www.elsevier.comrlocaterapplanim
The number of farm mates influences social and
maintenance behaviours of Japanese Black cows in
a communal pasture
Ken-ichi Takeda
),1, Shusuke Sato, Kazuo Sugawara
Graduate School of Agricultural Science, Tohoku UniÕersity, Kawatabi, Narugo, Miyagi 989-6711 JapanAccepted 30 November 1999
Abstract
The objective of the present study was to investigate optimal group size in cattle in communal
Ž .
pastures where cattle from various farms are turned out: 1 to ascertain if cows derived from each
Ž .
farm form an affiliative group and 2 to investigate the effect of the number of farm mates on social and maintenance behaviours when farm mates form an affiliative group. Farm mates were defined as cows from the same farm. A total of 27 cows having zero, one, two to four, or sixteen farm mates were selected as focal animals in two communal pastures. Each focal animal was followed by an individual observer for 8 h from sunrise. The nearest neighbour and the distance to the nearest neighbour were recorded at 10-min intervals, and maintenance behaviours at 1-min intervals; social behaviours and participants were recorded continuously. Most focal animals frequently choose cows from the same farm as their nearest neighbours allogroomed with these
Ž .
farm mates significantly more P-0.001 . Two focal animals did not form affiliative groups with farm mates. The mean distance to the nearest neighbours, whether farm mates or non-farm mates from focal animals having one and two to four farm mates was shorter than cows turned out with
Ž .
no farm mate 1.4 vs. 1.9 times as long as a cow’s body length, P-0.05 . Cows having two to
Ž . Ž
four farm mates performed 7.8 srhrcow, Ps0.06 and received allogrooming more 9.8
. Ž .
srhrcow and escaped less during agonistic encounters 0.1 frequenciesrhrcow, P-0.01 than ones having other numbers of farm mates. The mean duration of grazing behaviour per bout
Ž .
tended to increase with the number of farm mates 5.2 to 10.3 min, P-0.05 . The mean duration
)Corresponding author. Tel.:q81-229-84-7311; fax:q81-229-84-6490.
Ž .
E-mail address: k1take@bios.tohoku.ac.jp K. Takeda . 1
Current address: Faculty of Agriculture, Shinshu University, 8304 Minami-minowa, Kani-uria, Nagano 399-4598, Japan.
0168-1591r00r$ - see front matterq2000 Elsevier Science B.V. All rights reserved.
Ž .
Ž .
of recumbency behaviour of cows having two to four farm mates 44.0 minrbout was longer than the ones having other numbers of farm mates. Social and maintenance behaviours of cows having 16 farm mates were like those having zero or one farm mates, which suggested that the social bond among them was weak. It is concluded that a group of cows having two to four farm mates, that is, a group size of three to five cows, may be optimal for a stable life in a communal pasture.
q2000 Elsevier Science B.V. All rights reserved.
Keywords: Cattle; Optimal group size; Allogrooming; Social bond; Grazing behaviour
1. Introduction
Communal pastures into which cattle from various farms are turned out are very popular in Japan. Stress can be a problem on communal pastures. Being with unfamiliars
Ž .
increased the heart rate and changed behaviours in deer Pollard et al., 1993 , sheep ŽBaldock and Sibly, 1990 , and cattle Kenny and Tarrant, 1987 , which was interpreted. Ž . to mean that social mixing may be stressful. In addition, there is a big problem in communal pastures where the body weights of some animals are reduced during an early
Ž .
grazing period. Susaki and Hirota 1976 reported that there was a slight positive correlation between the body weight gains of calves during the grazing period in a communal pasture and the number of their farm mates reared together on each farm during the non-grazing period. They suggested that the social experience of indoor-group rearing in winter reduced social stress in a communal pasture in a grazing season.
Ž .
Warnick et al. 1977 reported that calves group-reared in their early days quickly learned competitive feeding after weaning.
