfami l iar i ty a m o n g m e m b e r s , s o m e groups exhibited linear hierarchies whi le others d i s p l a y e d non- l inear hierarchies. The drug may therefore not only delay the onset of aggress ion but a lso s e e m s to interfere with the normal deve lopment of a hierarchy. W i l s o n (1975) has suggested that for m o s t animals the f o r m a t i o n of a linear hierarchy impl ies the stabi l izat ion of the dominance order. Whi le this may be true of newly f o r m e d soc ia l hierarchies, the data f r o m this study indicate that this is not necessar i l y true in estab l i shed hierarchies. Non linear hierarchies were c o m m o n in groups containing famil iar groupmates (unmixed and 5;1 mixed groups) , c o m p l e t e with reversals and triangular re lat ionships . There are a number of p o s s i b l e reasons for this. First, the soc ia l hierachies of estab l i shed groups may be maintained through covert rather than overt agonist ic patterns. Consequent ly , the current methods of 45 a s s e s s m e n t may be unable to measure subtle behaviours used to maintain the hierarchies. S e c o n d , fami l iar individuals may be more tolerant of v io lat ions to their soc ia l pos i t ions by other group m e m b e r s ; and third, our measurement or understanding of the role and funct ioning of hierarchies may be inaccurate. Certainly there is much d i s c u s s i o n on the w a y we measure and interpret such soc ia l structures (Richards 1974; Craig 1986; M c G l o n e 1986b). Socia l Rank and Weight Gain D o m i n a n c e is said to confer advantages to high ranking individuals when resources are either l imited or loca l ized (Craig 1986). Usual ly individuals of high soc ia l status obtain most of a resource (McBride et al. 1964; Ewbank 1969a; Craig 1986), or the major i ty share a lmost equal amounts whi le the most subordinate rece ives little or none (Craig 1986). In this study s o m e pigs were o b s e r v e d to be act ive ly excluded f r o m the f e e d trough by the or ientat ion of the feed ing p igs ' b o d i e s . This behaviour w a s most o b v i o u s in the f irst week but a lso cont inued in subsequent w e e k s . A s a result, the d i f ferent ia l abi l i ty of high ranking individuals to acquire f o o d w a s re f lec ted in the weight gains of the 3:3 mixed groups during the f irst week and over the 3 w e e k s . Whi le this re lat ionship w a s not apparent in the S t resn i l - t reated groups, it is probably due to their de layed aggress ion and unsett led soc ia l structure. Other researchers have found that when animals are limit f e d , growth rate is related to soc ia l rank and f ight ing (Bryant and Ewbank 1972; Dantzer 1972a,b, c i ted in Sherritt et al. 1974). McBr ide and co workers (1964) 46 est imated that 13% of the total variance in growth is due to soc ia l rank, and other studies have s h o w n that dominant pigs spend more t ime at the f e e d trough (Baxter 1983/84; Hansen et al. 1982). A l though R a s m u s s e n et al. (1962) found no corre lat ion between soc ia l status and weight gain, they recogn ized that the degree of c o m p e t i t i o n for f o o d w a s not high in their exper iments . EFFECTS OF MIXING ON PRODUCTIVITY M a n y of the s t ress related changes in behaviour patterns were c lear ly a s s o c i a t e d with the p igs ' weight gains and feed c o n v e r s i o n s . Whi le s o m e researchers have been unable to f ind an e f f e c t of regrouping an imals on weight gains (Teague and Gr i fo 1961; J e n s e n 1971; Dantzer 1970, c i ted in Ewbank 1972; Graves et al. 1978) and others suggest a temporary improvement in feed e f f i c i e n c y (Friend et al. 1981), mixing s ign i f i cant ly d e p r e s s e d both the growth and f e e d e f f i c i e n c y of p igs in t reatments invo lv ing mixed pigs in this study. A l though d i f fe rences in average dai ly gain were o b s e r v e d only in the f irst w e e k , s imi lar to reports by M c G l o n e et al. (1986) for 55 - 58 kg pigs, there w a s a carryover e f fec t of the initial d i f fe rences in this s tudy. No s igni f icant d i f fe rences in weight gain occurred during either the s e c o n d or third w e e k s , but over the entire 3 week per iod s igni f icant d i f fe rences were apparent. S imi la r ly , mixing resulted in poor c o n v e r s i o n of f e e d , with the 3:3 mixed and Stresni l t reatments being the m o s t inef f ic ient . Teague and G r i f o (1961), have a lso o b s e r v e d that mixing during the g r o w i n g - f i n i s h i n g per iod decreases feed e f f i c i e n c y . These results suggest