Grazing behaviour of sheep in a situation of conflict ... - CiteSeerX

Behavioural Processes 49 (2000) 131 – 138 www.elsevier.com/locate/behavproc

Grazing behaviour of sheep in a situation of conflict between feeding and social motivations B. Dumont *, A. Boissy INRA Theix, Unite´ de Recherches sur les Herbi6ores, Centre de Clermont-Fd/Theix, 63122 Saint-Gene`s-Champanelle, France Received 16 November 1999; received in revised form 28 February 2000; accepted 1 March 2000

Abstract We investigated how food preferences and social bonds interact to determine the choice of grazing location in sheep. Ewes of INRA 401 breed were grazed in plots in which taller areas, i.e. preferred feeding sites, were left to grow at 15 or 50 m from a socially attractive site, i.e. familiar ewes placed in a public pen at one end of the plot. Eight experimental ewes were tested either alone or in groups with one, three or six accompanying animals chosen amongst 20 other familiar ewes. We used a Latin square design, in which the eight treatments (two distances × four group sizes) were balanced in 8 measurement days. We recorded, in 20-min tests, the behaviour of the experimental ewes by focal sampling, and the location of each animal in the groups by scan sampling. Foraging location, dietary choices and vigilance behaviour of ewes were affected by both the distance between the group of public peers and the preferred feeding site, and the size of their own group. Our results suggest that a sheep will move whether alone or with a few peers to a preferred feeding site located close to the core of its social group. In a small sub-group, its frequency of vigilance behaviour increases, probably to maintain social contact with the rest of the group. Conversely, a sheep will not leave its group to reach a preferred feeding site located further away unless it is followed by several other peers. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Grazing behaviour; Group size; Social attraction; Conflict; Sheep

1. Introduction Domestic animals construct responses to their environment that depend on experience and the integration of environmental features. Food preferences and social interactions are usually the main factors that influence diet and habitat selec* Corresponding author. Tel.: +33-473-624607; fax: + 33473-624118. E-mail address: [email protected] (B. Dumont)

tion in domestic grazing herbivores (e.g. Bailey et al., 1996; Roguet et al., 1998; Dumont and Boissy, 1999). Under range conditions, there are complex interactions between social grouping tendencies and foraging decisions. Early social experiences (Howery et al., 1998) and social cohesiveness (Lazo, 1992) affect the distribution patterns of cattle, together with fluctuating water and vegetation availability. Similarly, social bonds interact with forage availability to determine the grazing distribution of sheep (Arnold and

0376-6357/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 6 - 6 3 5 7 ( 0 0 ) 0 0 0 8 2 - 6

132

B. Dumont, A. Boissy / Beha6ioural Processes 49 (2000) 131–138

Dudzinski, 1978). Sheep recognise their group members, and under free ranging conditions avoid animals from the other groups (Hunter and Milner, 1963; Favre, 1979; Lawrence and WoodGush, 1988a). Animals of the same flock share the use of a common home-range in a similar way year to year, but with seasonal patterns according to vegetation availability and climatic variations. Despite this similar use of a common home-range, yearling sub-groups within a flock do not systematically follow the movements of older ewes to preferred foraging sites in both domestic (Lawrence and Wood-Gush, 1988b) and wild bighorn sheep (O6is canadensis, Geist, 1971). More recently, an experimental set-up was developed to test the interaction between food preferences and social grouping tendencies in sheep. Scott et al. (1995) conditioned lambs to prefer either wheat or milo, which were then placed at opposite ends of 1-ha pastures, and observed the animal activity and feeding location. They showed that the strength of social bonds determined the choice of feeding location. In groups of lambs that had been reared together, social attraction was stronger than the attraction for the preferred food. Conversely, in newly constituted groups, lambs left the rest of the group and expressed their own food preferences. The result of the conflict between social and feeding motivations was also shown to depend on animals’ knowledge of the environment. In new pastures lambs foraged together on the same food, whereas in familiar pastures they typically consumed different foods, according to whether they preferred wheat or milo (Scott et al., 1996). In the present study, we placed sheep in a conflict situation by exposing them simultaneously to a preferred feeding site and to a socially attractive site, i.e. familiar ewes placed in a public pen. These sites were located away from each other, and we observed the location and diet selection of ewes according to the distance between the two sites. As sheep are gregarious animals, it is likely that the expression of the motivation conflict will be affected by the size of the group in which an animal forages. Group size has already been shown to affect grazing behaviour in sheep. In groups smaller than four animals, sheep reduced their grazing time (South-