Social stress in a communal pasture may be reduced by not only early social experiences, but also by having familiar farm mates, which are defined in this paper as cows from the same farm. Familiar conspecifics can lower the stress responses of
Ž .
subjects Gunnar et al., 1980; Coe et al., 1982; Boissy and Le Neindre, 1990 . Arnone
Ž .
and Dantzer 1980 reported that the presence of a congener with which social bonds had been established played a protective role against behavioural and physiological consequences of frustration. It is well-known that calves reared together in the same pen
Ž
or on the same farm form a subgroup in outdoor rearing Ewbank, 1967; Broom and .
Leaver, 1978; Hayasaka, 1987 . Members in a subgroup allogroom each other, and graze
Ž .
and rest together Bouissou and Hovels, 1976; Sato et al., 1993 . It is recognized that
¨
Ž . Ž
allogrooming has a calming effect on cattle Sato and Tarumizu, 1993 , horses Feh and
. Ž .
De Mazieres, 1993 , and monkeys Schino et al., 1988; Boccia et al., 1989 . Thus, the presence of farm mates may have a buffering effect against multiple stressors. In addition, the suppressing effect of farm mates on multiple stresses may depend on the strength of their social bonds and the number of their farm mates. Cows having no or only a few farm mates may be more stressed than ones with many farm mates. Individual vigilance behaviour decreases and individual eating behaviour increases with
Ž .
In the present study, we tried to investigate if farm mates affiliate with each other and if the number of farm mates influences social and maintenance behaviours of cows in communal pastures.
2. Materials and methods
2.1. ObserÕation sites
Ž .
Two herds of cows were observed in the Onikoube communal pasture 388N, 1408E
Ž .
and the Nanashigure communal pasture 408N, 1418E at 300–500 m and 550 m altitudes, respectively, in northeastern Japan. Both pastures consisted of artificial
Ž .
grasslands and woodlands. Orchard grass Dactylis glomerata L. and broad-leaved
Ž .
trees maple Acer spp. and oak Quercus spp. dominated in artificial grasslands and woodlands, respectively. The Onikoube pasture and the Nanashigure pasture were divided into 10 paddocks of 59.3 ha and eight paddocks of 69.6 ha, respectively. Cows in the Onikoube pasture were observed in a paddock consisting of artificial grasslands of 6.8 ha and woodlands of 0.2 ha, and in the Nanashigure pasture in a paddock consisting of artificial grasslands of 10.3 ha and woodlands of 0.6 ha. There were two watering places and some mineral blocks were supplied for cows in each paddock. The paddock in the Onikoube pasture was flat while the paddock in the Nanashigure pasture was hilly.
2.2. Animals
Ž .
A total of 80 Japanese Black heifers and cows 1–17-years-old means7.5-years-old from 38 farms were turned out into the Onikoube pasture at the same time without their
Ž .
calves from 11 May to 12 October in 1995 Table 1 . Fifteen cows 3–5-years-old were selected as focal animals from the Onikoube herd: five cows from five farms which turned out one cow each, five cows from three farms which turned out two cows each,
Table 1
The number of farms which turned out cows and the number of cows from each farm into the Onikoube pasture and the Nanashigure pasture during observation periods
Number of cows derived from each farm Total
1 2 3 4 5 6 7 17
The Onikoube pasture
No. of farms 19 8 4 4 2 – 1 – 38
No. of cows 19 16 12 16 10 – 7 – 80
The Nanashigure pasture
No. of farms 9 14 4 8 4 – – 1 40
four cows from three farms which turned out four cows each, and one cow from one farm which turned out five cows. One hundred and eighteen Japanese Black heifers and
Ž .
cows 1–20-years-old means6.6-years-old from 40 farms were turned out into the Nanashigure pasture at the same time without their calves from 12 May to 6 November
Ž .
in 1997 Table 1 . Twelve cows 3–8-years-old were selected as focal animals from the Nanashigure herd: four cows from four farms which turned out two cows each, two cows from one farm which turned out three cows, one cow from one farm which turned out four cows, one cow from one farm which turned out five cows, and four cows from one farm which turned out 17 cows. Only three yearling heifers had not experienced any grazing and the remainder had experienced previous grazing in the Onikoube pasture. About one-third of the cows, including all focal animals, had experienced previous grazing in the Nanashigure pasture.