that though animals m a y appear to eat wel l after 47 regrouping, s t ress is sti l l present, and a f f e c t s product iv i ty . The c o m b i n e d f indings of weight gain and feed c o n v e r s i o n are o b v i o u s l y of cons iderab le importance to the producer because they demonstrate that mixing has covert but long term depress ive e f f e c t s on product iv i ty . M o v i n g to a new pen w a s a lso a s s o c i a t e d with a dec l ine in the product iv i ty of the an imals . This w a s readi ly apparent f r o m the depressed weight gain of the unmixed animals during the f irst week as c o m p a r e d to subsequent w e e k s . Mixed groups are therefore not on ly e x p o s e d to s t ress a s s s o c i a t e d with mix ing, but a lso to stress caused by being m o v e d to new surroundings. A l though the 3:3 mixed groups s h o w e d the lowest average dai ly gain in the f irst week, by the third week they were gaining proport iona l ly more than the other mixed groups. The intense f ighting d i s p l a y e d immediate ly after regrouping may have estab l i shed the new soc ia l hierarchy faster than in the other mixed treatments , and as a result, fewer d isputes and hence less s t ress , occurred in the last two w e e k s . T i n d s l e y and Lean (1984) have a lso suggested that a shorter per iod of hierarchy establ ishment may be benef i c ia l . In their exper iment, e v e n - w e i g h t groups s h o w e d more aggress ive retal iat ions than d iss imi lar weight groups , but establ ished a hierarchy and gained weight faster . Us ing animal product ion parameters as indicators of w e l l - b e i n g , S t resn i l - t reated groups appear to be a f f e c t e d by s t ress fac tors for a longer per iod than untreated 3:3 mixed groups. Initial tests of Stresni l in Europe did not measure animal product ion (Callear and van Geste l 1971; S y m o e n s and van den Brande 1969) whi le later ones have c l a i m e d that Stresni l 48 increases average weight gain (Ludvigsen 1970, c i ted in Cal lear and van G e s t e l ; S y m o e n s and Gaert , unpubl ished, c i ted in Cal lear and van G e s t e l ; C r o w n Chemica l C o . Ltd., Lamberhurst , Kent, c i ted in B lackshaw 1981; Porter and S lusser 1984), and f e e d e f f i c i e n c y (Porter and S lusser 1984). In this study and that of B lackshaw (1981), however , Stresni l w a s not e f f e c t i v e in improv ing relative growth rate. The weight gains o b s e r v e d in the 5:1 mixed group were rather unique. Initially, groups in this treatment s h o w e d g o o d average gains c o m p a r e d to the other mixed groups , as expected , because on ly one unfamil iar member w a s present in this treatment and so should have contr ibuted little to average va lues . O v e r the 3 w e e k s , however , these groups did not s h o w the s a m e improvement o b s e r v e d in the 3:3 mixed groups. A s in the S t resn i l - t reated groups , this indicates long term s t ress . This stress appeared to af fect the whole group, rather than just the s ingle individual or the introduced animals s ince no d i f fe rence w a s o b s e r v e d in the weight gains of the s ingle heavy or light p igs , and those in the rest of group. A s o b s e r v e d with agonist ic behaviour , the sex of the pigs had no e f fect on weight gains. The product iv i ty of both sexes appeared to be a f f e c t e d equal ly by the mixing cond i t ions i m p o s e d on them. CONCLUSIONS It is o b v i o u s f rom this study that mixing pigs f r o m dif ferent groups or litters adverse ly a f fec ts f in ish ing swine . Not on ly d o e s mixing promote aggress ion and f ighting but it a l so s ign i f i cant ly a f f e c t s product iv i ty , and hence, e c o n o m i c returns, both in the short and long term. Even after three w e e k s , the initial setback in product iv i ty w a s still apparent and in this study the addit ional c o s t s of mixing were substantial ($1.94/pig for 3:3 mixed, $3.50/pig for. Stresni l treated, and $1.54/pig for 5:1 mixed). Rather than e l iminate f ight ing, the tranquil izer Stresni l appeared to disrupt the an ima ls ' behavioural repertoire, de lay ing the peak aggress ive per iod , thus retarding the establ ishment of a "stable" soc ia l hierarchy, and further depress ing product iv i ty . The introduct ion of animals into o c c u p i e d pens a lso appears to disrupt the normal p r o c e s s of hierarchy f o r m a t i o n ; pr ior ity of ownership of space having a s igni f icant inf luence on aggress ive behaviour. Maintaining pigs in the s a m e groups , when m o v e d to a new pen, appeared to cause s o m e disrupt ions of the soc ia l hierarchy and product iv i ty . However these d isrupt ions were only s h o r t - l i v e d and consequent ly detr imental e f f e c t s on product ion were m i n i m i z e d . 