Do pigs ®nd a familiar odourant attractive
in novel surroundings?
J.B. Jones
a, C.M. Wathes
a,*, R.P. White
a, R.B. Jones
b aSilsoe Research Institute, Wrest Park, Silsoe, Bedfordshire MK45 4HS, UKbRoslin Institute (Edinburgh), Roslin, Midlothian EH25 9PS, UK
Accepted 16 May 2000
Abstract
Sixty-four DurocLandrace pigs (Sus scrofa) of both sexes were reared from birth to 6 weeks of age in either fresh air or an atmosphere containing an arti®cial odourant (Ambi-Pur). Their behavioural responses to a novel environment (a modi®ed open ®eld) were then observed over 1 h when the test arena contained either the familiar Ambi-Pur odourant or ammonia gas at 45 ppm. Although members of other mammalian and avian species are attracted to familiar odours in otherwise novel surroundings, the presence of Ambi-Pur had no effect on the pigs' readiness to enter the novel arena from a sheltered area or on other fear-related behaviours following entry (p>0.05). Pigs kept in fresh air from weaning until test were generally less active and spent less time at the centre of the open ®eld if the test was carried out in an ammoniated atmosphere rather than fresh air; this suggests that they found ammonia aversive. On the other hand, the presence of ammonia at test accelerated approach to the centre of the open ®eld and increased the number of entries into that area and the percentage of time spent there as well as overall activity in pigs that had also been exposed chronically to this gas at 43 ppm from weaning until testing at 6 weeks (p<0.05). Thus, after chronic exposure to either ammonia gas or an arti®cial odourant, ammonia was the only stimulus that appeared to act as an attractive agent for pigs placed in an otherwise novel environment.#2000 Elsevier Science B.V. All rights reserved.
Keywords: Pig; Ammonia; Familiarity; Odourants; Olfaction; Novel environment
1. Introduction
Laboratory rodents and chickens are attracted to familiar odourants in otherwise novel surroundings (e.g. Devor and Schneider, 1974; Carr et al., 1979; Galef, 1981; Jones and Faure, 1982; Jones and Gentle, 1985; Burne and Rogers, 1995; Porter and Picard, 1998;
Applied Animal Behaviour Science 70 (2000) 115±126
*Corresponding author. Tel.:44-1525-860000; fax:44-1525-861735.
E-mail address: christopher.wathes@bbsrc.ac.uk (C.M. Wathes).
0168-1591/00/$ ± see front matter#2000 Elsevier Science B.V. All rights reserved.
Jones and Carmichael, 1999). The presence of a familiar odourant also reduced fear in domestic chicks tested in an open ®eld (Jones and Gentle, 1985). Though pigs are attracted to odourants associated with a familiar food (e.g. Hafez and Signoret, 1969; Campbell, 1976; McLaughlin et al., 1983) or a familiar animal (e.g. Meese et al., 1975; Jeppesen, 1982; Morrow-Tesch and McGlone, 1990; Parfet and Gonyou, 1991), it is not known whether they are also attracted towards familiar odourants associated with their home environment or if these would reduce fear when they were encountered in an otherwise novel arena. A reluctance to emerge from a sheltered environment into an exposed, unfamiliar one is widely considered to re¯ect timidity in both avian and mammalian species (Russell, 1979; Jones, 1987). Similarly, behavioural inhibition in a novel environ-ment is widely considered to be positively associated with fear (e.g. Gray, 1987; Jones, 1996). Our use of a start box from which the pig could emerge into a novel arena in the present study enabled us to address both these issues. Thus, we investigated if pigs entered a novel environment more readily and showed less fear-related behaviour if an odourant with which they had been reared Ð and which was hence presumably familiar to them Ð was present. We used a commercially available liquid air freshener (Ambi-Pur) that provided a melange of ¯owery odours.
Poor air quality may interfere with the detection of other chemical stimuli either by masking during acute exposure or by desensitisation of receptors following chronic exposure. Previous work has shown that chronic exposure to 20 or 40 ppm ammonia gas interferes with the reception and recognition of pheromones regulating reproductive behaviour and performance in pigs (Malayer et al., 1988) and reduces sensory acuity for n-butanol in some pigs (J.B. Jones, unpublished). Herein, we examine if exposure to
44 ppm ammonia interferes with the perception of a familiar odourant.