There were no differences among farms in rearing methods during the non-grazing period. Cows were kept in single pens or tied in stalls. They were sometimes released into the arena as a group in each farm in good weather during the non-grazing period.
2.3. ObserÕation
The two herds were observed for 5 days from 26 to 30 July 1995 in the Onikoube pasture and for 4 days from 12 to 15 July 1997 in the Nanashigure pasture. The weather varied between fair, cloudy, and foggy and average temperatures at 8:00 were 27.58C in the Onikoube pasture and 22.98C in the Nanashigure pasture during the observation periods.
The three observers simultaneously followed three focal cows on each observation Ž
day, respectively 15 focal cows in 1995s3 observers=5 days; 12 focal cows in
. Ž
1997s3 observers=4 days . Each focal cow was observed for 8 h from sunrise 4:30 .
in 1995 and 4:00 in 1997 on one of the observation days. The ethogram used for recording is shown in Table 2.
Ž .
The nearest neighbour’s number, her posture standing, recumbency, and walking and the distance to the nearest neighbour as multiples of the cow’s body length were recorded at 10-min intervals in principle. These items, however, were not checked repeatedly if the nearest neighbour relationship with the focal animal did not change:
Ž
animals, the spatial relationship, and posture. Posture standing, recumbency, and
Table 2
Head throw, head butt, fight, escape Agonistic
. Ž .
walking and maintenance behaviour grazing, resting, and rumination of the focal cow were recorded at 1-min intervals. Social behaviour was observed continuously. The
Ž
number of occurrences of agonistic head throw, head butt, fight and escape from
. Ž .
aggression and affiliative behaviours rubbing and physical contact , except allogroom-ing, and the time spent allogrooming were recorded. The performers and receivers of social behaviour were identified.
2.4. Statistical analysis
Whether focal animals chose farm mates as the nearest neighbours and allogroomed with them was analysed using the binominal test.
The data on the frequenciesrcowrhour of agonistic and affiliative behaviours except Ž .
allogrooming and the total duration srcowrhour of allogrooming were a skewed distribution. This data was subjected to square root transformation after an addition of
Ž .
0.5 and then analysed statistically Martin and Bateson, 1986 . The mean distance to the nearest neighbour, the frequency of social behaviour and the mean duration of mainte-nance behaviour were analysed using a one-way analysis of variance to examine the effect of the number of farm mates on these items. Data on focal animals, which were turned out with two to four farm mates, were pooled because the sample size was small.
Ž .
If the effect was significant, Fisher’s protected least significance difference test PLSD was used among cows having no farm mate, one, two to four, and 16 farm mates.
As there was no difference in behavioural data between the two herds, the data from them were pooled. The data on one focal animal in oestrus, which was turned out with 16 farm mates, were excluded from the analysis.
3. Results
3.1. Social relationship of focal cows with farm mates
The nearest neighbours to each focal animal were observed from 23 to 35 occasions Žmeans29.3 in the Onikoube pasture and from 27 to 44 occasions mean. Ž s37.5 in. the Nanashigure pasture. Focal animals, except No. 4 and No. 85, frequently chose farm
Ž .
mates as their nearest neighbours P-0.001, Table 3 . No. 4 and No. 85 never chose farm mates as their nearest neighbours.
Ž .
Focal animals except six cows No. 18, No. 4, No. 7, No. 85, No. 149, No. 66
Ž .
allogroomed more with farm mates than with other cows P-0.001, Table 3 . Four
Ž .
cows No. 18, No. 7, No. 149, No. 66 performed affiliative behaviours to farm mates, such as rubbing and physical contact, although they did not allogroom. No. 4 and No. 85
Ž .
were not affiliative no allogrooming, rubbing, or physical contact with farm mates. Data from No. 4 and No. 85 were excluded from the remaining analysis because they did not form affiliative relationships with farm mates.