cott et al., 1962; Penning et al., 1993) and did more aimless walking (Southcott et al., 1962) than sheep tested in larger groups. Therefore, we tested the ewes in groups smaller, equal or bigger than four animals. In order to measure the result of the motivation conflict under more natural pasturelike conditions than in previous studies, the preferred feeding site was, in our experiment, a tall vegetative sward instead of a concentrate food.

2. Materials and methods

2.1. Animals The experiment ran from 10 May to 4 June 1999 at the INRA farm of Theix in the upland area of central France. We used 28 adult nonpregnant INRA 401 ewes from the same flock. INRA 401 is a French breed obtained by crossing Romanov ewes with Berrichon du Cher rams. The ewes weighed 63 kg (s.d. 8 kg) when the measurements began. Eight of them were used as experimental ewes, the remaining 20 ewes accompanying them in the tests.

2.2. Experimental set-up The experiment was carried out in a 1.57 ha cocksfoot (Dactylis glomerata) pasture. Seventeen 70× 10 m experimental plots were set up in this pasture. In each plot, the sward was kept mown to about 3.5 cm in height, except for a 6 × 10 m area that was left to grow to 10 cm for the tests (Fig. 1). Because diet preferences of herbivores at pasture increase with the height of vegetative swards (Illius and Gordon, 1990), this taller sward served as an attractive feeding site. It was established either close (15 m) or far (50 m) from a socially attractive site, namely a group of seven familiar ewes placed in a 3× 10 m public pen, surrounded by 1-m-high wooden panels with wide latticed openings. During the test period, the nontested animals were placed in a waiting pen enclosed with blind wooden panels and located behind the public pen. The observers were located on a 4-m-high platform so as not to influence the animals’ movements. Large numbered wooden


panels were placed every 5 m along the edge of the experimental plots to locate the animals.

2.3. Procedures Before measurements began, the ewes grazed for 3 days in some of the experimental plots to get accustomed to the test conditions. The observers were trained during this preliminary period to eliminate any observer bias. Then there were two or three measurement days per week during which each experimental ewe was tested for 20 min. In order to prevent a significant depletion of the taller sward during the tests, four experimental plots were used each day and were then left to grow again for at least one week before grazed again. Outside the measurement days, the 28 ewes grazed together on vegetative natural pastures. On measurement days, the animals were allowed to graze until 08:00 h in the morning and were then kept in the waiting pen with no access to food. Four tests were carried out between 10:00 and 12:00 h. All the animals were then allowed to graze until 13:00 h and were kept in the waiting pen again until 15:00 h. The four remaining tests were carried out between 15:00 and 17:00 h, all the animals then being turned out to pasture. We studied how a ewe’s activity and location varied according to the distance between the two attractive sites (15 or 50 m between the taller

Fig. 1. The experimental plots. The preferred feeding sites (taller sward) are established at 15 or 50 m from the socially attractive one (pen with the public ewes).

133

sward and the public pen), and the size of its own group (one, two, four or seven animals). The experimental ewes were always tested in the same order. Consequently, individual variations in behaviour were confounded with the effect of time of the day and fasting duration, but all observed changes in behaviour were due solely to the treatments. The 20 accompanying ewes were used to make up the different group sizes. Each of them was only used once a day, and was randomly chosen before the tests. The seven experimental ewes that were not being tested were placed in the public pen to form the socially attractive site. We used a Latin square design, in which we balanced the eight treatments (two distances× four group sizes) for residual effects in eight measurement days.