2. Materials and method
2.1. Animals, husbandry and accommodation
Sixty-four DurocLandrace crossbred pigs (Sus scrofa domestica) of both sexes were sourced from eight litters, two litters at a time. The piglets were weaned at 3±4 weeks of age. Each litter was reared in a farrowing ark, which was sited at an outdoor commercial breeding unit. Each ark had a volume of 4.1 m3, was bedded with wheat straw and was lit naturally.
After weaning, each litter of eight pigs was housed in a holding room of volume 23.7 m3. The rooms were bedded with wheat straw, which was replaced every 3 or 4 days to minimise the release of atmospheric pollutants. The rooms were lit arti®cially between 06.00 and 22.00 h by a single ¯uorescent tube. A weaner feed (Primary Select, Primary Diets Ltd., UK) and water were provided ad libitum. The pigs were free of clinical signs of respiratory disease before, during and after the experiment.
2.2. Experimental design
exposed continuously to the arti®cial odourant in the farrowing ark so that the animals would associate this odour with their home environment. Phase 2 lasted from weaning until between 14 and 19 days later at which time the pigs were tested in the open ®eld. In Phase 2, the exposure to the arti®cial odourant was the same as in Phase 1 but each litter was also exposed continuously to either fresh air or approximately 43 ppm ammonia to test for desensitisation. Each litter of eight pigs was subdivided in Phase 3, which comprised the period of the open-®eld test. In Phase 3, all four combinations of acute exposure to the arti®cial odourant and ammonia were used, thereby testing for masking of the arti®cial odourant by ammonia. Thus, four litters were divided into 16 groups to test various treatments. This protocol was repeated two times with eight litters in total, giving a total of four pigs on each of 16 treatments (n64).
2.3. Artificial odourant presentation and exposure
Both the farrowing arks and the holding rooms were ventilated mechanically in order to introduce either an arti®cial odourant and/or ammonia into their atmosphere. We used an Table 1
Design of the experiment, showing which groups of pigs were exposed to either a familiar artificial odourant (O)
and/or44 ppm ammonia (A) during each phasea
aSuperscripts `' and `ÿ' indicate exposure to O and A or no exposure, respectively.
arti®cial odourant in order to minimise the likelihood that it would resemble odours associated with food, social interactions, predatory encounters, etc. and because of its ready availability and constancy.
The arti®cial odourant was a proprietary liquid air freshener (Ambi-Pur fresh bouquet perfume, Sara Lee Corp., UK). This smelled of a melange of ¯owery odours, at least to the human nose. This odourant was released into the home environment, i.e. the farrowing ark and holding room of four of the eight litters of pigs, from birth to maximise the likelihood that the pigs would come to associate it with their familiar `home' environment. Olfactory acuity is particularly high in pigs and it is considered inconceivable that the pigs used here failed to perceive this odourant; it was perceived readily at the concentration used by humans. We did not formally test if the pigs found the Ambi-Pur odourant aversive but we saw no avoidance or disgust responses during its presentation.
An odourizer was constructed to release the arti®cial odourant into the atmosphere of the farrowing ark. The arti®cial odourant was volatilised from a wick, which was replaced every second day and was bathed in the liquid odourant. The liquid receptacle was heated to 658C in an aluminium block. The odourant from the odourizer was discharged into an open-ended pipe (116 mm diameter), through which air was moved by a small 12 V fan powered by a rechargeable battery. This pipe was attached to openings in the rear wall of the ark such that air was drawn from, and returned to the ark at an overall rate of 19.4 m3hÿ1
, which corresponded to a recirculation rate of 4.7 air changes per hour within the ark.
After weaning, exposure to the arti®cial odourant was continued in the holding room using commercially available odourizers, which comprised a wick saturated with the liquid odourant. The odourant from nine odourizers was discharged into the inlet pipe of the room's ventilation system. Each holding room was ventilated by an inlet fan at a rate of 17717.6 m3hÿ1
(meanS.E.), i.e.7.50.7 air changes per hour. The inlet air ¯ow was counter-balanced by an extractor fan situated in the opposite wall. All air supplied to the holding rooms was sourced from outside the building.
As a control, an identical but empty odourizer was placed in the ventilation unit servicing the second farrowing ark. The two ventilated arks were separated by six unventilated ones, sited so that the entrances faced away from the prevailing wind, in order to minimise cross-contamination. Plastic curtains over the entrances restricted further the entry of ambient air.
