EARLY WEANING IN SWINE: A BEHAVIOURALASSESSMENT A Thesis Presented to The Facuity of Graduate Studies of The University of Guelph by ERlN KATHLEEN SOPHIE WOROBEC In partial fuifilment of requirements for the degree of Master of Science August, 1997 O Enn Worobec, 1997
97
Embed
of - Library and Archives Canada · Piglets weafled at 14 days of age exhibited more behaviour problems than 28day ... The SEW pigs ranked the ... and Porcine Respiratory and Reproductive
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
EARLY WEANING IN SWINE: A BEHAVIOURAL ASSESSMENT
A Thesis
Presented to
The Facuity of Graduate Studies
of
The University of Guelph
by
ERlN KATHLEEN SOPHIE WOROBEC
In partial fuifilment of requirements
for the degree of
Master of Science
August, 1997
O Enn Worobec, 1997
National Library I*I of Canada Bibliothèque nationale du Canada
Acquisitions and Acquisiîions et Bibliographic Services services bibliographiques 395 Wellington Street 345. nie Wellmgton OttawaON KlAONQ -ON K l A W Canada CaMda
The author has granted a non- exclusive licence allowing the National Library of Canada to reproduce, loan, distribute or seil copies of this thesis in microfonn, paper or electronic formats-
The author retains ownership of the copyright in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author' s permission.
L'auteur a accordé une licence non exclusive permettant à la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/^ de reproduction sur papier ou sur format électronique.
L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.
ABSTRACT
EARLY WEANING IN SWINE: A BEHAWOURAL ASSESSMENT
Erin Kathleen Worobec University of Guelph, 1947
The relationship between weaning age, pigkt and sow behaviour and performance was
investigaiai. Piglets weaned at 7 days of age had difficdty adapting to weaning. They
exhibited redkcted nursÏng behaviour (beily-nosing), dec-d social contacf Iess interest
in theu environment and had minimal growth. This evidence suggests th& weaning at 7
days of age is too eady. Piglets weafled at 14 days of age exhibited more behaviour
problems than 28day-weaned anirnais but had Iess skin abrasions h m fighting. This
suggests that pigiets weaned at 14 days have both wetfare advantages and disadvantages
cornpared to those weaned at 228 days. They had similar growth rates to piglets weaned at 28
days indicating that they codd compete commercially with initiai f e d encouragement. Sow
behaviour and pdomance were not affected by weaning age in this study. AU these results
suggest bat weaning piglets between 14 and 21 days will be the best compromise between
performance and welfare.
Table of Contents
Literature Review
..................... ....*.......................................... The trend towards Early Weaning l.. 1
Disadvantages of SEW ............................................................................................ 3
Behaviour Feeding behaviour in Pigiets ..................................................................... 4
.......................................... Weaning behaviour - When the Sow Decides 6 Weaning behaviow at 3, 4 and 6 Weeks ................................................... 8
............................................................... Weaning behaviour at 2 Weeks 10
Feeding the Early-Weaned Piglet .......................................................................... 11
The Conflict between Nutritional Requirements and ................................................................................................. Behavioural Needs 12
.................... How Does Environmental Enrichment Affect Weaning Behaviour? 14
Does the Shortened Lactation Length Associated with Early Weaning affect Sow Behaviour and Welfare? ................................................ 16
Weaning Behaviour in Other Species ................................................................ 18
Pauer One: Earh Weaninn: the effects on tlidet behaviour
Table 1 : The mutually exclusive behaviour categories that were recorded d.yring the ....................... .................*. experiment together with their descriptions .... -3 5
Table 2: The percentage of the observed time piglets exhibited various behaviours ..... 36
Pauer Two: Does weanina age affect ~ ia le t uerformance?
Table 1 : Nutrient values and ingredients of the three diets ........................................... 52
Table 2: Mean feed disappearance ............................................................................... 53
Pawr Three: Effect of setrregated earlv weaning (SEW on sow behaviour
Table 1 : The mutuaiiy exclusive behaviour categories that were analyzed in the experiment together with their descriptions ...................................... ,. ............. 7 1
Table 2: The percentage of observational time the sows spent perforrning various behaviours ......................................................................................................... 72
iii
LIST OF FIGURES
Pa~er One: Earlv Weaninn: the effects on piglet behaviour
Figure I : The percentage of t h e piglets in the 3 treatment groups spent in various behaviours ............................................................................................... 38
Pa~er Two: Does weanine: age - affect pigIet mrformance?
Figure 1 : Bodyweight (kg) both before and after weaning over the six week experimental period ................... .... .............................................. 54
Figure 2: The mean skin integrity scores for each weaning age both before and after weaning ................................................................................................... 55
Figure 3: The percentage distribution of skin integrity scores one week after weaning for each treatment group ............................................................................ 56
P a ~ e r Three: Effect of semegated early weanina fSEW on sow behaviour
Figure 1 : The percentage of time the sows spent standing, lying ventrally and lying IateraUy ........................ .. ........................................................................ 74
The Trend Towards Eady Weaning
Over the last few decades weaning age in the mvine industry has been steadily
declining. Some of the proposed byefits of reducing weaning age include increased
facility utilization, increased numbers of litters per sow per year and a reduction in the
spread of disease. Brent et (1977) published the productivity results of the Cambridge
University sow herd in relation to weaning age. The nurnber of litters per sow per year
increased fiom 1.78 in wws weaned at 7&8 weeks to 1.98 in sows weaned at 5&6 weeks
to 2.24 in sows weaned at 3&4 weeks. In the last decade there has been a M e r decline
in the average weaning age on commercial pig farms. The catalyst for this recent decline
is the introduction of a management system know as Segregated Early Weaning (SEW).
SEW is a type of swine management system in which piglets are removed from the sow
at an eady age and raised in separate nurseries and grower-finisber facilities, away from
the breeding and farrowing areas.
In North America, before the introduction of SEW, conventionai weaning age was
21-28 days postpartum. This weaning age may be considered ''ear1y weaning" elsewhere
in the world. Most SEW f m s in Ontario wean their piglets between 14 and 17 days.
Some farms in the United States have average weaning ages of as early as 8-10 days
(Dial et al., 1995). There has been extensive research conducted regarding various
aspects of SEW. Researcti teams from the University of Guelph, Ridgetown College and
Agiculture Canada are currently coilaborating in an attempt to merge the knowledge
about SEW fiom rnany difFerent fields of study to formulate some genexal weaning
recornmendations. The project includes topics ranging from the econornic evaluation of
SEW, the anaiysis of immune and microbiological status, the evaluation of carcass
quality, the evaluation of reproductive performance, the assessrnent of nutritional needs
and an investigation into how early weaning affects piglet and sow behaviour.
Advantages of SEW
It is important to understand why SEW is emerging as a common management
practice in North America, and why this is happening before the specXc needs of the
anirnals have been established. The main goal of SEW is to limit the vertical transmission
of disease from the sow to her piglets (Clark 1995). SEW f m s attempt to wean the
piglets when they have received materna1 colostrum but have not yet received any
pathogenic organisms from the sow. The ideal weaning age of the piglet changes
depending on the potential pathogenic organisms (Dritz etal., 1994). Reduced weaning
age works in conjunction with more sanitary management practices (dl-in-dl-out
production) to produce healthier animals. All-in-dl-out production means that buildings
andior rooms are emptied, cleaned and disinfected before each new group of animais is
introduced. It helps to increase the health status of piglets by limiting the number of
pathogens present in the barn. This overall disease reduction may decrease the economic
impact of disease and may allow for greater exploitation of the genetic potential of lean
genotypes (Dial et., 1995). The anticipated higher health of auimals produced fiom
SEW units has been reported to result in increased feed efficiency (Clark 1995). This
proposed increase in feed efficiency wodd decrease the overall cost of production by
decreasing the total cost of feed on the f m .
Some of the disease benefits of SEW have been recently questioned.
Dionissopoulos et., (1997) compared the health status and feed efficiency of grow-
finish hogs obtained fkom a farrow-finish unit, an SEW nursery, and a specific pathogen
free (SPF) f m . The SEW pigs ranked the lowest in average daily feed intake (ADFE),
feed:gain ratio and contracted Mycoplasma hypopneumoniae @El) and Porcine
Respiratory and Reproductive Syndrome (PRRS) more often. One of the reasons that the
health status of the SEW pigs in this expriment was quite low may be because the
animais came fiom 6 dflerent sow herds. This means that piglets with many different
potentiai pathogens would be mixed together. Supplying piglets fiom several different
farrowuig herds for SEW nurseries is quite common in Ontario because Ontario's sow
herds are relatively small. This should be taken into consideration before the decision is
mzde to convert fiom a farrow-to-finish management system to SEW.
Disadvantages of SEW
The attainrnent of a hi& health status in a SEW herd requises superior
management skills, specialized food and housing and the CO-operation and CO-ordination
of pig producers. This speciaiized management system forces the mixing of piglets from
multiple sources, causes individual producers to lose their independence and results in
higher transportation costs (Ensrninger and Parker 1984). There has aiso been decreased
reproductive performance reported in sows that bave been weaned early. A decrease in a
sows' lactation length resulting fiom early-weaning may cause a drop in farrowing rate,
decrease the number of piglets born alive, decrease the number of piglets weaned per
litter and increase the number of non-productive sow days by increasing the weaning-to-
estrus interval (Deen 1994; Foxcrofi 1997). Field data based on 7-10 day weaning
suggested excellent piglet performance, but poor sow productivity (Dee 1995).
A possible disadvantage that is seemingly ignored in the overall cost-to-benefit
equation of segregated early weaning is the welfare of the aaimals. The welfate of both
the pigIets and tfle sows should be considered before SEW is accepted as a feasible mine
management system- An understanding of how the premature removal of piglets affects
their long term well-being and successful adaptation to their weaning environment should
be a critical step in the adoption of segregated early weaning.