2.4. Measurements During the tests an observer using a microphone recorded the activity (grazing on the short sward, grazing on the taller sward, walking, idling, vigilance or social interactions) of the experimental ewe, by focal sampling. An animal was considered to be grazing when it was biting or chewing grass or when it was walking with its muzzle close to the sward. Vigilance postures were recorded when an animal scanned all around with its head up and ears erected. Social interactions could be agonistic or non-agonistic, and directed towards accompanying or public animals. Focal sampling was preferred over scan sampling because behavioural measurements can be affected by the sampling pattern especially for rare and short-duration events (Zinner et al., 1997), such as vigilance and social interactions in this experiment. A second observer recorded the location of the experimental ewe and its accompanying animals every minute. Observed locations were reported on a map with a grid, each cell representing 2.5×2.5 m on the experimental plot. These data later allowed calculation of the proportion of the test time ewes were located on the taller site, and the average distance between the tested and public animals.

134


Fig. 2. Foraging location of ewes according to group size and distance between the two attractants. Within each distance, data with different superscripts differ statistically (PB 0.05).

test to detect differences in behaviour between treatments; (ii) measurements= b0 + b1ai + b2bj + b3cij + b4dk + b5el + oijkl, where ai, bj, dk and el represent differences due to distance between attractive sites, group size, animal and day, and cij represents the interaction between distance and group size. We used non-parametric statistics for the other parameters because no transformation could stabilise variances. To assess whether the behaviour of a ewe varied according to the treatments, we used Friedman’s two-way nonparametric ANOVA (SAS, 1989) on the proportion of test time ewes were located on the taller site, grazing on the taller sward, walking, idling, in vigilance or in social interaction. We used the Wilcoxon test for paired data to detect differences in behaviour between treatments. As there were no differences in location data between experimental ewes and the whole group, only the data from the experimental ewes are presented here.

3. Results

Fig. 3. Dietary choices of ewes according to group size and distance between the two attractants. Within each distance, data with different superscripts differ statistically (PB 0.05).

2.5. Statistical analyses We used the GLM procedure of SAS (1989) to analyse data on the distance from the public animals and on the proportion of test time spent grazing on the short sward. Two models were used: (i) measurements=a0 +a1ai +a2bj +a3ck + oijk, where ai, bj and ck represent differences due to treatment, animal and day. We used the Duncan

The average distance of the experimental ewe from the public animals varied according to the treatments (F7,63 = 5.68, PB 0.001). Interaction between group size and distance separating the two attractive sites, i.e. the public pen and the taller sward, was significant (F3,63 = 3.12, PB 0.05, Fig. 2). Group size did not affect the average distance of the experimental ewe from the public animals when the distance between the two attractive sites was short. Conversely, the larger the group, the further the experimental ewe moved from the public animals when the distance between the two attractive sites was greater. According to the treatments, there were significant differences in the time ewes were located on the taller sward (F7,63 = 5.81, PB 0.001) and grazed this taller sward (F7,63 = 5.37, PB 0.001, Fig. 3). Data closely matched those for the average distance of experimental ewes from the public pen. When the distance between the two attractive sites was short, ewes spent on average 0.66 of the test duration on the taller sward and grazed it for 0.63 of the time, regardless of group size. When the distance between the two sites was long, the


larger the group, the more time the ewes spent on the taller sward and the more they grazed it. Except for a group size of four animals, ewes spent significantly more time on the taller sward and grazed it longer when it was 15 m rather than 50 m away from the public pen (P B0.05). Whatever the group size, ewes spent significantly more time grazing on the short sward, when the taller sward was 50 m from the public animals (P B 0.05). Interaction between group size and distance separating the two attractive sites was significant (F3,63 = 4.41, PB0.01). With a 50-m-long distance, the proportion of time spent grazing on the short sward decreased from 0.84 to 0.52 with increasing group size (P B0.05). A distance of 15 m was without effect (on average 0.21 of test duration). The proportion of test time ewes spent in vigilance postures varied according to the treatments (F7,63 = 4.52, PB0.001). It decreased significantly with increasing group size when the distance between the two attractive sites was short (P B 0.05, Fig. 4), a similar tendency being observed when this distance was long (PB 0.10). The proportion of time spent walking was also significantly affected by the treatments (F7,63 =4.43, P B0.001). The larger the group, the more time the ewes

Fig. 4. Vigilance behaviour of ewes according to group size and distance between the two attractants. Within each distance, data with different superscripts differ statistically (PB 0.05).