2.4. Ammonia presentation and exposure
Anhydrous ammonia gas (BOC, UK) was released at a pressure of 1 atm into the air supply of one of the two holding rooms. The ¯ow of ammonia was regulated by a rotameter. Partial obstructions in the inlet ventilation pipe ensured that the ammonia and air were mixed thoroughly. The ammonia concentration was measured approximately 0.4 m above the ¯oor using an infrared gas analyser (BruÈel and Kjñr, Denmark) and was 43.30.1 ppm
(meanS.E.). Four of the eight weaned litters were exposed continuously to ammonia for between 14 and 19 days before testing (termed chronic exposure). These litters were chosen using a Latin square design.
ammonia at some stage during the experiment. The ammonia concentration was measured in two arks during the farrowing period using DraÈger diffusion tubes (DraÈger Sicherheit-stechnik, Germany) and was 1.90.5 ppm (meanS.E.), which is much lower than the concentration to which housed pigs are exposed routinely.
2.5. Novel environment (modified open field) and test procedure
Fig. 1 shows the layout of a sound-proof room that was used as the novel environment or open ®eld. It was located in the same building as the holding rooms; its ceiling sloped from 2.25 m at the entrance to 1.8 m at the opposite wall, and it was empty apart from a water bowl situated next to the entrance. The walls were lined with black StokbordTMto a height of approximately 1.25 m and then clad in chip-board; the concrete ¯oor was painted white. This room was lit by a single ¯uorescent tube at an average light intensity of 70.7 lx. An overhead video camera was used to record the pig's behaviour onto video tape. The room was ventilated by an inlet fan at a rate of approximately 176 m3hÿ1, i.e.7.3 air changes per hour. All air was sourced from outside the building and released at the centre of the room at approximately 0.5 m above the ¯oor. The inlet air ¯ow was counter-balanced by an extractor fan situated in the wall opposite the entrance and 1.5 m above the ¯oor. When Fig. 1. Floor plan of the open field (novel environment) showing the position of the start box, the inlet
ventilation pipe (*), the area of the open field that was classified as the centre (chequered pattern) and the 55
square grid superimposed over the video recording to assess a pig's activity.
required, the arti®cial odourant and/or ammonia were added to this air supply (termed acute exposure) using equipment identical to that used in the holding rooms. The ammonia concentration, measured 0.4 m above the ¯oor of the room using an infrared gas analyser, was 45.30.9 ppm (meanS.E.).
The test pig was placed in a wooden, unventilated, unlit start box measuring 1.2 m0.7 m1.0 m. It was allowed 15 min acclimatisation before a guillotine door into the open ®eld was raised. It is conceivable that exposure to the start box induced different levels of fear and/or separation distress in different pigs but this procedure was standardised across all test conditions. The test pig was not forced to leave the start box and the door was lowered once it had entered the open ®eld. After 1 h, the door to the start box was raised and the pig was allowed to leave the open ®eld. Four pigs, two from each litter, were tested on any 1 day in one of two sessions.
The following behaviours were recorded by continuous sampling using the tip of the snout to de®ne the pig's location; the latency to enter the open ®eld (also determined for the whole body), the latency to approach the centre of the open ®eld (Fig. 1), the number of times the pig entered the ®eld's centre, the percentage of time spent in the centre and the number of times the pig moved between the cells of an imaginary 55 square grid.
2.6. Statistical analysis
The data were log10 transformed before analysis of variance (ANOVA, GENSTAT5 Lawes Agricultural Trust, 1996) was used to compare the effects of arti®cial odourant and ammonia exposure on the pigs' behaviour. The litter from which the pigs were weaned, the pair of litters, the test day and test session were used as the blocking factors or random part of the analysis. Exposure to the arti®cial odourant in Phases 1, 2 and 3 and exposure to ammonia in Phases 2 and 3 were used as the treatment factors or ®xed part of the analysis. Table 2 summarises the hierarchical structure of the analysis of variance. Pair-wise comparisons of means were made subsequently using a two-tailed Student'st-test at a signi®cance level of 0.05 with 44 d.f.
3. Results and discussion
Novelty often elicits fear (Jones, 1996, 1997) and many animals are reluctant to approach novel stimuli, to venture into novel environments or to move much when they are in them. These effects are apparent in a wide range of species and open-®eld tests have been extensively used to measure fear not only in small animals, like chickens or rodents (Gray, 1987; Jones, 1987, 1989) that are potentially at risk from predators, but also in large ones, such as cattle (Kilgour, 1975; Boissy, 1995), sheep (Lachaux et al., 1983) and pigs (Mormede et al., 1984; Taylor and Friend, 1986). Simplistically, in the absence of any familiar cues, we might expect a frightened pig to be reluctant to enter a novel environment (like the open ®eld) and, having entered it, to shelter against the walls and to move infrequently across the open space (Jones, 1989, 1996).