49 RECOMMENDATIONS Regrouping pigs is c o m m o n l y pract ised on many swine operat ions , but it cannot be r e c o m m e n d e d f rom m y research. Regrouping adverse ly a f fec ts product iv i ty and a lso has a negative impact on the wel fare of the an imals . The stress a s s o c i a t e d with regrouping has a lso been found to p red ispose pigs to i l lness , injuries and accord ing to s o m e researchers , even death (Meese and Ewbank 1972). A t least one researcher has c o m m e n t e d that f ighting to the point of death s e e m s to have increased in recent years (Stone 1983). In m y study, one pig su f fe red a broken leg due to f ighting and another contracted an i l lness that necess i ta ted its remova l f r o m the group for a per iod of t ime. A f e w animals a lso suf fered temporary i l lnesses or d e v e l o p e d a b s c e s s e s as a result of injuries susta ined in f ight ing. Barr ing death, injury or i l lness , the losses in product iv i ty f r o m mixing pigs should be suf f ic ient to d iscourage producers f r o m this pract ice . A l though setbacks in weight gain and f e e d e f f i c i e n c y are usual ly only reported in the f irst week, m y results s h o w that they last as long as 3 w e e k s . In any product ion s y s t e m , the c o s t s of mixing w o u l d naturally have to be we ighed against those of any increased housing c o s t s for smal ler groups . Thus far, however , it appears that the e c o n o m i c e f f e c t s of mixing pigs has never been calculated in the literature. For the swine operator the e c o n o m i c c o s t s of mixing is a major cons iderat ion in making management d e c i s i o n s and in m y study, the addit ional housing and feed c o s t s to retain mixed pigs f r o m an initial weight of 76 kg to a f inal weight of 95 kg were substant ia l ; $1.43/pig for 3;3 mixed groups, $2.92/pig for 50 51 St resn i l - t reated groups, and $1.13 for 5:1 mixed groups (1987 Canadian dol lars) . Extrapolat ing to a market weight of 102 kg, the addit ional c o s t s a s s o c i a t e d with mixing were even greater (see A p p e n d i x 1 for cost analys is) . M y calculat ions do not include the extra c o s t s assoc ia ted with prov id ing pens to separately house animals that contract i l lnesses or are injured as a result of mix ing, medica l treatment, nor the labour that is needed during the extra d a y s that mixed pigs spend in the pens. In s i tuat ions where regrouping pigs is unavoidable , however , a number of r e c o m m e n d a t i o n s can be f o l l o w e d : 1. Feed pigs ad libitum during the f irst week f o l l o w i n g mixing The virtues of ad libitum feed ing versus restr icted feed ing and its inf luence on mixed pigs is contrad ic tory . Sherritt et al. (1974) and Graves et al. (1978) found that mixed litters that were initial ly l i m i t - f e d grew more s l o w l y than other groups , while pigs which were mixed but not l i m i t - f e d , init ial ly grew at rates comparab le to those of unmixed groups . Other studies have shown s igni f icant d i f fe rences in product iv i ty a m o n g mixed and unmixed treatments even with ad libitum feed ing s y s t e m s (Hines 1985; M c G l o n e and Curtis 1985; M c G l o n e et al. 1986), however , and at least one European producer has had re lat ively g o o d weight gains in mixed pigs with restr icted feed ing 4 t imes a day (Best 1971). In spite of these conf l i c t ing results , swine operat ions pract is ing limit feed ing may benefit by feed ing newly regrouped pigs ad libitum during the f irst week . During the f irst w e e k , when aggress ion is at its highest and subord inates are o f ten prevented f r o m feed ing , i m p o s i n g nutritional s t ress can on ly increase an imals ' suscept ib i l i ty to i l lnesses . A l though feed e f f i c i e n c y may sti l l suffer 52 with the implementat ion of an ad libitum feeding s y s t e m , newly mixed animals are not l ikely to lose as much condi t ion as they w o u l d under a limit feed ing s y s t e m ; a l so , the l i m i t - f e e d i n g s y s t e m can a lways be re i m p l e m e n t e d after the f irst week . 