135

spent walking (from 0.02 to 0.07 of test duration, PB 0.01), whichever the distance between the two attractive sites. Finally, neither the proportion of time spent in social interactions (on average 0.01 of test duration, F7,63 = 0.34, NS), nor the proportion of time spent idling (on average 0.07 of test duration, F7,63 = 0.45, NS) were affected by the treatments.

4. Discussion Foraging location, dietary choices and vigilance behaviour of ewes were affected by the distance between public peers and a preferred feeding site, as well as by the size of their own group. According to the distance between the two attractive sites, i.e. the public peers and a tall vegetative sward, group size had a different effect on the behavioural response of the ewes. Our results suggest that a sheep will move whether alone or with a few peers to a preferred feeding site located close to the core of its social group. In a small sub-group, its frequency of vigilance behaviour increases, probably to maintain social contact with the rest of the group. Conversely, a sheep will not leave its group to reach a preferred feeding site located further away unless it is followed by several other peers. The distance between the two attractive sites strongly influenced the behavioural response of the animals, as is generally the case in situations of conflict. For example, in domestic hens, preventing birds from feeding together reduced time spent feeding and increased frustration behaviour (Mills and Faure, 1989). Also, increasing the distance between a warm shelter and food offered at the cold end of a test arena decreased the hoarding behaviour of rats (Cabanac and Swiergiel, 1989). Tits (Parus caeruleus, Todd and Cowie, 1990) and squirrels (Sciurus niger, Brown and Morgan, 1995) have been shown to trade off foraging benefit and predation risk according to the distance of food from tree cover. Similarly, due to increased predation risk, foraging at a great distance from the edge of a forest caused moose (Alces alces gigas) to forage in larger

136


groups and browse willow less selectively (Molvar and Bowyer, 1994). In sheep, the ratio of edible biomass to the distance to a feeding patch dictated an animal’s decision to walk for a preferred forage when a poor-quality alternative was freely available (Dumont et al., 1998). Scott et al. (1995) also showed that lambs always expressed their learned preferences for wheat or milo when the two grains were placed at opposite ends of a small pasture (0.25 ha), whereas social cohesiveness sometimes overrode dietary preferences in a larger pasture (1 ha). Unfortunately, the effects of the distance between attractants were confounded with the strength of dietary preferences in this last experiment, whereas the results of our experiment prove that it affects the choice of grazing location by sheep. Sheep are typically gregarious animals (Arnold, 1985), and their social motivation remains strong when grazing (Penning et al., 1993). Here, the ewes were reluctant to separate clearly from their peers placed in the public pen to exploit a preferred feeding site located 50 m away, especially when they grazed alone or in a group of two. In this situation, a conflict arose between the motivations to maintain social contact and to express a food preference, the ewes shifting between the two attractants as the size of their group varied. Genetic factors can partially influence social attractiveness (Launay et al., 1991; Hahn and Wright, 1998), and have been shown to interact with environmental factors in the determination of grazing distribution in free-ranging sheep (Dwyer and Lawrence, 1999). Due to the genetic origin of the INRA 401 breed, the expression of this motivation conflict is probably particularly strong since both Romanov and Berrichon du Cher crossbred animals have been shown to be very sensitive to social isolation (Boissy et al., 1996; Le Neindre et al., 1998). In this experiment, the tested and the public ewes had been bred together since a young age, which also may have aggravated the motivation conflict. There is indeed evidence that the influence of social peers on individual behaviour depends on the quality of the relationship between the animals. It is now well established that social