Table 2
Hierarchical structure of the analysis of variance used to assess the data (given in Tables 3 and 4) and the probability that the result at each level is different from chance for each behavioural response
Phases 1 and 2 odourantPhase 2 ammonia 1 0.774 0.899 0.909 0.308 0.263 0.964
Residual 2
Litter pairlitter stratum
Phases 1 and 2 odourant 1 0.208 0.543 0.559 0.056 0.162 0.014
Phase 2 ammonia 1 0.296 0.222 0.432 0.070 0.137 0.451
Residual 2
Litter pairlittertest daytest session stratum
Phase 3 odourant 1 0.968 0.894 0.525 0.735 0.749 0.100
Phase 3 ammonia 1 0.698 0.941 0.612 0.727 0.606 0.737
Phases 1 and 2 odourantPhase 3 odourant 1 0.601 0.718 0.962 0.786 0.520 0.656
Phase 2 ammoniaPhase 3 odourant 1 0.139 0.082 0.162 0.794 0.597 0.612
Phases 1 and 2 odourantPhase 3 ammonia 1 0.185 0.206 0.711 0.057 0.028 0.346
Phase 2 ammoniaPhase 3 ammonia 1 0.765 0.580 0.047 0.008 0.044 0.007
Phase 3 odourantPhase 3 ammonia 1 0.604 0.739 0.815 0.225 0.534 0.269
Phases 1 and 2 odourantPhase 2 ammoniaPhase 3 odourant 1 0.632 0.758 0.112 0.736 0.366 0.753
Phases 1 and 2 odourantPhase 2 ammoniaPhase 3 ammonia 1 0.565 0.812 0.656 0.244 0.182 0.506
Phases 1 and 2 odourantPhase 3 odourantPhase 3 ammonia 1 0.665 0.396 0.557 0.876 0.899 0.534
Phase 2 ammoniaPhase 3 odourantPhase 3 ammonia 1 0.856 0.700 0.225 0.774 0.549 0.616
Phases 1 and 2 odourantPhase 2 ammoniaPhase 3
odourantPhase 3 ammonia
1 0.095 0.334 0.237 0.594 0.715 0.916
residual d.f. used to test the majority of treatment effects. Table 3 summarises the behaviour of the pigs in the open ®eld according to whether or not they had been exposed to the arti®cial odourant in Phases 1 and 2 before test. Table 4 summarises the behaviour of the pigs following chronic exposure to 43 ppm ammonia before test. The means in Tables 3 and 4 are those for four (of the six) two factor interactions shown in Table 2. None of the three factor or the four factor interactions were signi®cant (p>0.05, Table 2) and these data are therefore not presented.
In general, the behaviour of the pigs that had received previous exposure to the arti®cial odourant in the farrowing ark and holding room was not affected by its presence or absence in the novel test room (Table 2,p>0.05 for all relevant factors). There was, however, one exception. Pigs which had been exposed to the arti®cial odourant in Phases 1 and 2 were generally more active (as indicated by the number of moves across a 55 grid) than unexposed controls (Table 2, Phase 1 and 2 odourant,p0.014). This probably re¯ected a Type 1 error because the other behavioural responses of these pigs showed no consistent trend. Contrary to ®ndings that laboratory rodents and chickens were attracted to, and reassured by, familiar odours presented in otherwise unfamiliar surroundings (Devor and Schneider, 1974; Carr et al., 1979; Galef, 1981; Jones and Gentle, 1985; Burne and Rogers, 1995; Porter and Picard, 1998; Jones and Carmichael, 1999), the responses of the pigs used here were unaffected by the presence or absence in the novel environment of an arti®cial odourant (Ambi-Pur) with which they had been reared from birth. There are at least three possible explanation for this ®nding. Firstly, the animals may have developed a transient hyposmia for the components of this odourant and thereby ceased to sense it. Secondly, although no disgust or avoidance responses to Ambi-Pur were seen in the home environ-ment, it is conceivable that either some of its components or its overall olfactory gestalt had aversive properties that counterbalanced the development of attraction through familiar-isation. Thirdly, and perhaps more likely, the pigs may not have formed a suf®ciently strong association between this olfactory cue and the familiar home environment, possibly because the compounds did not occur naturally and have biological relevance.