2. Ensure that there is suf f ic ient feed trough space in the pen Feeder space is espec ia l l y important with mixed p igs because of the high level of aggress ion that is d i s p l a y e d at the feeder . In estab l i shed groups less agonist ic act iv i ty is s h o w n but a certain basel ine level is usual ly maintained and that level tends to increase if the area decreases (Craig 1986). S p a c e is, therefore , a prized p o s s e s s i o n and feeder space that w o u l d normal ly be suf f ic ient for estab l i shed groups probably wil l be insuff ic ient for mixed groups . Th is is particularly true in the first week f o l l o w i n g mix ing. In these exper iments , subordinate animals were o f ten prevented f r o m feeding due to the c l o s e proximity of aggress ive indiv iduals . Even with an ad libitum feed ing s y s t e m , Hansen and Hagelso (1980) exper ienced p r o b l e m s with pigs guarding the feeders . Part i t ioned feed troughs or mult iple feed ing areas may therefore aid in decreas ing a g g r e s s i o n . 3. M o v e animals into a new pen It is apparent that pr ior ity o f p o s s e s s i o n or use of space p lays a role in aggress ion when animals are introduced into an occup ied pen. In these exper iments , f ierce aggress ion w a s d isp layed by original pen occupants . Under certain c i r cumstance , this behaviour may be used to advantage by the manager. For example , it w o u l d probably be more 53 benef ic ia l to mix a large group into the pen of a smal l group. 4. A v o i d use of the tranquil izer "Stresn i l " The use of Stresni l and related tranquil izers as management aids have rece ived much publ ic i ty in recent years . In spite of the c la ims to the contrary, Stresni l is not cost e f f e c t i v e when used in conjunct ion with regrouping. Initial tests of Stresni l in Europe (Callear and van Geste l 1971; S y m o e n s and van den Brande 1969) s h o w e d that the drug w a s s u c c e s s f u l both in prevent ing aggress ion when pigs were mixed and s t o p p i n g f ight ing once it had started. However , the drug d o e s not e l iminate f ight ing entirely but mere ly de lays it. In addit ion, no improvement in weight gain or feed e f f i c e n c y is obtained with St resn i l . Stresni l may play a better role in other aspects of swine management such as reducing s o w aggress ion towards piglets and reducing the inc idence of s t r e s s - r e l a t e d diarrhoea in p ig lets during weaning ( S y m o e n s 1975). 5. P lac ing large animals into a group of f in ish ing pigs serves no purpose Introducing large individuals into a group of smal l pigs has been r e c o m m e n d e d by a f e w individuals ( A n o n y m o u s 1974; Houpt and W o l s k i 1982; S tone 1983). Within the weight range of f in ishing p igs , no improvement is ach ieved by mixing a large individual with a group of f in ishing p igs , although -if a f in ishing pig were mixed with a group of weaners or growers the results might be d i f ferent . FUTURE RESEARCH NEEDS The c o m m o n pract ice of mixing f in ish ing pigs sti l l deserves further invest igat ion. Based on these results , mixing has long term e f f e c t s on the product iv i ty of the an imals , caus ing d e p r e s s i o n s in p e r f o r m a n c e that are detectable even after 3 w e e k s p o s t - m i x i n g . Much of the current research is a imed at reducing aggress ion in the short term (e.g. p h e r o m o n e s , tranqui l izers, odour mask ing c o m p o u n d s , etc.), but as results with Stresni l s h o w , reduct ions in v is ib le aggress ion m a y not necessar i l y be a val id indicator of reduced s t ress . In order to opt imize product iv i ty , p ig producers w o u l d probably benef i t f r o m f irst , trying to establ ish the new soc ia l hierarchy in the shortest t ime poss ib le (e.g. by initial contact through adjacent pen f e n c e s etc.) and s e c o n d , prov id ing a means for individuals of all soc ia l ranks to obtain f o o d and escape f r o m aggress ion . In this regard, protected individual feed ing and escape areas m a y help reduce the stress of mixing for pigs of low soc ia l status. Currently, s o m e work is being carr ied out along these l ines. Hide boxes (McGlone and Curtis 1985), pens uti l iz ing a maze s y s t e m (Nehring 1981) and two level pig pens (Philips and Fraser 1987>-are being examined. Research should a lso be d i rected at deve lop ing s y s t e m s where the mixing of unfamil iar p igs is unnecessary . Mult ip le group farrowing with a c o m m o n area for the piglets w o u l d a l low pigs to b e c o m e a c c u s t o m e d to each other at an early age. One concept that def in i te ly deserves future cons iderat ion is that of kin recogni t ion a m o n g f in ishing pigs and the existence of an "interlitter" soc ia l order. Pre l iminary data f r o m m y study suggests that interlitter soc ia l 54 55 organizat ion exists and may have a greater inf luence on mixing than do other factors such as sex and initial starting weight . In this study f e w intralitter f ights were o b s e r v e d f o l l o w i n g mixing and on o c a s s i o n , two pigs f r o m one litter were o b s e r v e d f ight ing with one pig f r o m another litter. In addi t ion , the pigs f r o m one litter were o f ten found to o c c u p y the top ranks of the soc ia l hierarchy. The full impl icat ions of this new concept wil l not be known without further research. LITERATURE CITED A d a m s , D B . 