Behaviour
Feedina Behaviour in Piglets
When weaned at an early age, piglets have to adapt to several changes in their
environment. The most critical of these is makuig the switch fiom sucking their mother's
milk to consuming solid food. An understanding on the ontogeny of sucking behaviour
wi11 therefore probably give indications as to how piglets h t learn to recognize and
consume food.
In fact, milk intake in piglets is dependent on the closely integrated behaviour of
the suckling sow and the sucking piglets. Suckling behaviour in swine is more complex
then in many other mammals because sows give birth to large litters of fairly precocial
Young. The sow m u t be able to assemble al1 of her offspring at the udder and adjust her
milk production for varied litter sizes. Research studies suggest that behaviour and
physiology help to dictate suckling patterns (Algers 1993). Sows change their suckling
behaviour as the piglets age. Irnmediately after parturition the sow has continuous milk
flow, allowing the piglets to suck mythe. About 10.4 4 4.5 hours after the birth of the
fht piglet, the sow has short bouts of mik availability (Lewis and H d 1985). When
the piglets are young (a few days old) the sow initiates most of the sucklings. The
initiation of these suckiings is signaleci by the sow gninting and rolling on her side to
expose her udder (Fraser 1980, Jensen et., 1991). A sucking bout is divided into four
phases: initial massage, an inactive phase, "me sucking", and a final massage phase
(Houpt 1982). The initiai massage phase (1 -3 min) is termed the "pre-ejection massaging
phase" which concludes with the release of oxytocin. Approximately 20 seconds &et
oxytocin release there is an increase in mammary pressure which causes miik let-down
(E l l endorE~ . , 1982). Fraser (1980) suggested that the massage period before milk let-
down allows the piglets time to assemble at the udder and find a teat to suck. The sow
speeds up her rate of gninting a few seconds before milk ejection as a signal for the
piglets. Newbom piglets "sample" a few dBerent teats before they find "the?' teat h m
which they will suck until they are weaned (Algers 1993).
MiUc letdown correlates with the "tme sucking" phase, which lasts about 10 to 20
seconds (Fraser 1980). The reason that miik is ody available for such a short t h e is
because the sow does out have a teat cistem to store the milk. Typical posture of the
piglets during milk let-down ("true sucking") includes a rigid stance with their ears back
and their tails tightly curled (Houpt 1982).
The h a 1 massage phase Iasts anywhere from 2-15 min, controversy exists
surroundhg its purpose. McBride (1963) suggests that the final massage scent marks
each individual teat as a type of "ownership". More recent research proposes the
"restaurant hypothesis", where piglets stimulate the development of the mammary gland
of their section of the udder. f i s would mean that piglets which performed more
forcefd pst-massage khaviour would receive more milk in their next sucking bout
(Algers and Jensen 1985). AIgers and Jensen (1991) observed that the duration and
intensity of the finai massage on a specific teat affected the mik production of that teat in
the first few days of lactation Spinka et., (1996) found that the length of udder
massage influenced hormone levels (prolacth, insulin, somatotropin) but did not affect
milk production.
Weaning Behaviour-When The Sow Decides
TJatural" weaning in M e is a gradual process. When housed in a semi-natural
environment a sow will seek out a favourable nest site one or two days before parturition.
The nest sites are usually isolated fiom where the main group activities take place (Stolba
and Wood-Gush 1989). During the first week postpartum the sow spends most of her
time with her piglets, Ieaving them ody for short spans of time (30 min on average).
Piglets as young as two weeks of age start to foliow the sow away fkom the nest. The
initiation of the weaning process starts, at the earliest, about ten days postparturn with a
decrease in suckling fiequency (Jensen et al., 1991). This decreased suckling time
corresponds to when the sow and her litter leave the nest to move closer to the group
(Jensen and Redbo 1987). When the piglets leave the nest with the sow they change fiom
"hiding" in the nest to "following" the sow when she leaves the nest. This "hider" to
"follower" shift causes a change in the interactions between the sow and her young. The
fiequency of nasal contact decreases and the frequency of auditory contact increases just
before nest Leaving time.
By weeks 3 and 4 the piglets are more active, engaging in ninning and playing
activities. Play fighting and aggressive behaviour between piglets starts to occur in weeks
5 and 6. Entire iitters are not completely weaned at the same time. By 6 weeks of age,
many unweaned piglets receive less than haif of their dry matter intake fkom mik
(English et., 1988). The strongest pigs in the Litter wean themselves before some of the
weaker animals in the litter. Weaning cannot be characterized by a specific point in time
but is a gradua1 shift fiom a reliance on the sow's miik to a reliance on solid feed. At 3
months postpartum, the sows in a Scottish study were suckling their piglets at 2 hour
intervals (Stolba and Wood-Gush 1989). There is some discrepancy as to when sows
have completely weaned their litters when they are housed in semi-natural environments.
For example, Stolba and Wood-Gush (1989) estimated this had occurred by 12.5 weeks
whereas Jensen and Recen (1989) thought that it was not completed until 17.2 weeks..
Similar weaning behaviour is seen in enriched pens (those with a separate area for
the sow and an area for the sow and her litter) as in semi-naturai environments. Sows
spend less time with their piglets and allow fewer sucklings as her piglets age. B0e
(1 99 1) found that the sow initiated 20% of the sucklings at 2 weeks postpartum versus
less then 10% of the sucklings at 10 weeks postpartum. Some sows started to stand for
nursing when their piglets were 4 weeks old. Standing for nursing could mark the
commencement of the weaning process because sows let down less milk while standing
compared with when they are lying (Gill and Thomson 1956). In one study, the final
stages of weaning were completed at 1 1-12 weeks in enriched pens ( B0e 199 1).
W d g Behaviour at 3.4 And 6 Weeks
The age at weaning in commercial pig operations has been steadily declining in
recent years. Weaning piglets at 4 or 6 weeks seems very late in cornparison to North
Arnerican "early-weaning" yet rem& quite young when compared with 'sow
controlled' weaning at 12-1 7 weeks. Abrupt weaning at any age causes the pigiets'
behaviour to change. They are moved fiom an environment where feeding and resting
behaviour of the litter is synchronized by the sow, to an UXLfamiIiar environment, with
strange pen-mates, and a novel food source where they have to initiate their own
maintenance behaviour. Weaning is a 'stressful' event at any age with indications that it
is more stressful at younger ages. Fraser (1978) noticed that piglets weaned at 3 weeks
showed an increase in overall activity and aggression, and appeared to have difficulty
lying together comfortably, compared with piglets weaned at 6 weeks of age. Wood-Gush
and Csermely (198 1) compared the diumal activity of a group of early weaned pigiets at
3 weeks and again at 6 weeks of age. They observed that the percentage of time that the
piglets spent lying, exploring, feeding, and drinking were essentiaily the same at 3 and 6
weeks of age. Only the scores for aggressive behaviour differed significantly between the
two ages. The piglets weaned at 6 weeks showed much less aggression than those weaned
at 3 weeks. This lower kequency of aggressive acts was most likely due to the piglets
being able to form a stable social hierarchy at 6 weeks and not 3 weeks.
As research to establish the relationships between weaning age and piglet
behaviour continued, the behaviour patterns studied becarne more specific. Researchers
began to measure the amount of behaviour that was directed towards the environment in
relation to the amount of behaviour that was directed towards pen-mates. The
performance of increased levels of redirected sucking-like activities in weaned piglets bas
been well documented (BIackshaw 1981; Bse 1993; Dybkjaer 1992; Worsaae and
Schmidt 1980). B8e (1993) found that sucking and massaging pen mates is seen
proportionately more when piglets are weaned at 4 weeks (sucking:O.l + 0.2% ;
rnassaging:0.7 f 1.6%) compared with 6 weeks (0.0 k 0.1% ; 0.1 + 0.2%). Other
beliaviours such as s ~ g / r o o t i n g pen-mates, chewing-nibbling fittings, and tail biting
are performed approximately the same arnount in piglets weaned at 4 and 6 weeks of age.
It is important to understand how to interpret the behaviourai dif5erences that
occur with early weaning. Dybkjaer (1992) attempted to identq behavioural indicators
of 'stress' in early-weaned piglets by exposing groups of piglets to two diEerent levels of
environmental and social stressors. Half of the piglets were placed in pens with eight
littermates (an area of 0.30 square metres per piglet) and given straw each day. The other
half were mixed with non-littermates (16 piglets in each pen with an area of 0.15 square
metres) and received no straw. Al1 of the piglets were weaned at four weeks of age and
the treatment groups were compared at two and four weeks after weaning. The
overcrowded, 'stressed' piglets performed more redirected oral behaviours such as belly-
nosing and manipulating ears, tails and other body parts in comparison to their less
'stressed' counterparts. It is important to recognize that some of these oral behaviour
patterns can aiso be seen in piglets housed in semi-natural environments but usually at a
much lower rate (Newberry and Wood-Gush 1988). The fiequency of redirected chewing
behaviour is substantially diminished in animals housed outdoors compared to intensively
housed animais (0.33 t i m d 30 min in a semi-natural environment; 2.2 timed 30 min in
'unstressed' treatment; 5.8 timesI30 min in 'stresse8 treatment).
BIackshaw (198 1) also studied persistent inguinal nose thnisting, and tail and ear
bithg in piglets weaned at 4 weeks of age. Biting was displayed by high ranking pigs and
may be fiinctional in the establishment of dominance hierarchies. Biting behaviour was
spread between pens where there was o d y visuai contact, suggesting that pigs imitate the
sight of another pig biting.
The motivation behind persistent inguinal nose t h d g (belly-nosing) is
unknown. Older studies have interpreted it as a displaced rooting urge ( Schouten 1986;
van Putten and Dammers 1976) or a resdt of hierarchical aggression (Blackshaw 1981).