Table 3
Odourant Ammonia1 1.28 1.55 2.29 1.31 0.11 b 2.87
No ammonia1 1.12 1.39 2.22 1.45 0.36 a 2.90
No odourant Ammonia1 1.21 1.46 2.17 1.34 0.19 ab 2.78
No ammonia1 1.49 1.63 2.16 1.14 0.03 b 2.72
Odourant Odourant2 1.16 1.46 2.23 1.41 0.28 2.91
No odourant2 1.24 1.49 2.28 1.36 0.19 2.85
No odourant Odourant2 1.40 1.57 2.14 1.24 0.09 2.80
No odourant2 1.31 1.51 2.19 1.24 0.12 2.70
Standard error of difference between means
0.181 0.161 0.148 0.092 0.109 0.053
fMean (log
10) open field behaviour of pigs exposed/not exposed to an artificial odourant before test, listed according to the different environmental conditions in the
open field (absence/presence of odourant and/or45 ppm ammonia). Letters `a' and `b' indicate differences significant atP<0.05 between means within a two factor
interaction.
1Means pooled across odourant exposure in Phase 3.
2Means pooled across ammonia exposure in Phase 3.
Table 4
Ammonia exposuref
Phase 2 (experience) Phase 3 (open field
environment)
Ammonia Odourant2 1.10 1.32 2.07 1.40 0.28 2.88
No odourant2 1.34 1.53 2.22 1.35 0.20 2.77
No ammonia Odourant2 1.46 1.72 2.30 1.25 0.09 2.84
No odourant2 1.21 1.47 2.22 1.24 0.11 2.78
Standard error of difference between means
0.181 0.161 0.148 0.092 0.109 0.053
fMean (log
10) open field behaviour of pigs exposed/not exposed to43 ppm ammonia before test, listed according to the different environmental conditions in the
open field (absence/presence of odourant and/or45 ppm ammonia). Letters `a' and `b' indicate differences significant atP<0.05 between means within a two factor
interaction.
1Means pooled across odourant exposure in Phase 3.
2Means pooled across ammonia exposure in Phase 3.
choice between an atmosphere containing approximately 0, 10, 20 or 40 ppm ammonia pigs reared in fresh air preferred the lowest concentration of ammonia and minimised the time spent in the highest concentration (Jones et al., 1996).
Ammonia is a natural constituent of the atmosphere for the pig kept outdoors and indoors; it is obviously found at urination sites in ®elds where its concentration will vary over time. Chronic exposure to ammonia at approximately 43 ppm for 14±19 days before the open ®eld test did not impair the pig's ability to perceive a similar concentration of ammonia in the open ®eld, even though it is known to interfere with the olfactory perception of reproductive pheromones (Malayer et al., 1988) and sensory acuity for n-butanol (J.B., Jones, unpublished). Details of odourant perception in the pig are poorly understood but ammonia stimulates the trigeminal as well as the olfactory sense in man and it may be that the former sense is less likely to be harmed by ammonia exposure than the latter. We had also planned to test whether chronic exposure to ammonia in Phase 2 and acute exposure to ammonia in Phase 3, affected the pigs' responses to a familiar odourant in the home environment. However, these effects could not be tested because the presence of the arti®cial familiar odourant did not affect open-®eld behaviour.
Acknowledgements
The authors would like to thank Mr. I.R. Meeks for his technical assistance and Mr. J.D. Rickatson for allowing unrestricted access to his commercial breeding unit. This study was funded by the Biotechnology and Biological Sciences Research Council Ð Agri-Food Committee, as part of project 204/A07440 to study olfactory perception in pigs and the in¯uence of aerial pollutants.
References
Boissy, A., 1995. Fear and fearfulness in animals. Q. Rev. Biol. 70, 165±191.
Burne, T.H.J., Rogers, L.J., 1995. Odors volatiles and the approach-avoidance behaviour of the domestic chick (Gallus gallus domesticus). Int. J. Comp. Psychol. 8, 99±114.
Campbell, R.G., 1976. A note of the use of a feed flavour to stimulate the feed intake of weaner pigs. Anim. Prod. 23, 417±419.
Carr, W.J., Marasco, E., Landauer, M.R., 1979. Responses by rat pups to their own nest versus a strange conspecific nest. Physiol. Behav. 23, 1149±1151.