1979. Brain m e c h a n i s m s for o f f e n s e , de fense and s u b m i s s i o n . Behav. Brain S c i . 2:201 A l t m a n n , J . 1974. Observat iona l study of behaviour: sampl ing m e t h o d s . Behav. 49:227-267. A n o n y m o u s . 1974. Soc ia l behaviour of the p ig . Vet . Rec. 95:123. A r n o n e , M. and Dantzer, R. 1980b. D o e s frustrat ion induce aggress ion in p igs? A p p l . A n i m . Etho l . 6:351-362. A p p l . A n i m . Etho l . 11:74-75 (abstract). A s h f i e l d , G . 1984. Pigs at play are less aggress ive , more product ive . F e e d s t u f f s 56(10):1. Barrette, C. 1986. Fighting behaviour of wi ld Sus scrofa. J . M a m m . 67(1): 177-179 Baxter, M.R. 1983/84. Feeding and aggress ion in p igs . A p p l . A n i m . Etho l . 11:74-75 (abstract). Beilharz, R.G. and Cox, D.F. 1967. Soc ia l dominance in swine . A n i m . Behav. 15:117-122. Best , P. 1971. Making light of mix ing . P ig Farming 19(5);30-33. B lackshaw, J.K. 1981. The e f fec t of pen des ign and the tranqui l is ing drug, A z a p e r o n e , on the growth and behaviour of w e a n e d p igs . Aus t r . Vet . J . 57:272-276. B r o w n , J.L. 1970. The neural contro l of a g g r e s s i o n . In A n i m a l A g g r e s s i o n : S e l e c t e d Readings. Southwick , C.H. (ed). Van Nostrand Reinhold C o . N e w York . pp. 164-186 Bryant, M.J. and Ewbank, R. 1972. S o m e e f f e c t s of s tock ing rate and group size upon agonist ic behaviour in groups of growing p igs . Br. Vet . J . 128:64-69. Cal lear, J.F.F. and van G e s t e l , J .F.E. 1971. A n analys is of the results of f ie ld exper iments in pigs in the U.K. and Ireland with the sedat ive neurolept ic A z a p e r o n e . Vet . Rec. 89:453-458. Co l l i as , N.E. 1944. A g g r e s s i v e behaviour a m o n g vertebrate an imals . P h y s i o l . Z o o . 17(1):83-123 Co l l i as , N.E. 1953. Soc ia l behaviour in an imals . E c o l o g y 34:810-811. 56 57 Cra ig , J.V. 1986. Measur ing soc ia l behaviour: Soc ia l d o m i n a n c e . J . A n i m . S c i . 62:1120-1129 Dantzer, R. and M o r m e d e , P. 1979. E f f e c t s of l ithium on aggress ive behaviour in d o m e s t i c p igs . J . Vet . Pharmaco l . Therap. 2:200-303. Ewbank, R. 1969a. Soc ia l behaviour and intensive animal product ion . Vet. Rec. 85:183-186. Ewbank, R. 1972. S o c i a l env i ronment of the p ig . In: D.J.A. C o l e (ed.), Pig Product ion . Pennsy lvan ia State Un ivers i ty P ress , Un ivers i ty Park, P A , pp. 129-139. Ewbank, R. and Bryant, M.J . 1972. A g g r e s s i v e behaviour amongst groups of d o m e s t i c a t e d pigs kept at var ious s tock ing rates. A n i m . Behav. 20:21-28. Ewbank, R. and M e e s e , G.B. 1971. A g g r e s s i v e behaviour in groups of d o m e s t i c a t e d pigs on remova l and return of indiv iduals . A n i m . P r o d . 13:685-693. Fradrich, H. 1974. A c o m p a r i s o n of the behaviour of the Suidae. pp. 133-143, In: The Behaviour of Ungulates and its Relat ion to Management (eds. V. Ge is t and F. Walther), IUCN N e w Ser ies , no. 24 Fraser, D. 1974. The behaviour of growing pigs during exper imental soc ia l encounters . J . A g r i c . Sc i . , C a m b . 82:147-163. Fraser, D. and Rushen, J . 1987. A plea for prec i s ion in descr ib ing observat iona l m e t h o d s . A p p l . A n i m . Behav. S c i . 18:205-209. Fr iend, T.H., Tanks ley , T .D. and Krabe, D.A. 1981. Behaviour and per formance of p igs grouped by three di f ferent methods at w e a n i n g . J . A n i m . S c i . 53 (suppl. 1):127 (abstract). Gauthreaux, Jr., S .A . 1978. The eco log ica l s ign i f i cance of behavioural d o m i n a n c e . In: Bateson & Klopfer (eds) Perspect ives in E tho logy . V o l . 3: Chapt. 2. P lenum Press , N e w York . Graves , H.B., G r a v e s , K.L. and Sherritt, G.W. 1978. S o c i a l behaviour and growth of p igs f o l l o w i n g mixing during the g r o w i n g - f i n i s h i n g per iod . A p p l . A n i m . Ethol . 