More recent work has alluded to a possible association between nose-thnisting behaviour
and the initiation of feeding behaviour (Gonyou 1997). Further discussion about the
motivation behiad belly-nosing behaviour and its reIationship to weaning age is more
hlly expiored in the next section where it is related to 2-week-weaning.
Weaning Behaviour at 2 Weeks
As weaning age decreases, the proportion of abnormal and redirected behaviours
seems to increase. Few studies have examined the behaviour of piglets weaned at 2
weeks of age or younger. Metz and Gonyou (1990) compared the effect of age and
housing conditions on the behavioural and hemolytic reaction of piglets to weaning. The
behaviour of piglets weaned at 2 and 4 weeks were compared both after weaning and
while the piglets were with their dam. There were many différences in the feeding and
sucking behaviuur of piglets weaned at 2 versus 4 weeks when both were with the dam.
The 2 week old piglets spent significantly more tirne sucking than the 4 week old piglets.
The younger pigiets also had more contact with the teats after massage, often f a l h g
asleep beside i t The ciifferences between the behaviour of the 2- and 4-week-old piglets
were accentuated after weaning. Although the 2-week-weaned pigiets were younger and
presumably need more rest, the lying time of the 2-week-old pigiets was sigaifîcantly
shorter thau the 4-week-old piglets at both 4 and 6 days post-weaning . The younger
piglets also sbrted to beiiy-nose. The older piglets engaged in significady Iess belIy-
nosing behaviour. Any belly-nosing thai the 4-week-old pigiets did was less h t i c and
patterned d-erentiy than the belly-nosing of the younger piglets. The cause of the nosing
in the older piglets may be a rootiag urge rather than a sucking urge (Schouten 1986).
Belly-nosing and naval sucking in 2-week-weaned piglets does not start until about day 4
after weaniag. Gonyou (1997) suggests a connection between the initiation of feeding
and bely-nosing behaviour. This rnay have similar principles to the hding that veal
calves 'butt' more fiequently when artificiaiiy fed rr;ilk at a slow flow rate than at a
faster, easier to attain flow rate (Haley et., 1996). Research to establish the motivator
of belly-nosing and naval sucking behaviour must be conducted before proposed
soiutions to eliminate these behaviours can be established.
Feeding The EarIy Weaned Piglet
Before a relationship between the initiation of feeding and belly-nosing can be
postulated an understanding of how early-weaned piglets are currently fed must be
established. The nutritional requirements of the early-weaned piglet have been well
studied. C h e (1991) explains that early-weaned piglets have simiIar nutrient
requirements as growing-finishing piglets. AU young piglets require energy, amino acids,
fatty acids, minerais, vitamins and water. The objectives of the formulation of SEW dies
are expressed quite ciearly by Drïtz et (1997) as 1) the selection of ingredients that
stimulate feed intake 2) the provision of a substantial amount of highly available amino
acids in the proper proportions and 3) the preparation of pigs to utilize Iess expensive
diets in subsequent phases of growth. The feeding management system which is
recommended to achieve these objectives is a four phase diet including an SEW diet, a
transition diet, a phase two and a phase three diet. The shift from one feding phase to the
next brings a decrease in the amount of added fat in the diet, a decrease in the amount of
spray-dried plasma and an inmase in the amount of less digestible ingredients such as
soybeans and grain (Dritz et al., 1997). This ingredient shift causes the SEW diet to move
h m a highly palatable, easily -digested form to a coarser, less digestible, Iess costly
fom.
The Conflict Between Nutritional Requirements And Bebavioural Needs
It is essential to understand both what nutrients piglets require in different stages
of their life and what stirnuiates piglets to feed and establish feeding patterns. Piglets
establish their feeding cycles in their first few weeks of life fiom the vocal and
behavioural cues of the sow. The sow provides piglets with approximately 24 equally
spaced me& per day (Goodband et., 1995). Sow's d is highiy digestible and aii of
the piglets learn to feed at the same time and only when the sow allows them to eat. It is
important that nutritionists consider the development of this feeding behaviour when
designing feed and feeding schedules for the earty-weaned piglet Bark et al., (1986)
found that piglets weaned at 21 days and fed a dry diet either ad-libitum or at 2,4 and 6
hour meal intervals did not consume sufficient feed for maintenance during the first 3
days post-weaning. This suggested to the researchers that the number of meals provided
was not adequate. In a second experiment Bark et ai., (1986) compared the feed intake of
weaned pigiets that were allowed to consume food in a familiar environment via a
familiar method (sucking) 24 hours after weaning both with unweaned piglets and with
weaned piglets put into an unfamiliar environment and fed a dry diet. The weaned group
fed a dry diet consurned less and had signi£icantly higher plasma free fatty acid (FFA)
levels which may suggest a reliance on body fat stores during this period. The reduction
in feed intake which is seen directiy following weaning is likely caused by both an
unfaLniliar method of food acquisition and an unfarniliar food source.
Auditory and thermal cues are also very important in the feeding pattern of
piglets. Petrie and Gonyou (1 988) found that newly weaned piglets given the auditory
stimulus of a sow grunting spent more tirne feediag than did piglets without the auditory
stimulus (1 27.1 vs. 104.2 min, respectively). The initial seeking of a food source seems to
be in response to the surface of the udder. Welch and Baxter (1986) found that the body
temperature of sows rises before fmowing and when milk is present in the teats. The
researchers camed out preference tests with models using 3 different combinations of
temperature and 'sofhess' of udder. Piglets demonstrated a clear preference for a warm,
soft surface, both by the amount of tirne spent at the models and by the sucking actions
directed towards the models.
The development of feeding behaviour in young piglets bas been studied
intensively and a solid knowledge base has been established. The current irnplementation
of very early weaning practices on commercial swine f m s rnakes it al1 the more
important that this knowledge base be increased and used. Ignoring the learned feeding
behaviour of piglets may cause problems such as redirected sucking behaviour and
digestive discodort post-weaning. Both the nutritional and behavioural needs of the
early-weaned piglet must be met to achieve the maximum benefit fiom SEW systems,
How Does Environmental Enrichment Affect Weanhg Behaviour?
Piglets weaned at an early age may have many other behavioural needs apart from
those associated with feeding. One way that commercial pig producen may be able to
meet some of the behavioural needs specifically associated with abrupt weaning without
compromising overall productivity is through a concept known as environmental
enrichment. Many researchers have studied how different kinds of environmental
enrichment affect the behaviour of piglets following weaning. Schaefer et al., (1 990)
compared the arnount of aggressive behaviour exhibited by newly weaned piglets when
they were provided with objects such as a car tire suspended on a chah, a sugar-mineral
block suspended in a metal basket or a teeter-totter (metal bar with rubber belts on the
ends). Less aggressive acts were displayed when newly weaned piglets were given any of
the three "toys" in cornparison to the control (no toys provided) animais. The researchers
noted that the larger toys not only provided an aggressive outlet for the dominant anirnals
they also gave a hiding area for the submissive piglets. However, it is important to
acknowledge that familiarity with toy objects may reduce their effectiveness in
controlling aggression (Schaefer et., 1990). Waran and Broom (1993) studied the
influence of a galvanized metal barrier and two different housing systems (conventiond
flat deck; pen with straw) on the behaviour and growth of early weaned piglets and found
sindar results to the toy study. The provision of an opaque barrier both decreased the
fiequency of aggressive interaction by 40% the week after weaning and increased the
piglets' growth rate by 0.15 proportionately. Surprisingly, the piglets housed in a
conventionai flat deck demonstrated sunilar IeveIs of aggression as the piglets housed in
a pen bedded with straw. Btae (1993) used a 2 x 2 factorid design to compare piglets
weaned at 4 or 6 weeks with piglets housed in either a flatdeck or a farrowing pen
(bedded with sawdust). Piglets weaned in flat-decks spent more time rooting at objects in
the pen and chewinghibbling and tail-biting pen-mates than piglets weaned in the
farrowing pen. The fiequency of massaging and sucking pen-mates was higher amongst
piglets weaned at 4 weeks compared to 6 weeks and was also found to be higher among
piglets weaned in farrowing pens compared to flatdecks. FIat-deck housed piglets had a
higher fkequency of belly and tail lesious than animals in farrowing pens.
The behaviourd effects of early weaning are difficult to separate fiom the effects
of the weaning environment. Studies suggest that both factors contribute to the formation
of abnormal behavioural patterns post-weaning. The research to be described in this
thesis attempts to separate the effects of early weaning fiom the weaning environment by
housing al1 three weaning ages in identical environments and cornparhg the frequency of
abnormal behaviour exhibited.
Does The Shortened Lactation Length Associated With Eariy-Weaning Anert SOW
Behaviour And Welfare?
Previous research has established the patterns of nursing behaviour of piglets, the
behaviourai changes associated with the process of weaning (both sow controlled and
&ficial), the feeding systems of commercial SEW systems and how these feeding
systems relate to the behavioural needs of the piglets, and the use of environmental
enrichment to attempt to lesson the 'stress' of weaning. If early-weaning causes
behaviourai changes in piglets, how does it effect the sows?
Sows have complex social interactions with their Liners that govern the cycle of
piglet behaviour. They fom strong materna1 bonds with their piglets. Jensen et al., (1991)
observed that mothers and their young spent almost all resting time in close contact. As
the piglets aged, the sow spent less time in close proximity to them but the tirne she
spent close to her young during resting did not change. This suggests that the maternd
bond between a sow and her young continues until the completion of weaning. There is
practically no information available on how the early temination of this relationship
affects the sow. As previously mentioned, a disadvantage of SEW is reduced sow
reproductive performance. Lactation length seems to be inversely related to the weaning-
to-breeding i n t e d .
increase weanhg to
explains that one of
Each 10-day decrease in lactation length has been estimated to
breeding time by one day (Friendship 1987). Foxcrofk (1997) C
the reasons that sow productivity drops due to early-weaning is
because the reproductive system needs to recover fiom the effects of pregnancy. The
uterus has to r e m to its non-pregnant state and the brah and pituitary need to be
capable of producing the hormones (LH, FSH) needed for reproductive success. The sow
seems to have diffîculty adapting physiologicaily to early weaning. How does this lack
of adaptation relate to the animal's behaviour and weffare?