contacts during development influence social behaviour in adulthood. For example, heifers in groups formed at birth expressed closer associations than in groups formed at the age of 6 months (Bouissou and Andrieu, 1978). These preferential associations between individuals can affect their feeding behaviour: penned calves reared together from birth associate later during feeding (Bouissou and Ho¨vels, 1976). In addition, ewes reared together from a young age stayed with their peers, even when the location of their preferred grain in a plot differed (Scott et al., 1995). As early social experiences have a marked influence on individual behaviour, it would be interesting to test under these more natural conditions whether non-familiar animals would graze the taller sward irrespective of its distance from the public pen. The expression of the motivation conflict could have been further enhanced by other factors that would have led animals to express their dietary preferences. When offered a range of choice between vegetative patches differing in sward heights, sheep showed less sensitivity to height differences than goats and cattle (Illius and Gordon, 1990). This was explained by differences in grazing style and in body size, respectively. When group size was increased, the strength of the shift towards the expression of food preference could thus have been stronger with other herbivore species. In addition, fasting sheep for at least 16 h has been shown to affect their grazing behaviour and diet preferences (Newman et al., 1994; Dumont et al., 1995). Here, increasing the feeding motivation of experimental ewes by longer food deprivation before the tests could have caused them to graze more on the taller sward, which allows a higher intake rate. This would have affected their adjustment to the proposed conflict. When the preferred site was located near the public peers, the ewes grazed it independently of their group size, but their vigilance behaviour varied. Vigilance postures were on average six times more frequent when the ewes grazed alone than in a group of seven animals. The presence of peers provides social support, which has positive consequences for the individual exposed to a


stressful situation (Boissy and Le Neindre, 1990). Group size and vigilance behaviour usually show a negative correlation in wild animals (e.g. Hoogland, 1979; Underwood, 1982; Hunter and Skinner, 1998), though not always (Molvar and Bowyer, 1994). When running predation risk, wild bighorn sheep in smaller groups grazed less efficiently than those in larger groups (Berger, 1991). From an adaptive perspective, vigilance provides a benefit by increasing the chances of detecting a potential predator, thus enhancing animals’ survival (Elgar, 1989; Boissy, 1998). Underwood (1982) suggested that even when the risk of being preyed upon is low, vigilance in ungulates appears to be affected by the possibility of predation. Although our sheep never ran any predation risk, they may have perceived themselves to be under greater threat alone than in a group. This confirms that domestic herbivores can still express ancestral behaviour correlated with the level of perceived danger of predation (Price, 1984; Penning et al., 1993). In conclusion, when sheep are in a situation of conflict between feeding and social motivations, group size and distance between the two attractants interact to determine the choice of grazing location. Our experimental approach provides a useful tool for further testing of the effects of social bonds, feeding motivation and animal species or breed on individual behaviour when competition arises between motivation to maintain social contact and expression of food preferences. Other studies have begun to investigate the tradeoffs between forage availability and micro-climatic variations in grazing sheep (Duncan et al., 1997). There is now a need to quantify the relative importance of these different features under contrasting environmental conditions, in order to model the formation of subgroups in grazing herbivores, and their use of heterogeneous pastures.

Acknowledgements We are grateful to I. Calvo and P. Dacheux for help with measurements, and to M. Auguergouz, L. Lavelle, B. Mallet and H. Tournadre for setting up the experimental plots.