4 :169-180. Hansen, L.L. and Hage lso , A . M . 1980. A general survey of environmental inf luence on the soc ia l hierarchy funct ion in p igs . A c t a A g r i c . Scand inav ica 30:388-391. Hansen, L.L., Hage lso , A . M . and M a d s e n , A . 1982. Behavioural results and p e r f o r m a n c e of b a c o n p igs fed "ad l ib i tum" f r o m one or several s e l f - f e e d e r s . A p p l . A n i m . Etho l . 8 :307-333. 58 Hi l lyer, M . 1972. Mixing l itters. Pig Farming 20(5):103 Hines, R.H. 1985. E f fect of mixing on per fo rmance of f in ishing swine . A g r i c . Expt. Stn. , Kansas Stat. Univ. P rog . Rep. 486:94-97 Houpt, K.A. and W o l s k i , T.R. 1982. D o m e s t i c A n i m a l Behaviour for Veter inar ians and A n i m a l Sc ient i s t s . Iowa State Univ . Press . Iowa. p. 45. J a m e s , J . W . 1967. The value of soc ia l status to cattle and p igs . Proc . E c o l . S o c . Aust . 2:171-181. J e n s e n , A . H . 1971. B io log ica l impl icat ions of intensive swine rearing s y s t e m s . J . A n i m . S c i . 32:560-565. J e n s e n , P. 1982. A n analys is of agonist ic interaction patterns in g r o u p - h o u s e d dry s o w s - aggress ion regulation through an "avoidance order". A p p l . A n i m . Etho l . 9 :47-61 . J e n s e n , P. 1984. E f fec ts of conf inement on soc ia l interact ion patterns in dry s o w s . A p p l . A n i m . Behav. S c i . 12:93-101. J e n s e n , P. and W o o d - G u s h , D.G.M. 1984. Soc ia l interact ions in a group of f r e e - r a n g i n g s o w s . A p p l . A n i m . Behav. S c i . 12:327-337. Ke l ley , K.W., M c G l o n e , J . J . and Gask ins , C.T. 1980. Porc ine a g g r e s s i o n : M e a s u r e m e m t -and e f f e c t s of c rowding and fas t ing . - J . A n i m . S c i . 50:336-341. Krsiak, M., S u l c o v a , A . , Donat, P., T o m a s i k o v a , Z., D l o h o z k o v a , E., Kosar , E. and M a s e k , K. 1984. Can soc ia l and agonist ic interact ions be used to detect anxiolyt ic act iv i ty of drugs? Ethopharmac. A g g r e s s i o n Res . A l a n R. L i ss , New York , NY pp. 93-114 Kurz, J .C . & March inton , R.L. 1972. Rad iote lemetry studies of feral hogs in South Caro l ina . J . W i l d l . manage. 36:1240-1248. Lehner, P.N. 1979. Handbook of Etho log ica l M e t h o d s . Gar land Press . N e w York . McBr ide , G., J a m e s , J .W. and Hodgens , N. 1964. Soc ia l behaviour of d o m e s t i c an imals . IV. G r o w i n g p igs . A n i m . P rod . 6:129-139. M c C o r t W.D. and Graves , H.B. 1982. S o c i a l dominance re lat ionships and spac ing behaviour of swine . Behav. P r o c e s s e s 7:169-178. M c G l o n e , J . J . 1985. A quantitative ethogram of aggress ive and s u b m i s s i s v e behaviours in recent ly regrouped p igs . J . A n i m . S c i . 61:559-565. 59 M c G l o n e , J . J . 1986a. Influence of resources on pig aggress ion and d o m i n a n c e . Behav. P r o c e s s e s 12:135. M c G l o n e , J . J . 1986b. A g o n i s t i c behaviour in f o o d an imals : Rev iew of research and techniques. J . A n i m . S c i . 62:1130-1139. M c G l o n e , J . J . and Curtis, S.E. 1981a. A behav iour/per formance study to evaluate an alternative nursery pen des ign for s w i n e . J . A n i m . S c i . 53 (Suppl . 1):129-130 (abstract) M c G l o n e , J . J . and Curtis, S.E. 1981b. E f f e c t of gender on a g g r e s s i o n - inf luencing propert ies in swine urine. J . A n i m . S c i . 53 (Suppl. 1): 130 (abstract). M c G l o n e , J . J . and Curt is , S.E. 1985. Behaviour and p e r f o r m a n c e of weanl ing pigs in pens equipped with hide areas. J . A n i m . S c i . 60:20-24. M c G l o n e , J .J . , Curt is , S.E. and Banks, E.M. 1981. A g g r e s s i o n - i n f l u e n c i n g pheromones in s w i n e . J . A n i m . S c i . 53 (Suppl . 1): 130 (abstract) M c G l o n e , J .J . , Stansbury , W.F. and Tr ibb le , L.F. 1986. A e r o s o l i z e d 5 * - a n d r o s t - 1 6 - e n - 3 - o n e reduced agonis t i c behaviour and temporar i ly i m p r o v e d p e r f o r m a n c e of growing p igs . J . A n i m . S c i . 63:679-684. M a r d a r o w i c z , L. 1987. M a r d a r o w i s c , I. Pig News and Information 8(1): (abstract) M a r s b o o m , R. 1969. On the pharmaco logy of A z a p e r o n e , a neuroleptic used for the restraint of w i ld an imals . A c t a Z o o l . Path. A n t . 48:155-161. Matschke , G.H. & Hardister, J .P. 1966. M o v e m e n t s of transplanted European wi ld boar in North Carol ina and T e n n e s s e e . Proc . A n n . Con f . S.E. A s s n . G a m e & Fish C o m m . 