This question is reIativeIy novel and essentiaiiy uncharted, Pajor et ai., (1996)
studied the individuai difEerences and housing effects on the response of sows to
weaning. Sows were housed in either a pen confined with their litter or in a "get away"
pen where the sow couid leave her litter. They found a large variation in the amount of
tixne sows choose to spend with their young. Two categories of sows were defined based
on how much time they spent with their litters: "leaver" sows spent more than 50% of
their time away, "stayer" sows spent less than 50% of their time away. "Stayer" sows
were reported to vocalize more post-weaning, were more active, and investigated the
piglet creep area more often. These possible indicators of sow distress wouid be
important considerations in the design of an expriment attempting to establish the
relationship between lactation length and sow behaviour.
The identification of the signals that indicate the initiation of the process of
weaning are dso important. Boe (1991) postdates that the process of weaning is the
result of a confiict between mother and offspriag. If this is me, then an artificiaiiy
weaned sow is likely to never experience this conflict. This is not to say that such conflict
shouid somehow be induced or simulated at the same time the sows are artificially
weaned fiom their piglets. However, artificiai weaning practices such as SEW may be
depriving the sow of a stimulus which is essential to her recognition of the termination of
the relationship between her and her offspring.
Some may argue that the process of a sow suckiing her young is just as 'stressful',
if not more so, as t a h g ail of her young away relatively early in her lactation. Bse
(1991) opposed this theory by observing that given the opportunity for a sow to open and
dose a b&er which separates her fiom her litter, she wilI consistently and voluntarily
open this barrier and move to an area where she can suckle her young. AIthough suckhg
may be physioIogica.Uy 'sîresstùi' for the sow, she may have a behavioural need or
motivation to perform it.
The design of commercial management systems ofien ignores the animals'
behavioural needs. There is an obvious lack of information about how sows respond
behaviourdiy to reduced lactation length. Knowledge of this area may help us to
understand some of the reproductive failures of SEW sows and would also aid in an
assesment of the. well-being of early-weaned sows.
Weaning Behaviour In Other Species
Behavioural patterns associated with weaning across various species are similar.
The removal of young animals fiom their mother is understood to be a 'stressful' event in
many different species. The offspring of most animals that are artificially housed and
managed are abruptly weaned fiom their dam.
In anUnals f i e d for their fur such as mink, early weaning has been shown to
cause abnormal behaviour patterns. Mason (1996) found that mink removed from their
mother at 7 weeks compared with 11 weeks exhibited more stereotypic behaviour which
continued over the 9 month observation period. Laboratory mice weaned at 17 days with
a low weaning weight compared with 20 days with a higher weaning weight had higher
corticosterone levels 48 hours d e r weaning and developed a higher Level of stereotypic
wire-gnawing behaviour (Wurbel and Stauffacher 1996). Female hooded rats weaned
their litters eariy when they were food-restricted (Smith 1991). These early-weaned rat
pups exhibited more play behaviour than the controt animais- Martin and Bateson (1985)
observed that kittens aiso responded to early weaning by playing more and that diEerent
types of play behaviour were shown depending on when weaning occurred. The
researchers suggested that weaning age may affect the way that young kittens develop.
Early-weaning red deer calves at 3-4 months ( compared with the 7-8 months that occurs
naturally) aiso caused differences in the caifbehaviour. The weaned animals spent 12%
of their time pacing at the fencehe as opposed to the 1% pacing time seen in the
unweaned group (Pollard et., 1992). Human infants also exhibit changes in their
behaviour post-weaning. Prolonged crying, generai restlessness, and excessive pacifier
sucking are al1 possible results of the human weaning process (Gesell and ilg 1979).
In summary, weaning results in prolonged changes in behaviour in many different
species. The knowledge and appreciation of these changes is an essential part of strong
animal hus bandry skills.
Early weaning: The effects on piglet behaviour
To be submitted to Applied Animal Behaviour Science
Abstract
Piglets were weaned at 7, 14, and 28 days, housed in groups of ten and obsenred
until they were six weeks of age. Piglets weaned at 7 days spent much more time in
activities directed towards theu pen-mates, particuIarly belly-nosing behaviour @<O.OS),
showed more escape behaviour (j! <0.005), and less contact @ < 0.005), feeding
@<O.OS), and exploratory behaviour @<0.0005) compared with piglets weaned at 14 or
28 days. Piglets weaned at 14 days exhibited more belly-nosing behaviour than piglets
weaned at 28 days @<0.05) and spent less t he at the feeder directly foilowing weaning
than the 28 day weaned animals @0.05).
General aggression levels, aggression at the feeder and drinker and the
performance of social nosing and chewing behaviour were similar in piglets weaned at 7,
14 and 28 days of age. These results suggest that the welfare of piglets weaned at 14 days
and particuiarly 7 days is compromised. The welfare of early-weaned piglets must be
considered as part of the assessment of the feasibiiity of early-weaning systems.
Keywords: piglet behaviour, early- weaning, w e k e assessment
Introduction
[n North America, weaning age on commercial M e fanns has been steadily
decreasing. Recently, eariy weaning has been combined with a segregation of the piglets
fiom the sow in a system known as Segregated Early Weaning (SEW). Usually, several
sow breeding f m s attempt to synchronize fmowing. When aii the piglets on these
supply farms are about 14 days old they are weaned and transported to a growing f m
which has been thoroughly cleaned and disinfected. This unit is run on an all-in all-out
basis with strict biological security. Weaning ages in SEW systems range fiom 16-17
days to a s early as 7-10 days post partum. The main advantage of this system is the
decrease in vertical transmission of disease from the sow to her litter (Clark 1995). SEW
is being rapidly adopted by the swine industry in North America, suggesting that it works
well cornmercially. However, weaning piglets as early as 14 days raises questions of
animai welfare. Although their health status may be enhanceci, do piglets sufEer in other
ways when separated fiom their mother at 14 days?
Natural weaning in swine is a gradual process. When housed in semi-natural
environments the sow will seek out a favourable nest site one or two days before
parturition (Stolba and Wood-Gush 1989). During the first week postpartum, the sow
spends most of her tirne with her piglets, leaving them for oniy short spans of tirne (30
minutes on average) (S tangel and Jensen 199 1). The initiation of the weaning process
starts, at the earliest, 10 days postpartum with a decrease in suckling fiequency (Jensen
al 1991). This decreased suckling fiequency corresponds to when the sow and her litter 2-
leave the nest to move closer to the rest of the herd (Jensen and Redbo 1987).
Piglets graduaily become more independent. By weeks 34, they are more active,
engaging in ninaing and playing activities. Play fighting and aggressive encounters
between uiglets start to occur in weeks 5-6. Weaning cannot be defined as a specific time
but is a gradua1 shift Erom a reliance on the sow's miik to a reliance on other food. Entire
litters are not completely weaned at the same time. By 6 weeks of age, many unweaned
piglets receive less than half of their dry-matter intake f?om milk (English et al., 1988).
Descriptions of sows in semi-natural environments Vary with regard to when weaning is
completed. In a Scottish study, Stolba and Wood-Gush (1989) reported that aIi pigIets
were weaned by 12.5 weeks. In a Swedish population of pigs, Jensen and Recen (1989)
reported that weaning was not completed until 17.2 weeks. Bae (1991) found that with
sows and piglets housed in enriched pens, weaning was finished by 1 1-12 weeks.
The behavioural implications of abrupt artificial weaning at 3 weeks of age have
been weIl studied. Fraser (1978) observed that piglets weaned at 3 weeks showed an
increase in overall activity and aggression and appeared to have difficulty lying together
comfortably compared with piglets weaned at 6 weeks of age. This early work on the
behavioural effects of early weaning led into research about how piglets signal their
needs and what signals can be accurately used to assess piglet well-being (Weary and
Fraser 1995). Dybkjaer (1992) identified redirected oral behaviour in early-weaned
piglets, such as belly-nosing and oral manipulation of pen-mates' ears, tails and other
body parts, as possible indicators of 'stress'. There is, therefore, a good knowledge base
conceming the types of behaviour that may be useful in determining how disturbed,
distressed or frustrated early-weaned piglets might be. There are also a few reports on
the effects of very early weaning fiom 20 years ago (van Putten and Dammers, 1976; van
Putten, 198 l), but these studies pre-date advances in piglet nutrition and management and
so are less relevant.
Few studies have examined the behaviour of piglets weaned at 2 weeks of age or
younger. Metz and Gonyou (1990) compared the effect of age and housing conditions on
the behaviourd and immunological (neutrophiylymphocyte ratio) reaction of piglets to
weaning. The behaviour of piglets weaned at 2 and 4 weeks was compared both while
the piglets were with the sow and d e r weaning. There were many ciifferences in the
feeding and sucking behaviour of piglets weaned at 2 versus 4 weeks of age. The piglets
weaned at 2 weeks spent signiflcantly more time at the teat before weaning than those
weaned at 4 weeks. Weaning caused an increase in beily-nosing behaviour two days
post-weaning in both groups. More belly-nosing was performed by the early-weaned
animds. The authors suggest that nosing litter-mates in the weaned piglets acted as a
substitute for teat contact with the sow.
In order to gauge whether or not very early weaning leads to serious problems, ten
SEW nurseries in Ontario were surveyed and asked to describe any behavioural
abnormalities on their f m s (Dewey et al., 1997). Producers reported increases in the
arnount of belly-nosing, navel-sucking and aggression in piglets that had been weaned
early.
This study examines the effects in a controlled experiment of weaning at various ages on
piglet behaviour.