137

References Arnold, G.W., 1985. Associations and social behaviour. In: Fraser, A.F. (Ed.), Ethology of Farm Animals. Elsevier, Amsterdam, pp. 233 – 248. Arnold, G.W., Dudzinski, M.L., 1978. Ethology of FreeRanging Domestic Animals. Elsevier, Amsterdam 198 pp. Bailey, D.W., Gross, J.E., Laca, E.A., Rittenhouse, L.R., Coughenour, M.B., Swift, D.M., Sims, P.L., 1996. Mechanisms that result in large herbivore grazing distribution patterns. J. Range Manage. 49, 386 – 400. Berger, J., 1991. Pregnancy incentives, predation constraints and habitat shifts: experimental and field evidence for wild bighorn sheep. Anim. Behav. 41, 61 – 77. Boissy, A., 1998. Fear and fearfulness in determining behavior. In: Grandin, T. (Ed.), Genetics and the Behavior of Domestic Animals. Academic Press, San Diego, pp. 67 – 110. Boissy, A., Le Neindre, P., 1990. Social influences on the reactivity of heifers: implications for learning abilities in operant conditioning. Appl. Anim. Behav. Sci. 25, 149 – 165. Boissy, A., Le Neindre, P., Orgeur, P., Bouix, J., 1996. Genetic variability and psychobiological reactivity in lambs reared under open-range management. In: Duncan, I.J.H., Widowski, T.M., Haley, D.B. (Eds.), Proceedings of the 30th International Congress of the ISAE. Guelph, Canada, p. 59. Bouissou, M.F., Andrieu, S., 1978. Etablissement des relations pre´fe´rentielles chez les bovins domestiques. Behaviour 64, 148 – 157. Bouissou, M.F., Ho¨vels, J.H., 1976. Effet des conditions d’e´levage sur le comportement des ge´nisses dans une situation de compe´tition alimentaire. Ann. Zootech. 25, 213 – 219. Brown, J.S., Morgan, R.A., 1995. Effects of foraging behavior and spatial scale on diet selectivity: a test with fox squirrels. Oikos 74, 122 – 136. Cabanac, M., Swiergiel, A.H., 1989. Rats eating and hoarding as a function of body weight and cost of foraging. Am. J. Physiol. 257, R952 – R957. Dumont, B., Boissy, A., 1999. Relations sociales et comportement alimentaire au paˆturage. INRA Prod. Anim. 12, 3 – 10. Dumont, B., Petit, M., D’hour, P., 1995. Choice of sheep and cattle between vegetative and reproductive cocksfoot patches. Appl. Anim. Behav. Sci. 43, 1 – 15. Dumont, B., Dutronc, A., Petit, M., 1998. How readily will sheep walk for a preferred forage? J. Anim. Sci. 76, 965 – 971. Duncan, A.J., Mayes, R.W., Young, S.A., Wilson, P., Lamb, C.S., 1997. Preliminary study on the effect of micro-climatic variation on behavioural choices made by grazing sheep. In: Forbes, J.M., Lawrence, T.L.J., Rodway, R.C., Varley, M.A. (Eds.), Animal choices. Occasional Publication no. 20, BSAS, pp. 94 – 95. Dwyer, C.M., Lawrence, A.B., 1999. Ewe – ewe and ewe – lamb behaviour in a hill and a lowland breed of sheep: a study using embryo transfer. Appl. Anim. Behav. Sci. 61, 319 – 334.