20:74-84. M a r s b o o m , R. 1969. On the p h a r m a c o l o g y of A z a p e r o n e , a neurolept ic used for the restraint of w i ld an imals . A c t a z o o l . path. A n t w e r p . 48:155-161. M e e s e , G.B. and Ba ldwin , B.A. 1975. The e f f e c t s of ablat ion of the o l f a c t o r y bulbs on aggress ive behaviour in p igs . A p p l . A n i m . Ethol . 1: 251-262 . M e e s e , G.B. and Ba ldwin , B.A. 1977. S e n s o r y factors in the soc ia l behaviour of p igs. A p p l . A n i m . Ethol . 3:203 (abstract). M e e s e , G.B. and Ewbank, R. 1972. A note on instabi l i ty of the dominance hierarchy and var iat ions in levels of aggress ion within groups of fattening p igs . A n i m . Prod. 14:359-362. 60 M e e s e , G.B. and Ewbank, R. 1973. The estab l i shment and nature of the dominance hierarchy in the d o m e s t i c a t e d p ig . A n i m . Behav. 21:326-334. M iczek , K.A. and G o l d , L.H. 1983. d - A m p h e t a m i n e in squirrel m o n k e y s of d i f ferent soc ia l status; E f fec t on soc ia l and agonist ic behaviour, l o c o m o t i o n and s t e r e o t y p i e s . P s y c h o p h a r m . 81:183-190. M iczek , K.A., D e B o l d , J.F. and T h o m p s o n , M.L. 1984. P h a r m a c o l o g i c a l , hormona l , and behavioural manipulat ions in the ana lys is of aggress ive behaviour, pp. 1-26 In; E thopharmaco log ica l A g g r e s s i o n Research. A l a n R. L i ss , Inc. New York . M o s s , B.W. 1978. S o m e o b s e r v a t i o n s on the act iv i ty and aggress ive behaviour of p igs when penned prior to slaughter. A p p l . A n i m . Etho l . 4 :323-339. Nehring, A . 1981. One answer to the conf inement pig p r o b l e m . Int. J . S tud . A n i m . Prob. 2(5):256-259. Phi l ips, P.A. and Fraser, D. 1987. Des ign , cost and p e r f o r m a n c e of a f r e e - a c c e s s , t w o - l e v e l pen for g rowing f in ish ing p igs . Can . A g r i c . Eng . 29(2):193-195. P o n d , W . G . and Maner , J .H. 1984. Swine Product ion and Nutr it ion. AVI Publ . C o . Connet icut . pp. 124-138. Porter, D.B. and S lusser , C .A. 1984. Reducing stress in swine . A n i m . Nutr. Health Nov./Dec . :14-23. R a s m u s s e n , O.G., Banks, E.M., Berry, T .H. and Becker, D.E. 1962. S o c i a l dominance in g i l ts . J . A n i m . S c i . 21: 520-522. Richards, S . M . 1974. The concept of d o m i n a n c e and methods of a s s e s s m e n t . A n i m . Behav. 22:914-930. Rowel l , T .E . 1966. Hierarchy in the organizat ion of a capt ive b a b o o n group. A n i m . Behav. 14:430-443. S c h e e l , D.E., G r a v e s , H.B., and Sherritt , G.W. 1977. Nursing order, soc ia l dominance and growth in swine . J . A n i m . S c i . 45:219-229. Sherritt, G.W., Graves , H.B., G o b b l e , J.L. and Hazlett, V .E . 1974. E f f e c t s of mixing pigs during the growing f in ishing per iod . J . A n i m . S c i . 39:834-837. S inger , F.J., O t to , D.K., T i p t o n , A.R., and Hable, C.P. 1981. H o m e ranges, m o v e m e n t s and habitat use of European wi ld boar in T e n n e s s e e . J . W i l d l . Manage . 45:343-353. 61 Stephens , D.B. 1980. S t ress measurement in d o m e s t i c an imals . A d v . Vet. S c i . C o m p . M e d . 24:192-193 S t o n e , M.W. 1983. Cann ib i l i sm and other bad habits III. W e s t e r n Hog Journal Winter/1983 S y m o e n s , J . 1975. Psych ic s t ress as a cause of p o s t - w e a n i n g diarrhea in p ig lets . A p p l . A n i m . Ethol . 1:205 (abstract). S y m o e n s , J . and Van Den Brande, M. 1969. Prevent ion and cure of a g g r e s s i v e n e s s in pigs using the sedat ive A z a p e r o n e . Vet . Rec. 85 :64-67 . Teague , H.S. and G r i f o , A .P . 1961. M o v e m e n t and resort ing of pigs during the g r o w i n g - f i n i s h i n g per iod . A n . S c i . M i m e o 124. Oh io A g r i c . Exp. S t a . T i n d s l e y , W.E.C. and Lean, I.J. 1984. E f f e c t s of weight range at a l locat ion on product ion and behaviour in fattening pig groups . A p p l . A n i m . Behav. S c i . 12:79-92. T i s d e l l , C .A . 1982. W i l d P igs : Env i ronmenta l Pest or E c o n o m i c Resource? P e r g a m o n Press . S y d n e y , Aust ra l i a . Van Kreve ld , D. 1970. A se lec t ive rev iew of dominance subord inat ion relat ions in an imals . G e n . P s y c h o l . 81:143-173. W h i t t e m o r e , C.T. 1980. Pig Product ion . Longman Group L imi ted . New York, pp. 15-24. W i l s o n , E.O. 1975. S o c i o b i o l o g y : The N e w Synthes i s . Belknap Cambr idge , M a s s . pp. 279-297. Winer , B.J. 1962. Stat ist ica l Pr inc ip les in Experimental D e s i g n . M c G r a w - H i l l , • •New York . pp. 8 8 - 8 9 . W o o d , G.W. & Brenneman, R.E. 1980. Feral hog m o v e m e n t s and habitat use in coasta l South Caro l ina . J . W i l d l . Manage. 