Materials And Methods
The 180 piglets used in this experiment were of pure-bred Yorkshire stock. Three
trials were conducted with nine litters in each triai. In each tri& three litters were
assigned at random to be weaned at 7 days, t h e at 14 days, and three at 28 days
postpartum. From each pool of three litters, two groups of 10 piglets were assembled by
weight and each group placed in a pen m e a s u ~ g 1.83 x 1.22 m, giving two pens per
weaning age per trial. Each pen had two nipple drinkers, two galvanized steel feeders,
two rubber mats, an infrared heat lamp and tender foot flooring. The temperature in the
room varied between 24 and 2 7 ' ~ and was monitored ddy. Ail of the pens were washed
and disinfected between each triai.
The study was approved by the animal case cornmittee at the University of
Guelph. Animals were cared for under the guidelines of the Canadian Council on Animal
Care ( CCAC 1993).
Three different types of starter feed were used in the experiment (Types 1,2 and
3) (TabIe 1, pp. 52). The piglets weaned at 7 days old were fed each type of starter for
two week periods in succession . The piglets weaned at 14 days old were fed Type 1
starter for one week, Type 2 for two weeks and Type 3 for two weeks. The piglets
weaned at 28 days old were fed Type 3 for two weeks. Type 1 starter was provided as a
creep feed in the farrowing crates to al1 of the pigiets fiom day 5 postpartum. Creep
feeders were used for the piglets weaned at 7 days for the fïrst week after weaning
because the animals were too smail to reach the feed in the large gdvanized steel feeders
used throughout the rest of the experiment.
The animals were d l weighed at birth and then weekly. Feed intake was
estimated weekly on a pen basis by measuring feed disappearance. AU piglets had their
teeth ciipped on &y 1 postpartum and were castrated and ear notched on day 4.
Observations
The piglets were videotaped for 48 h penods on different occasions. Piglets
weaued at 7 days old were videotaped on days 7 & 8,11& 12, 14 & 15,28 & 29, and 42
& 43. Piglets weaned at 14 days old were videotaped on days 14 & 15,28 & 29, and 42
& 43. Piglets weaned at 28 days old were videotaped on days 28 & 29, and 42 & 43.
The videotapes were scanned every five minutes to obtain an instantaneous behavioural
sample. Ten mutually exclusive behavioural categories were recorded and are shown in
Table 1 dong with descriptions.
S tatisticai Analvsis
The experimentai design was a completely randomized one-way ciassification
design. The number of piglets engaged in each of the ten behavioural categories was
counted for each scan, and adjusted for any missing piglets by determining the % of
piglets engaged in each behaviour at each sarnpling tirne. These percentages were then
averaged for the entire two-day observation penod. And fiom this, the amount of t h e
spent in each behavioural category was calculated and expressed as a least square mean
value. Analysis of variance was used to compare statistically the behaviour of piglets
weaned at 7 days of age with piglets weaned at 14 days of age on days 14 & 15,28 & 29
and 42 & 43. Piglets weaned at 7 and 14 days were compared with piglets weaned at 28
days old on days 28 & 29 and 42 & 43. The GLM (generai h e a r models) procedure was
used with SAS to perform the analysis (Statistical Analysis Systems Institute 1985).
Analysis of variance was also used to compare the amount of time spent performing each
behaviour for the fht 48 hours following weaning for each treatment group. The data
was aiso tested for normality (Statistical Analysis Systems Tiistitute 1985).
Results
The times spent by the piglets engaged in the ten mutually exclusive behaviour
categories are shown in Table 2. The most important of these, with regard to the amount
of time spent, were belly-nosing, feeding, general aggression, exploratory behaviour and
lying.
Figure 1 shows that in piglets weaned at 7 days, the incidence of belly-nosing,
although fairly low the day after weaning, had increased dramaticdy by days 1 1 & 12.
These piglets continued to belly-nose at a high level through to days 14 & 15 when it was
higher than that shown by piglets weaned at 14 days (PcO.0005) and through days 28 &
29 when it was higher than piglets weaned at 28 days (P<0.005). At 6 weeks of age the
piglets weaned at 7 and 14 days were still belly-nosing at a higher Level than the piglets
weaned at 28 days (Pc0.05).
Feeding Behaviour
Figure 1 shows that on the day of weaning and the foliowing &y the 3 treatment
groups spent varied amounts of time at the feeder. Pigiets weaned at 28 days spent 4.6 * 0.4% of their time at the feeder, piglets weaned at 14 days spent 2.8 1 .O% of their time
at the feeder, and pigiets weaned at 7 days spent 0.6 * 0.2% of their time at the feeder
ee0.05). At six weeks of age, the piglets weaned at al1 of the 3 weaning ages spent
similar arnounts of tirne at the feeder @O. IO).
Generai Aegression and Agmession at the feeder/drinker
The overall amount of generai aggressive encounters and aggressive encounters at
the feederldrinker did not m e r with weaning age. Figure 1 shows the general pattern of
aggressive behaviour in the 3 weaning ages.
Ex~loratorv Behaviour
During the fust 48 hours after weaning piglets spent significady different
amounts of time exploring dependhg on weaning age. Newly weaned 7 day old animais
spent 4.1 rt 0.71% of their tirne explorhg as opposed to 9.9 * 0.71% and 8.8 0.71% for
14 and 28 day weaned piglets respectively @ < 0.001). At two weeks of age piglets
weaned at 7 days spent less t h e exploring than piglets newly weaned at two weeks
@<0.0001). At 4 weeks of age, piglets weaned at 7 days spent less t h e exploring than
piglets newly weaned at four weeks @<0.001) (Figure 1). The pattern of temporal
changes in exploratory behaviour is similar across treatments. When piglets in al1 3
treatment groups were initially weaned into their pens, exploratory behaviour was at its
highest point and graduaiiy deched over the. This deciiie in exploratory behaviour was
sharper in piglets weaued at 7 days compared with piglets weaned at 14 and 28 days.
Social Nosinrr and Chewing
The piglets in al1 three treatment groups spent l e s then 3% of their time
exhibiting nosing and chewing behaviour during the six week observation period. Piglets
weaned at 7 7 28 days spent more tirne nosing and chewing pen-mates on day 28 & 29
than did piglets weaned at 14 days @<O.OS).
Esca~e Behaviour
Escape behaviour was seen almost exclusively in piglets weaned at 7 days of age.
Forty-eight hours following weaning, pigiets weaned at 7 days spent significantly more
time performing escape behaviour than piglets weaned at 14 @<0.0005) or 28 days
(P<0.005). On days 7 & 8 piglets weaned at 7 days spent 2.5% of the observed tirne
performing escape behaviour. Table 2 shows that as the piglets aged, the amount of
escape behaviour diminished.
Contact Behaviour
Following wedng, pigiets weaned at 14 and 28 days spent more time in social
contact behaviour than piglets weaned at 7 days @<0.005).
Drinkllig Behaviour
Pigiets weaned at 7 days spent siightiy more tirne at the nipple drinker than piglets
weaned at 28 days on days 28 and 29 (P < 0.10). PigIets weaned at 7 days spent 1.6 -1: 0.3%
of their time drinking on days 7 & 8. By week six, pigiets weaned at 7 days were spending
more tirne at the drinker than the later weaned animals (P4.05).
Lving Behaviour
Piglets in aii 3 treatments groups spent most of their time lying down (Table 2). Ali
of the animals in this study spent a p p r o h t e l y 70-80% of the observed time lying.
Discussion
Weaning age had a profound effect on piglet behaviour. The most ciramatic
ciifferences occurred in pigiets weaned at 7 days which spent much more tirne in activities
directed towards their pen-mates (belly-nosing and social nosing and chewing), more tirne in
escape behaviour, very smaii amounts of time at the feeder directIy after weaning, and
seemed to have less interest in their environment (decreased exploratory and contact
be haviour).
Piglets w m e d at 14 days showed simiiar behaviour patterns as piglets weaned at 28
days except that they showed more belly-nosing behaviour and they spent less time at the
feeder directly foliowing weaning. Other studies have found sirnilar results when
camparing piglets weaned at later ages. B a (1 993) found that sucking and massaging pen-
mates is proportionately more frequent when piglets are weaned at 4 weeks (sucking: 0.1 * 0.2% ; massaging 0.7 k 1.6% ) compared to 6 weeks (sucking 0.0 * 0.1% ; massaging 0.1 -1:
0.2% ). Mer behaviour such as sni£Eng and rooting at pen-mates, chewing and nibbling on
fittings, and tail biting, were performed at approximately the same rate by piglets weaned at
4 and 6 weeks of age.
Weaning at both 7 and 14 days was characterized by the occurrence of a large
amount of beiiy-nosing. This beiiy-nosing is undesirable since it can lead to skin lesions in
the piglets that receive it (Worobec et., 1997). However, before attempting to fmd ways
of eliminaîing or decreasing the belly-nosing behaviour, it is worthwhile to consider what is
motivating this behaviour to occur. The methods used to reduce the incidence of belly-
nosing may be very different depending on whether beiiy-nosing is part of the feeding or
social contact or escape motivational systems.
It seems most likely that belly-nosing is connected in some way to the feeding
motivationai system. The motor patterns involved appear very Sitnilar to the massagllig
movements that piglets direct towards the sow's udder both immediately before milk Iet-
down and for some tirne aîlerwards. In the case of feeding behaviour in calves, it has been
show that the appetitive element of sucking has become dissociated f?om the
consummatory element of ingestion of milk When a calf is aven milk fiom a bucket, it
stilI has a "need" to suck (Rushen and de Passillé, 1995). if it is provided with an artificial
dry teat, it will suck on it for some minutes afler drinking milk lÏom a bucket. Moreovei,
the performance of this non-nutritive sucking may have beneficial consequences for the caif
in that it stimulates secretion of certain digestive enzymes (de Passiiié et al., f 993). If belly-
nosing behaviour is some element of appetitive feeding in the piglet, such as redirected
massaging or pre-sucking behaviour, then it may be important for piglets to be abIe to
perform this behaviour. The solution may not be to prevent the occurrence of the motor
patterns but to direct them such that their perf0111lil~1ce does not cause injury.