138

B. Dumont, A. Boissy / Beha6ioural Processes 49 (2000) 131–138 Le Neindre, P., Murphy, P.M., Boissy, A., Purvis, I.W., Lindsay, D., Orgeur, P., Bouix, J., Bibe´, B., 1998. Genetics of maternal ability in cattle and sheep. In: Proceedings of the 6th World Congress on Genetics Applied to Livestock Production. Armidale, Australia, pp. 23 – 30. Mills, A.D., Faure, J.M., 1989. Social attraction and the feeding behaviour of domestic hens. Behav. Proc. 18, 71 – 85. Molvar, E.M., Bowyer, R.T., 1994. Costs and benefits of group living in a recently social ungulate: the Alaskan moose. J. Mammal 75, 621 – 630. Newman, J.A., Penning, P.D., Parsons, A.J., Harvey, A., Orr, R.J., 1994. Fasting affects intake behaviour and diet preference of grazing sheep. Anim. Behav. 47, 185 – 193. Penning, P.D., Parsons, A.J., Newman, J.A., Orr, R.J., Harvey, A., 1993. The effects of group size on grazing time in sheep. Appl. Anim. Behav. Sci. 37, 101 – 109. Price, E.O., 1984. Behavioral aspects of animal domestication. Quat. Rev. Biol. 59, 1 – 32. Roguet, C., Dumont, B., Prache, S., 1998. Selection and use of feeding sites and feeding stations by herbivores: a review. Ann. Zootech. 47, 225 – 244. SAS, 1989. SAS/STAT® User’s Guide (Release 6.11). SAS Inst. Inc., Cary, NC. Scott, C.B., Provenza, F.D., Banner, R.E., 1995. Dietary habits and social interactions affect choice of feeding location by sheep. Appl. Anim. Behav. Sci. 45, 225 – 237. Scott, C.B., Banner, R.E., Provenza, F.D., 1996. Observations of sheep foraging in familiar and unfamiliar environments: familiarity with the environment influences diet selection. Appl. Anim. Behav. Sci. 49, 165 – 171. Southcott, W.H., Roe, R., Turner, H.N., 1962. Grazing management of native pastures in the New England region of the New South Wales. Aust. J. Agri. Res. 13, 880 – 893. Todd, I.A., Cowie, R.J., 1990. Measuring the risk of predation in an energy currency: field experiments with foraging blue tits, Parus caeruleus. Anim. Behav. 40, 112 – 117. Underwood, R., 1982. Vigilance behaviour in grazing African antelopes. Behaviour 79, 81 – 107. Zinner, D., Hindahl, J., Schwibbe, M., 1997. Effects of temporal sampling patterns of all-occurrence recordings in behavioural studies: many short sampling periods are better than a few long ones. Ethology 103, 236 – 246.

Elgar, M.A., 1989. Predator vigilance and group size in mammals and birds: a critical review of the empirical evidence. Biol. Rev. 64, 13 – 33. Favre, Y., 1979. Etude de l’organisation sociale et de l’utilisation de l’espace par des ovins (O6is aries L.) en liberte´. The`se Univ. Aix-Marseille II, 200 pp + annexes. Geist, V., 1971. Mountain Sheep. A Study in Behaviour and Evolution. University of Chicago Press, Chicago, 383 pp. Hahn, M.E., Wright, J.C., 1998. The influence of genes on social behavior of dogs. In: Grandin, T. (Ed.), Genetics and the Behavior of Domestic Animals. Academic Press, San Diego, pp. 299 – 318. Hoogland, J.L., 1979. The effect of colony size on individual alertness of prairie dogs (Sciuridae: Cynomys spp.). Anim. Behav. 27, 394 – 407. Howery, L.D., Provenza, F.D., Banner, R.E., Scott, C.B., 1998. Social and environmental factors influence cattle distribution on rangeland. Appl. Anim. Behav. Sci. 55, 231 – 244. Hunter, L.T.B., Skinner, J.D., 1998. Vigilance behaviour in African ungulates: the role of predation pressure. Behaviour 135, 195 – 211. Hunter, R.F., Milner, C., 1963. The behaviour of individual, related and groups of south country Cheviot hill sheep. Anim. Behav. 11, 507–513. Illius, A.W., Gordon, I.J., 1990. Constraints on diet selection and foraging behaviour in mammalian herbivores. In: Hughes, R.N. (Ed.), Behavioural Mechanisms of Food Selection. NATO ASI Series, vol. G20, pp. 369–393. Launay, F., Mills, A.D., Faure, J.M., 1991. Social motivation in Japanese quail Coturnix coturnix japonica chicks selected for high or low levels of treadmill behaviour. Behav. Proc. 24, 95 – 110. Lawrence, A.B., Wood-Gush, D.G.M., 1988a. Home range behaviour and social organization of Scottish Blackface sheep. J. Appl. Ecol. 25, 25–40. Lawrence, A.B., Wood-Gush, D.G.M., 1988b. Influence of social behaviour on utilization of supplemental feedblocks by Scottish hill sheep. Anim. Prod. 46, 203–212. Lazo, A., 1992. Facteurs de´terminants du comportement gre´gaire de bovins retourne´s a` l’e´tat sauvage. Rev. Ecol. 47, 51– 66.

.

.