Devor, M., Schneider, G.E., 1974. Attraction to home-cage odor in hamster pups: specificity and changes with age. Behav. Biol. 10, 211±221.
Galef Jr., B.G., 1981. Development of olfactory control of feeding-site selection in rat pups. J. Comp. Physiol. Psychol. 95, 615±622.
Gray, J.A., 1987. The Psychology of Fear and Stress, 2nd Edition. Cambridge University Press, Cambridge, UK. Hafez, E.S.E., Signoret, J.P., 1969. The behaviour of swine. In: Hafez, S.E. (Ed.), The Behaviour of Domestic
Animals. BallieÁre, Tindall and Cassell, London, pp. 341±30.
Jeppesen, L.E., 1982. Teat-order in groups of piglets reared on an artificial sow. 2. Maintenance of teat-order with some evidence for the use of odour cues. Appl. Anim. Ethol. 8, 347±355.
Jones, R.B., 1987. The assessment of fear in the domestic fowl. In: Zayan, R., Duncan, I.J.H. (Eds.), Cognitive Aspects of Social Behaviour in the Domestic Fowl. Elsevier, Amsterdam, pp. 40±81.
Jones, R.B., 1989. Avian open-field research and related effects of environmental novelty: an annotated bibliography 1960±1988. Psychol. Rec. 39, 397±420.
Jones, R.B., 1996. Fear and adaptability in poultry: insights, implications and imperatives. W. Poul. Sci. J. 52, 131±174.
Jones, R.B., 1997. Fear and distress. In: Appleby, M.C., Hughes, B.O. (Eds.), Animal Welfare. CAB International, Wallingford, UK, pp. 75±87.
Jones, R.B., Carmichael, N.L., 1999. Domestic chicks are attracted to a familiar odourant in a novel test situation: a brief report. Appl. Anim. Behav. Sci. 61, 351±356.
Jones, R.B., Faure, J.M., 1982. Domestic chicks prefer soiled substrate in an otherwise novel environment. IRCS Med. Sci. 10, 847.
Jones, R.B., Gentle, M.J., 1985. Olfaction and behavioural modification in the domestic chick (Gallus
domesticus). Physiol. Behav. 34, 917±924.
Jones, J.B., Burgess, L.R., Webster, A.J.F., Wathes, C.M., 1996. Behavioural responses of pigs to atmospheric ammonia in a chronic choice test. Anim. Sci. 63, 437±445.
Kilgour, R., 1975. The open-field test as an assessment of the temperament of dairy cows. Anim. Behav. 23, 615±624.
Lachaux, M., Bouissou, M.F., Berges, J.C., Orgeur, P., 1983. Etude du comportements en open-field de beliers Ile-de-France soumis a differentes conditions d'elevage. Biol. Behav. 3, 257±269.
Lawes Agricultural Trust, 1996. GENSTAT5 Release 3.2. Rothamsted Experimental Station, UK.
Malayer, J.R., Brandt, K.E., Green, M.L., Kelly, D.T., Sutton, A.L., Diekman, M.A., 1988. Influence of manure gases on the onset of puberty in replacement gilts. Anim. Prod. 46, 277±282.
Meese, G.B., Conner, D.J., Baldwin, B.A., 1975. Ability of the pig to distinguish between conspecific urine samples using olfaction. Physiol. Behav. 15, 121±125.
McLaughlin, C.L., Baile, C.A., Buckholtz, L.L., Freeman, S.K., 1983. Preferred flavours and performance of weanling pigs. J. Anim. Sci. 56, 1287±1293.
Mormede, P., Dantzer, R., Bluthe, R.M., Caritez, J.C., 1984. Differences in adaptive abilities of three breeds of Chinese pigs. Behavioural and neuroendocrine studies. Genet. Sel. Evol. 16, 85±102.
Morrow-Tesch, J., McGlone, J.J., 1990. Sources of maternal odours and the development of odour preferences in baby pigs. J. Anim. Sci. 68, 3563±3571.
Parfet, K.A.R., Gonyou, H.W., 1991. Attraction of newborn piglets to auditory, visual, olfactory and tactile stimuli. J. Anim. Sci. 69, 125±133.
Porter, R.H., Picard, M., 1998. Effects of early odor exposure in domestic chicks. Reprod. Nutr. Dev. 38, 441± 448.
Russell, P.A., 1979. Fear-evoking stimuli. In: Sluckin, W. (Ed.), Fear in Animals and Men. Van Nostrand Reinhold, London. pp. 22±54.