The speed with which animds begin to feed on solid food is an important
indicator of how well they are adapting to weaning. Pigiets weaned at 7 days spent less
than 1% of their time at the feeder in the week following weaning compared with 5% of
their time when weaned at 28 days suggesting that the early-weaned piglets were not
adapting as well. Fourteen-day-weaned piglets spent 3% of their time at the feeder the
week of weaning. This suggests that piglets weaned at two weeks can adapt to dry feed
more easily than those weaned at one week but that work still needs to be doue to
motivate these animals to begin feeding. The prime Iimiting factor to performance
directiy after weaning is the delay of the initiation of feeding and the limited consurnption
of solid food in the week following weaning pettigrew et., 1994). Therefore, methods
of stimulating feeding behaviour directly following weaning will greatly contribute to the
welfare and performance of early-weaned piglets.
Reduced welfare would also be indicated by the performance of behaviour
associated with States of suffering. Escape behaviour may be an indicator of distress or
fear in young piglets. It is important to note that escape behaviour was seen only in
newly weaned 7 days old piglets. When it did occur, it was accompauied by high levels
of vocalizations. Over the last few years the analysis of piglet vocalizations has been
used as an indicator of well-being. Weary et al., (1996) found that piglets which were the
smallest and slowest growing (needy piglets) vocaiized at a higher rate and higher
kquency, and for a longer duration than "un-needy" piglets.
It has been suggested that animais that are stressed may pay less attention to theu
environment than animals that are relaxed (Wkpkema et., 1983). Therefore the amount
of exploratory behaviour and social contact that piglets show may be positively correlateci
with their weKare. Piglets newly weaned at 7 &YS spent only haif the time explorhg
compared with piglets newly weaned at 14 or 28 days. Social contact in this study is
defined as nose-to-nose contact with piglets in a neighbouring pen. Piglets newly weaned at
7 days performed seven times less contact behaviour than did pigiets newly weaned at 14
days and nine times less contact behaviour than did piglets newly weaned at 28 days. The
difference in exploratory and social contact behaviour seen in pigiets weaned at 7 days
compared with the later weaned piglets suggests strongly that very early weaned piglets
may be stressed, more apathetic, and less responsive to their environment. This
unresponsiveness or "apathy" was described by several SEW nurseries as ''the loss of the
will to iive" (Dewey et., 1997) and is a potentiai weKare concern.
The lack of establishment of stable dominance hierarchies in very early-weaned
piglets may also be cause for concern. Although general aggression levels were similar
when piglets were first weaned at different ages, the overall trend was different. The
piglets in al1 three treatrnent groups displayed similar peak levels of general aggression
when they were initially grouped together following separation fiom the sow. This
aggression decreased over time in the 14 and 28 day weaned animais. However piglets
weaned at 7 days did not show t h s steady decline in general aggression with age. The
resultant increased level of social fiction has the potential to reduce the welfare of these
piglets. More research is required to determine the exact age at which social mixing will
ailow the formation of a stable hierarchy.
In summary, it can be said tbat pigiets do not adapt weII to being weaned at 7
days. The merences in behaviour that were seen in these piglets cornpared with that
seen in piglets weaned at 14 and 28 days suggests that very eary weaning causes them
severe problems and greatly reduces their welfitre. Piglets weaned at 14 days seem to
adapt more easily to weaning, but still have some problems compared with piglets
weaned at 28 days. In fact, there has been a trend in SEW famis in Canada over the past
few years to wean slightiy later, at 16-17 &YS rather than 12-14 days. The present study
suggests tbat this trend to ctn older weaaiog age may have positive b e h a v i o d
ramifications and welfare benefits.
I f earIy weaning systems, induding SEW systems, are to become successfûl
commercially protect the welfare of the piglets, then the most pressing need is to find
a way to motivate the piglets to consume food as soon afkr weaning as possible.
Acknowledgements
The financial support of the Ontario Ministry of Agriculture, Food and Rural Affairs and
of Ontario Pork is appreciated. Many thanks to the staff of the Arkell Swine Unit,
University of Guelph.
TabIe 1. The mutually exclusive behaviour categories that were recorded during the
experiment together with their descriptions.
Feeding Dfhkin9;
--
General aggression
Aggression at the
feeder and drinker Escape
Exploratory behaviour
Social contact
Social nosing and chewing
Body weight not supported by legs Head in feeder Snout against nipple drinker A distinctive sequence in which one piglet
rubs a pen-mate's belly with rhythmic, up
and down rnovements of its snout (Fraser
Pushing, head-thnisting, biting and chasing (Blackshaw 1987). Forcetid displacement from the feeder or drinker
Jumping against the sides of the pen with
hind feet on the ground and fore feet on the
side of the pen.
Oral-nasal contact with parts of the pen or
objets in the pen
Nose-to-nose contact with piglets in a
neighbouring pen.
Nosing and chewing at the body of a pen-
mate (distinct fiom belly-nosing) (Fraser
Table 2: The percentage of the observed time (* s.e.m.) pigiets exhibited various behaviours
Days 7 & 8 I l & 12 14 & 15 28 & 29 42 & 43
Feeding Wean 7 0.6*0.5, 5.7 0.9 6.5 1 0 9 8.6 0.5' 6.1 * O S b
Wean 14
Wean 28
Expiore Wean 7
Wean I4
Wean 28
Belly-nosing
Wean 7
Wean 14
Wean 28
Drinking
Wean 7
Wean 14
Wean 28
Gen. Agg
Wean 7
Wean 14
Wean 28
Agg. FeedDrink
Wean 7
Wean 14
Wean 28
Days
Wean 7
Wean 14
Wean 28
Contact
Wean 7
Wean 14
Wean 28
Nose & Chew Wean 7
Wean 14
Wean 28
Lyhg
Wean 7
Wean 14
Wean 28
The letter supe-pts which appear in the columns and the numerid subscnpts that appear across columns ( boided
Different lener superscripts within colurnns indicate statistical difference @< 0.05)
Dclvs Wean 7
Wean 14
Wean 28
7-14 14-21 21-28 28-36 36-43 14-43 28-43 2.0 * 0.4 1.4 * 0.2" 1.9 & 0.7 1 .O A O. 1 1.2 k 0.9 1.4 * O. 1 1.4 i 0.1
0.2 * 0.2 2.1 * 0.7 0,9 I O. 1 1.3 * 0.1 1.4 * 0.1 1.3 rt0.I
0.6 ,: O. 1 1.2 * 0.1 .. 1.3 * 0.1
-e- Wean 14
W e a n 28
Figure 1 : Bodyweight (kg) both before and after weaning over the six week experimental period. Different letter vertical superscripts indicate skttisticai difference: a b (P ~0.05); a* b @<O.OOS).
Arrows point IO the Lhree weaiung ages.
Figure 2: The mean skin lesion scores for each weaning age both before and afier weaning . Arrows represent the three weaning ages.
Wean 7
Score 3 Score O
Wean 14
Wean 28
Score 3
Score 1
Figure 3: The percentage distribution of skin lesion scores one week after weaning for each treatment group.
Effect of segregated early weaning (SEW) on sow behaviour
To be submitted to the Journal of Applied Animal Behaviour
Abstract
Segregated early weaning (SEW) is a mine management system in which
piglets are removed h m the sow at an eariy age and tmsported to a separate growing
facility. Average weaning aga in SEW systems range h m 16-17 days to as low as 7-
10 days postpartum. North Amencan commercial swine operations are rapidly shifting
to SEW systems with very little information about the behavioural and weIfare
implications of this switch. The behaviour and weKm of sows in early weaning
systems has been vimially ignored.
The objective of this study was to observe the behavioural effects of separation
from piglets and reduced lactation length on sows. The snidy was conducted on a
commercial SEW sow breeding and gestation f m . Two trials were conducted.
Twenty-four sows were weaned at 12-15 days of lactation (early-weaned) and twenty-
four sows were weaned at 19-22 days of lactation (Iate-weaned). m e r weaning the
sows were placed in pens in groups of six with two pens per treatment group per trial.
The sows were videotaped on days one, two, five and six pst-weaning. The weanhg-
to-breeding interval for each sow was recorded.
The behaviour of eariy- and late-weaned sows was very similar. The amounts of
time spent performing social behaviour (social grooming, nosing penmates, sniffing
Table 2: The percentage of observational tirne the sows spent perfoming various behaviours
Day 1 Day 2 Day 5 Day 6 FEEDINC Early Weaned Late Weaned
DEUMCWG Early Weaned Late Weaned
SOCLAL EarIy Weaned Late Weaned
Gmoming Early Weaned Late Weaned Nosing wnmates Eariy Weaned Late Weaned Snifiinp enm mates Early Weaned Late Weaned Mounting Early Weaned Late Weaned
POSTURE Early Weaning Late Weaning
Sitting Early Weaning Late Weaning Standing Early Weaning Late Weaning Ventral Lving Early Weaning Late Weaning Lateral Lving Early Weaning Late Weaning
ACGRESSION Early Weaning Late Weaning
Head knocking EarIy Weaning Late Weaning
Threat EarIy Weaning Late Weaning Attack Early Weaning Late Weaning Fight Early Weaning Late Weaning EXaE Early Weaning Late Weaning
ABNORMAL Early Weaning Late Weaning
Bar biting Early Weaning Late Weaning Sham Chewing Early Weaning Late Weaning
EXPLORATORY Early Weaning Late Weaning
The letter superscripts represent statistical diflkrences @< 0.05)
Standing
2 5
Days post-weaning
Ventral Lying
2 5 Days post-weaoing
Laterai Lybg
2 5 Days post-weaning
Figure 1 : The percentage of t h e the sows spent standing, tying venaally and lying laterally
GENERAL DISCUSSION
In the experiments d e s c r i i in this thesis, age at weaning affected the behaviour
and production of pigiets but not sows. Early weaned pigiets budgeted their time differently
than did pigiets weaned at 28 days, both directly after weaning and up to six weeks of age.