Table 3
Total number of farm mates which were chosen as the nearest neighbours by the focal animal, and the total
Ž
time spent allogrooming of focal animals with farm mates parenthesized figures are the total number of the
.
nearest neighbours and the total time spent allogrooming during an 8-h observation
Farm mates Cow no. No. of choice Time spent allogrooming
Žtimes. Žseconds.
UUU UUU
Ž . Ž .
The Onikoube pasture 1 6 8 25 60 60
UUU UUU
The Nanashigure pasture 1 2 37 44 37 37
UUU to four farm mates were significantly shorter than cows having no farm mates 1.4 vs.
.
1.9 times as long as a cow’s body length, P-0.05 . The mean distance of cows having
Ž .
16 farm mates 1.6 times as long as a cow’s body length was between the distances of cows having no farm mate and of cows having one or two to four farm mates. The mean distances of No. 4 and No. 85 to the nearest neighbours were 1.9 and 1.2 times as long as a cow’s body length, respectively.
3.2. Effect of the number of farm mates on social and maintenance behaÕiours
Table 5 shows the frequencies of social behaviour of cows having various numbers of
Ž .
farm mates. Cows having two to four farm mates performed Ps0.06 and received
Table 4
Mean distances from focal cows having various numbers of farm mates to their nearest neighbours
The number of farm mates Significance
0 1 2–4 16
Table 5
Time spent allogrooming and frequencies of affiliative and agonistic behaviours of focal cows having various numbers of farm mates
Behaviours The number of farm mates Significance
0 1 2–4 16
Means followed by different letters b,c differ significantly in each line.
a
Affiliative behaviour does not include allogrooming.
allogrooming more than any other focal animals. Cows having 16 farm mates rarely performed and received allogrooming. There were no significant differences in affilia-tive behaviours except for allogrooming among cows having various numbers of farm mates. There was also no effect of the numbers of farm mates on aggressive behaviour.
Ž .
Cows having two to four farm mates 0.1 timesrcowrh escaped less during agonistic
Ž .
encounters than cows having no farm mates 0.7 timesrcowrh, P-0.01 . Cows having Ž
16 farm mates escaped as often as cows having one farm mate 0.4 vs. 0.3 .
occurrencercowrh . No. 4 escaped 0.6 timesrh and No. 85 did not exhibit escape. Affiliative and aggressive behaviours were never observed in the two focal animals
Ž .
excluded from analysis No. 4 and No. 85 .
The number of farm mates affected the mean duration of maintenance behaviours ŽTable 6 . Mean duration of grazing behaviour per bout tended to increase with the. number of farm mates up to two to four. Mean durations of grazing behaviour in cows having two to four and 16 farm mates were longer than those of cows having zero or one farm mate. Mean duration of recumbency behaviour in cows having two to four
Ž . Ž .
farm mates 44 min was longer than and in cows having no farm mate 30 min was less than the other groups. These differences were not significant. Mean duration of grazing and recumbency behaviours of No. 4 and No. 85 were 4.6 and 8.4 min, and 11.4 and 44.2 min, respectively. Total duration of walking was not different among cows
Table 6
Ž .
Mean duration min of grazing and recumbency behaviours per bout in focal cows having various numbers of farm mates
Behaviours The number of farm mates Significance
0 1 2–4 16
a a b b
Grazing behaviour 5.2"0.4 5.3"0.7 9.4"1.2 10.3"4.0 P-0.05 Recumbency behaviour 29.5"5.7 32.7"3.7 44.0"7.8 35.3"7.1 NS Figures are mean"S.E.M.
Ž .
Ž
having various numbers of farm mates 24.3 min with no farm mate, 28.0 min with one . farm mate, 25.1 min with two to four farm mates and 21.3 min with 16 farm mates . No. 4 and No. 85 walked for 24 min and 71 min during 8 h of observation.
4. Discussion
4.1. Focal cows formed affiliation with farm mates
Most focal cows closely related with farm mates in the communal pastures. It is well-known that animals reared indoors as a group or on the same farm live gregariously
Ž . Ž . Ž
outdoors; including heifers Hayasaka, 1987 , calves Ewbank, 1967 and sheep Win-.
field et al., 1981 . Formation of a subgroup depends more on social history than on
Ž .
kinship Ewbank, 1967; Murphey, 1990 .