44:420-427. APPENDIX I ECONOMICS OF MIXING FINISHING PIGS A T THE UBC SWINE FACILITIES In calculating the extra costs of the various mixed treatments relative to the unmixed treatment, two main components can be identified. Given that the pigs start at a standard initial weight and are marketed at a predetermined finishing weight, the only difference between the treatments is 1) the amount of feed that must be fed in order for the pigs to achieve a common finishing weight and 2) the extra days that mixed pigs spend in the pens before the standard finishing weight is achieved. A. FEED COSTS E q u a t i o n : F/G x (Common f i n a l w e i g h t - a v e r a g e s t a r t i n g w e i g h t ) x C o s t o f f e e d = T o t a l f e e d c o s t p e r p i g Where: F/G of unmixed groups = 3.35 kg/kg F/G of 3:3 mixed groups = 3.74 kg/kg F/G of S t r e s n i l - t r e a t e d groups = 3.85 kg/kg F/G of 5:1 mixed groups =3.65 kg/kg Average s t a r t i n g weight = 76.218 kg Weight a f t e r 21 days = 95 kg Market weight = 102 kg 16% protein hog grower feed p r i c e = $ 187.54/tonne = $ O.188/kg T o 95 k g T o 102 k g Unmixed g r o u p s : $ 1 1 . 8 3 / p l g $ 18 2 4 / p t g 3 : 3 m i x e d g r o u p s : $ 13 .21/p1g $ 18 1 3 / p l g S t r e s n i l - t r e a t e d g r o u p s : $ 1 3 . 5 9 / p l g $ 18 6 6 / p l g 5 :1 m l x e d g r o u p s : $ 1 2 . 8 9 / p l g $ 17 8 9 / p l g B. SPACE COSTS a) Days to reach a standard f i n a l weight of 95 kg (based on regression equations of average weight) Unmixed groups : 95 kg = 0.91823 t + 75 645 t = 21 08 days 3:3 mixed groups : 95 kg = 0.91414 t + 74 050 t = 22 92 days Stresnl1-treated : 95 Kg = 0.86018 t + 74 674 groups t - 23 63 days 5:1 mixed groups : 95 kg » 0.81500 t 75 698 X t = 23 68 days 62 Days to reach a market weight of 102 kg (based on regression equations of average wel.ght) Unmixed groups 3:3 mixed groups Stresm 1 -treated groups 5:1 mixed groups 102 kg = 0.91823 t + 75.645 t = 28.70 days 102 kg = 0.91414 t + 74.050 t = 30.58 days 102 kg = 0.8G018 t + 74.674 t = 31.77 days 102 kg = 0.81500 t + 75.698 t « 32.27 days b) Pr i c e for grower-finisher barn = $ 15.74/sq. f t . * = $ 169.42/sq. m. * A. Wahl, B.C. M1n. Agric. Fish. Swine Special 1st. 1987 c) Estimated average l i f e of buildings = 20 years * = 7305 days * B.C. Pork Production Home Study Course, 1984 E q u a t i o n : Days t o Common f i n a l w e i g h t x C o s t o f p e n s p a c e = T o t a l s p a c e c o s t p e r p i g Where: Price per grower-finisher pen In U.B.C. Swine unit = (1.91m x 3.48m)/pen x $169.424/sq. m. = $ 1126.13/6 pigs/20 years = $ 187.69/pig/7305 days = $ 0.026/plg/day (minimum cost estimate since the assumption is that 6 pigs occupy the pen at a l l times and are marketed simultaneously) To 95 k g To 102 k g Unmixed g r o u p s : $ 0 . 5 5 / p i g $ 0 . 7 5 / p l g 3 : 3 m i x e d g r o u p s : $ 0 . 6 0 / p l g $ 0 . 8 0 / p i g S t r e s n i l - t r e a t e d g r o u p s : $ 0 . 6 1 / p i g $ 0 . 8 3 / p i g 5 :1 m i x e d g r o u p s : $ 0 . 6 2 / p l g $ 0 . 8 4 / p l g EXTRA COSTS ASSOCIATED WITH MIXING To 95 k g To 102 k g Unmixed g r o u p s T o t a l A d d i t i o n a l C o s t = $ 0 . 0 0 / p l g $ 0 . 0 0 / p l g 3 : 3 m i x e d g r o u p s Extra feed cost = $ 1.38/pig $ 1.89/plg Extra pen space cost = $ 0.05/p1g $ 0.05/p1g T o t a l A d d i t i o n a l C o s t = $ 1 . 4 3 / p i g $ 1 . 9 4 / p i g S t r e s n i l - t r e a t e d g r o u p s Extra feed cost = $ 1.76/plg $ 2.42/pig Extra pen space cost » $ 0.06/plg $ 0.08/plg Drug cost ($ 5.25/20 ml) Recommended dose (1 ml/18.2 kg l i v e wt.) = $ 1.10/plg (average 76.218 kg) T o t a l A d d i t i o n a l C o s t = $ 2 . 9 2 / p i g $ 3 . 5 0 / p i g 5:1 m i x e d g r o u p s Extra feed cost = $ 1.06/plg $ 1.45/p1g Extra pen space cost = $ 0.07/p1g $ 0.09/p1g T o t a l A d d i t i o n a l C o s t = $ 1 . 1 3 / p i g $ 1 . 5 4 / p i g These calculated costs are minimum estimates since the following costs have not been included: 1) labour associated with the extra days that mixed pigs spend in the pens, 2) medical treatment for pig(s) that sustain injuries or contract illnesses as a result of mixing and 3) extra pen space for sick and injured pig(s) should it be necessary to remove animal(s) to a separate pen until recovery is complete.

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
France 2 0
United States 2 0
Germany 1 3
Italy 1 0
Japan 1 0
City Views Downloads
Unknown 4 0
Tokyo 1 0
Sunnyvale 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items