Piglets weaned at seven days of age spent 10% of their time beiiy-nosing a few days atler
weaning, spent Iess then 1% of their time at the feeder, performed a type of distress
behaviour named "escape behaviouf and spent oniy 1% of their time exploring their
environment At six weeks of age these early-weaneci piglets continued to spend more time
bdly-nosing than the 28-day-weaned animals. This dtered behavioural time budget in
conjunction with low feed intakes and weight gains suggests a weLfare problem for piglets
weaned at seven days of age. The time budget of piglets weaned at 14 days was different
h m piglets weaned at 28 days. Piglets weaned at 14 days spent more tirne exploring
directiy after weaning, less t h e at the feeder and p e r f o d more belly-noshg behaviour
than piglets weaned at 28 days.
The motivation behind beliy-nosing is still unclear, however this research suggests a
strong comection between age at weaning and belly-nosing. Piglets weaned at earlier ages
displayed greater amounts of belly-nosing. Beliy-nosing behaviour appeared when the very
early-weaned piglets (7 days) started to increase their c o m p t i o n of feed. Thetefore intake
of feed rnay be associated with the motivation of beiiy-nosing behaviour. n i e implications
of increased levels of belly-nosing are undetermined. There may be a relationship between
belly-nosing and the skin integrity of piglets. Piglets weaned at 7 and 28 days had a higher
frequency of skin markings one week &er weaning than piglets weaned at 14 days. The
type of markings found on pigIets weaned at 7 days seemed to be more "suck" marks as
oppsed to scratches resulting fkom aggression seen on the pigiets weaned at 28 days.
ûther performance indicators such as feed intake, weight gain and feed-to-gain ratio
were also affected by weaning age. By week six, piglets weaned at 7 and 14 days were
consuming more feed than the 28 day weaned piglets but the 7-day-weaned piglets weighed
significantly less than the later weaned animds. Piglets weaned at 7, 14 and 28 days and
kept on-site had simiiar feed efficiençies during the last two weeks of the observation period
(weeks 4-6).
Early weaned pigiets must be stimulated to feed directly d e r weaning to improve
their feed efficiency. Phase feeding systems have been developed to meet the nutritional
demands of the early-weaned piglet (Dritz et ai., 1997). However, the behaviouraI
abnormalities such as high levels of beUy-nosing and low feed consumption observed in
early-weaned pigiets suggest that even if their physiologid needs are king met their
behavioural needs are not.
This study demonstrated an inverse relationship between weaning age and altered
behaviour patterns. Piglets weaned at 7 days displayed the most severe behaviour
ciifferences foliowed by 14-day-weaned pigIets and the control28-day-weaned animals. B e
(1993) observed the same type of graded response to weaning with the amount of redirected
oral behaviours performed by pigiets weaned at four as opposed to six weeks.
If piglets display behavioufal abnormalities and poor performance as a result of
early-weaning, how does it affect the sow? Few studies have considered the behavioural
ramifications of the premature removal of piglets fiom the sow. Pajor et (1996) observed
the individual variation in sows weaning behaviour when given a chance to separate
foiiowing weaning. Pigiets' performance suggests again that 7-day-weaning is too young
and that 14-day-weaned piglets can compete with 28-day-weaned animais but they need
encouragement to feed directiy foiiowing weaning. Sow behaviour and performance were
not affected by weaning age in this study. Taking into consideration the cwrent knowIedge
of both piglet and sow behaviour and performance, weaning between 14-21 days is
recommended.
Aiexander, T.J.L., Thomton, K., Boon, G., Lysons, R.J. and Gush, A.F. 1980. Medicated early-weaning to obtain pigs k e h m pathogens endemic in the herd of origin. Vet-Rec. 106: 114.
Algers, B. and Jensen, P. 1985. Communication during suckhg in the domestic pig. Effects of continuous noise. AppI.Anim.Behav.Sci. 14: 49.
Aigers, B. and Jensen, P. 1991. Teat Stimulation and m i k production in early lactation of sows. Can.J.Anim.Sci. 71 :5 1.
Aigers, B. 1993. Nursing in pigs: Comrnunicating needs and distributhg resources. J.Anim.Sci. 71: 2826.
Arellano, P. E., Pijoan, C ., Jacobsen, L.D. and Algers, B. 1992. Stereotyped behaviour; Social interactions and suckling patterns of pigs housed in groups or in single crates. Appl. Anim. Behav. Sci. 35: 157.
Bark, L.J., Crenshaw, T.D. and Leibbrandt, V.D. 1986. The effect of meal intervais on feed inîake of early weaned pigs. J.Anim.Sci. 62: 1233.
Barnett, J. L., Wuifield, CG., Cronin, G.M. and Dewar, A.M. 1984. The weIfare of adult pigs: the effits of five fiousing treatments on behaviour, plasma corticosteroids and injuries. Appl. Anim. Behav. Sci. 12: 209.
Blackshaw, J.K. 198 1. Some behavioural deviations in weaned domestic pigs: persistent inguina1 nose thrusting, and tail and ear biting. Anirn-Prod. 33: 325.
Bne, K., 1 99 1. The process of weaning in pigs: when the sow decides. Appl. Anim. Behav. Sci. 30: 47.
B B ~ , K., 1993. The effect of age at weaning and post-weaning environment on the behaviour of pigs. Ph.D. thesis, Agricultural University of Norway. Publication 3: 1 o.
Brent, G., Hoveil, D., Ridgeon, R.F. and Smith, W.J. 1977. Early-weaning of pigs. Farrning Press Ltd. Ipswich, Suffolk. pp. 124.
Canadian Council on Animal Care. 1993. Guide to the care and use of experimental animals. Ed by Olfert, E.D., Cross, B.M., McWilliam, A.A. pp.34.
Clark, K.L., 1 995. SEW: program, problems, performance, potential profits and methods of implementation for various herd sizes. Proceeding of the George A. Young Swine Conference, University of Nebraska, Lincoln, Nebraska, U.S.A. pp. 1.
C h e , T.R. 199 1. Feeding pigs weaned at 3 to 4 weeks of age. In: Swine Nutrition. Chapter 30. Ed by M e r , E.R, U h y , D.U. and Lewis, A.J. pp. 497.
Csemely, D. and Wood-Gush, D.G.M. 198 1. Artificial stimulation of ingestive behaviour in early-weaned piglets. BiolBehav. 6: 159.
Dee, 1995. Sow productivity before and after SEW. PIGS- Misset, June-July. pp. 13.
Deen, 1. 1994. Controllhg the costs of early weaned sows. Proceedings of the George A. Young Swine Conference. pp. 1.
de Passiiié, A. M. B. and Robert, S. 1989. Behaviour of lactating sows: inauence of stage of lactation and husbandry practices at weaning. Appl. Anim. Behav. Sci. 23: 315.
de Passillé, A. M. B., Robert, S., Dubreuil, P., Pelletier, G. and Brazeau. P. 1990. Effect of hypothalamic factor treatments on the behaviour of sows during lactation and on their behavioural cortisol responses to weaning. Appl. Anim. Behav. Sci. 27: 23 1.
de PassiIlé, A.M.B., Christopherson, R.J. and Rushen, J. 1993. Nonnutritive sucking and the postprandial secretion of insulin, CCK and gastrin in the caif. Physiol. Behav. 55: 1069.
Dewey, C., Deckert, A. and Friendship, RM., 1997. Ontario SEW Survey Results. Centralia Swine Research Update, Kirkton, Ontario, Canada. pp. 155.
Dial, G., Moore, C., Wiseman, B., Connor, J. and Aheme, F. 1995. Factors influencing SEW success (Part 2). Pigletter, Dec:.37.
Dio ~ssopoulos, L., delange, C.F.M., Dewey, C., Friendship, R.M. and Maches , J. 1997. Effect of management strategy during rearing on grower-finisher pig performance and uIdicators of immune system activation. Ontario Swine Research Review. pp. 13.
Dritz, S.S., Nelssen, J.L., Goodband, RD., Tokach, M.D. and Chengappa. M.M. 1994. Establishing and maintaining controlled health status in M e production systems. Proceedings of the Canadian Veterinary Medicai Association. Quebec City. pp. 441.
Dritz, S.S., Tokach, M.D., Goodband, R.D. and Nelssen, J.L. 1997. Nutrition programs for SEW pigs- Part II. Compendium. Food animal medicine and management. 19: 10.
Duncan, 1. J. H. 198 1. Animal rights - animai welfâre: A scientist's assessrnent Poult. Sci., 60: 489.
Duncan, I.J.H. 1996. Animal Welfàre defined in terms of feelings. Acta Agric-Scand. Sect A, Suppl. 27, pp. 29.
Dybkjaer, L. 1992. The identification of behavioural indicam of stress in early weaned piglets. Appl. Anim. Behav. Sci. 35: 135.
Ellendorff, F., ForsLing, M.L. and Poulain, D.A. 1982. The mdk ejection reflex in the pig. J.Physiol.333: 577.
English, P.R., Fowler, V.R., Baxter, S. and Smith, B. 1988. In: The growing and finishing pig: lmptoving efficiency. Farming Press, Ipswich. pp. 555.
Ensminger, M.E. and Parker, R.O. 1984. Swhe Science. The interstate printers and publishers hc . Danville, Ihois .
Fraser, D., 1978. Observations on the behav iod development of suckling and early- weaned piglets during the £kt six weeks d e r birth. Anim.Behav. 26: 22.
Fraser, D. 1980. A review of the behavioural mechanism of mille ejection of the domestic pig. Appl.Anim.Ethol.6: 247.
Friendship, KM. 1987. A reduction in litter size- a case study. Can.Vet. J. 28: 120.
Foxcroft,. G.R. 1997. Reproductive implications of early-weaning. Proceedings of the Satellite Conference. Prairie Swine Centre. pp. 9.
Gesell, A. and Ilg, F.L. 1979. Feeding Behaviour of Infants. LB. Lippincott Company. pp. 116.
GU, J-C. and Thomson, W. 1956. Observations on the behaviour of suckling pigs. Br. J.Anim.Behav. 4: 46.
Gonyou, H. W. and Whittington, D.L. 1997. Segregated Early Weaning: effects of weaning at12 days of age on subsequent behaviour of piglets. Proceedings of the
North American Regional Meeting of the International Society for Applied Ethology. Lennoxville, Quebec, Canada. p. 5.
Goodband, R.D., Tokach, M.D., Dritz, S.S. and Nelssen, J.L. 1995. Practical Nutrition for the Segregated Early Weaned Piglet. in: Proceedings of the Saskatchewan Pork Industry Symposium. Saskatchewan. pp. 15.
Haley, D .B ., Rushen, J., Duncan, 1. J.H. and dePassiIie, A.M.B. 1996. What makes a calf butt? in: Proceedings of the 3on International Congress of the International Society for Applied Ethology (Ed by Duncan, 1.J.K Widowski, T.M. and Haley, D.B.). Guelph. pp. 97.
Houpt, K. 1982. Domestic animai behaviour for veterinarians and scientists. University Press. Ames: Iowa State. U.S.A.
Jensen, P. and Redbo, 1. 1987. Behaviour during nest leaving in he-ranging domestic pigs. Appl.Anim.Behav.Sci. 1 8: 355.
Jensen, P- and Recen, B., 1989. When to wean - observations h m fiee-ranging domestic pigs. App. Anim. Behav. Sci. 23: 49.
Jensen, P., Stangei, G. and Algers, B. 1991. Nursing and suckling behaviour of semi- nattiraily kept pigs during the 6rst 10 days postpartum., Appl. Anim. Behav. Sci. 31: 195.
Kuehl, R.O. 1994. Statisticai principles of research design and anaiysis. Duxbury Press. California. U.S. A
Leibbrandt, V.D., Ewan R.C., Speer, V.C. and Zimmermaa, D.R 1975. Effect of weaning and age at weaning on baby pig performance. J.Anim.Sci. 40: 1077.
Lewis, N.J. and Hurnik, J.F. 1985. The development of nursing behaviour in swine. Appl.Anim.Behav.Sci. 14: 225.
Marchant, J. N., Mendl, M.T., Rudd, A.R. and Broom, DM. 1995. The effect of agonistic interactions on the heart rate of group-housed sows. Appl. Anim. Behav. Sci. 46: 49.
Martin, P.H. and Bateson, P.P.G. 1985. The influence of experimentaüy manipulation a component of weaning on the development of play in domestic cats. Anim.Behav. 33: 511.
Mason, G. 1996. Eariy-weanuig enhances the Iater development of stereotypy in mink. Proceedings of the International Society for Applied Ethology, Guelph. pp. 16
Maurer, R. R, Ford, JJ and Christenson, RK. 1989. Interval to fïrst pst-weaning oestrus and causes for leaving the breeding herd in Landrace, Large White, Yorkshire, and Chester White females after 3 litters. Swine Res. Prog. Rep. 3: 44.
McBride, G. 1963. The "teat-order" and communication in young pigs. AnimBehav. 11: 53.
Metz, J.H.M. and Gonyou, H.W., 1990. Effect of age and housing conditions on the behavioural and haemolytic reactions of piglets to weaning. Appl. Anim. Behav. Sci. 27: 299.
Newberry, R.C. and Wood-Gush, D.G.M. 1988. Development of some behaviour patterns in piglets under semi-naturai conditions. Anim.Prod. 46: 103.
Pajor, E.A., Weary, D.M., Fraser, D. and Kramer, D.L. 1996. individuai ciifferences and housing effects on the response of sows to weaning. In: Proceedings of the International Society for Applied Ethology, Guelph. pp.33.
Patience, J.F., Thacker, PA. and de Lange, C.F.M. 1995. in: Swine Nuûition Guide (Second Edition). Ptaicie Swine Centre hc. pp. 12.
Petrie, C.L. and Gonyou, H. W. 1988. Effects of auditory, visual and chernical stimuli on the ingestive behaviour of newly weaned piglets. J.Anim.Sci. 66: 66 1.
Pettigrew, J.E., Johaston, L.J., Shurson, G.C. and Hawton, J.D. 1994. Feeding nursery pigs. Proceedings of the Arnerïcan Association of Swine Practitioners, Rosemont, Illinois, U.S.A.
Pollard, J.C., Littlejohn, R.P. and Suttie, J.M. 1992. Behaviou. and weight change of red deer calves during different weaning procedures. Appl.Aaim.Behav.Sci. 35: 23.
Rushen, J. and de Passillé, A.M., 1995. The motivation of non-nutritive sucking in calves, Bos taum. Anirn. Behav. 49: 1503.
Schaefer, A.L., Saiomons, M.O., Tong, A.K.W., Sather, A.P. and Lepage, P. 1990. The effect of environment enrichment on aggression in newly weaned pigs. Appl.Anim.Behav.Sci. 27: 41.
Scholman, G.J, Sas, H.M. and van Wesel, A.A.M. 1996. Exploring inside the prestarter piglet. PIGS-Misset. 12: 29.
Schouten, W.G.P. 1986. Rearing conditions and behaviour in pigs: Dissertation, Agricultural University of Wageningen, The Netherlands.
Smith, E. 199 1. The influence of nutrition and postpartum mating on weaning and subsequent play behaviour of hooded rats. Anim.Behav. 41 : 5 13.
Spinka, M., Ilimam, G., Stëtkov5, Z., Krejci, P., SeciMc, L., Lidiclq', J. and Minarik, P. 1 996. The innuence of intervals between nursing and udder massage on hormone levels in the domestic sow. Proceedings of the International Society for Applied Ethology, Guelph. Canada pp. 141.
Stangel, G. and Jensen, P., 1991. Behaviour of semi-naturally kept sows and pigiets (except suckling) during 10 days post-partum. Appl. Anim. Behav. Sci. 3 1 : 21 1.
Statistical Analysis Systems Institute, 1985. SAS. User's Guide: Statistics. SAS. Mtute , Cary, NC.
Stolba, A. 198 1. A M y system in enriched pens as a novel method of pig housing. Alternatives to Animal Husbandry Systems. Potters Bar, UK. pp. 57.
Stolba, A. and Wood-Gush, D.G. 1989. The behaviour of pigs in a semi-natural environment. Anim.Prod. 48: 4 19.
Stott, G. H. 1981. What is animal stress and how is it measured? 1. Anim. Sci., 52: 150.
Taylor, RE. 1988. ScientSc f m animal production. Macmillan Publishing Company, New York, U.S.A. pp. 322.
Tokach, M.D., Goodband, R.D. and Nelssen, J.L. 1994. Recent developrnents in nutrition for the early weaned pig. Compendium of Food Animal Medicine and Management. Marck 407.
van Putte- G. and Dammers, J., 1976. A comparative study of the weli-being of piglets reared conventionally and in cages. Appl. Anim. Ethol. 10: 339.
von Borell, E., and Humik, J.F. 199 1. Stereotypic behaviour, adrenocortical function, and open field behaviour of individually confined gesbting sows. Physiol.and Behav. 49: 709.
van Putten, G. 198 1. Welfare of piglets in flatdeck-cages. In: The welfare of pigs. Ed by Sybesma, W., Nijhoff, M. The Hague. pp. 185.
Walker, RD, and Wiseman, B.S. 1994. Growth perforrnauce of segregated early weaned (SEW) pigs compared to their conventionally weaned littermates. J.Anim.Sci. 72: 3 77.
Waran, N.K. and Broom, D.M. 1993. The influence of a barrier on the behaviour and growth of early-weaned piglets. Anim.Prod. 56: 1 15.
Weary, D.M. and Fraser, D., 1995. Signahg need: costly signals and animal welfare assessment. App. Anim. Behav. Sci. 44: 159.
Weary, D.M., Lawson, G.L. and Thompson, B.K., 1996. Sows show stronger responses to isolation c d s of piglets associated with greater levels of pigiet need. Anim. Behav. 52: 1247.
Welch, A.R and Baxter, M R 1986. Responses of newbom piglets to thermal and tactile properties of their environment ApplAnh.Behav.Sci. 15: 203.
Wiepkema, P.R., Broom, D.M., Duncan, I J.H. and van Putton, G. 1983. Abnormal Behaviom in Farm Animals, Commission of the European Communities, Brussels. pp. 1.
Williams, NH., Stanley, T.S. and Zimmermaa, D R 1994. Impact of immune system activation on growth and amino acid needs of pigs from 6 to 1 14 kg bodyweight. J.Anim.Sci. 72: 57 .
Wood-Gush, D.G.M. and Cserrnely, D. 198 1. A note on the diumal activity of eariy- weaned piglets in flat-deck cages at 3 and 6 weeks of age. 33: 107.
Worobec. E.K., Duncan, I.J.H. and Widowski, T.M. 1997. Effects of age at weaning on belly-nosing behaviour in pigiets. Proceedings of the International Society for Applied Ethology Conference, Prague.pp. 245.
Worsaae, H. and Schmidt, M. 1980. Plasma cortisol and behaviour in early weaned piglets. Acta.Vet.Scand. 21 : 640.
Wurbel, H. and StaufFacher, M. 1996. Age and weight at weaning affect weaning stress and the development of stereotypy in laboratory mice. In: Proceedings from the International Society for Applied Ethology. Guelph. pp. 18.
Xue, J. L., Dial, GD., Marsh, W.E., Davies, P.R. and Momont, H.W. 1993. Muence of lactation length on sow productivity. Livest. Prod. Sci. 33: 253.