Spatial distribution of individuals is one aspect of their social bond. Some observa-tions concerning the mean distance to the nearest neighbour in cattle have been reported.
Ž . Ž . Ž
Most animals keep 1 m Sato et al., 1987 , 1–2 m Kondo, 1987 and 2–3 m Fraser and
. Ž .
Broom, 1997 during lying, and 4–10 m during grazing Fraser and Broom, 1997 . In this study, mean distances during grazing and resting recumbency from focal animals to
Ž .
their nearest neighbours was about 1.5 times as long as a cow’s body length about 2 m . When calves lay close to one another, they sniffed the nearest neighbour possibly in
Ž . Ž .
order to identify her as familiar Sato et al., 1987 . Broom and Leaver 1978 have found that group-reared animals licked, sniffed, or nuzzled with members of their own group more than with members of another group. It may be that a close relationship in which affiliative behaviours are performed represents a stronger social bond than a proximate relationship. In our study, focal animals allogroomed farm mates and were allogroomed
Ž .
by them for a longer time than non-farm mates P-0.0001 . These results suggest that farm mates not only gather with each other spatially, but are also close together behaviourally in communal pastures. These results also show that farm mates form an affiliative group in a herd.
Two focal animals, however, No. 4 and No. 85, did not form such relationships. No. 4 was very nervous on the farm, and its mean distance to its nearest neighbour was the same as that of cows having no farm mates. No. 85 might have been in oestrus because its walking time was 71 min, which was the same duration as when the cow was in
Ž .
oestrus 78 min in this study. Temperament and physical conditions of cows may influence affiliative relationships with farm mates.
4.2. Effect of the number of farm mates on social relationships
Mean distances to the nearest neighbours in cows having one or two to four farm mates were shorter than cows having no farm mates. Therefore, it is considered that social bonds affected the spatial pattern of cows in communal pastures.
Ž .
distance to the nearest neighbour of cows having 16 farm mates was an intermediate length between those in cows having no farm mate and in cows having one or two to four farm mates. In addition, allogrooming of cows having 16 farm mates was hardly observed. Therefore, intra-cow social bonds in cows having 16 farm mates might be
Ž .
loose. In a study by Sato et al. 1998 , the mean of subgroup sizes of Japanese Shorthorn cows in each month of the study ranged between 5.8 and 10.4 in the communal forest range. It may be considered that cows do not form large affiliative groups, and a large number of cows will divide into small affiliative groups. Further research is needed on social relationships in large groups of cows because we observed only one group of 17 farm mates in the present study.
Cows having two to four farm mates allogroomed them and were allogroomed by farm mates significantly more than those having the other numbers of farm mates. Sato
Ž .
and Tarumizu 1993 reported that heart rates of cows decreased significantly while receiving allogrooming. In addition, positive correlation between the time spent
receiv-Ž . Ž
ing allogrooming and milk yields of cows Wood, 1977 or weight gains of calves Sato, .
1984 have been reported. Moreover, not only time spent receiving but also performing
Ž .
allogrooming correlated significantly to milk yield in cows Sato et al., 1991b . These results suggest that the receiving of allogrooming influences welfare directly to improve it. The performance of allogrooming may also influence welfare indirectly through the formation and maintenance of social bonds. Thus, it is considered that cows having two to four farm mates are more psychologically stable than those having other numbers of farm mates.
Ž .
Sato et al. 1991a reported that the novelty of a new cow faded out 45–60 days after an unfamiliar cow was introduced into an established cow herd. Since cows were observed 1 month after turning out into a communal pasture in this study, there would still be agonism among cows. Cows having two to four farm mates escaped less during agonistic encounters than ones having other numbers of farm mates. Broom and Leaver Ž1978 reported that in competitive situations, the mean proportion of wins in agonistic. encounters of group-reared calves was higher than in individually reared calves.
Ž .
Bouissou and Hovels 1976 found that animals kept in a group for some time were
¨
more likely to have similar ranks when their group was combined with others. These results suggest that cows reared in a group have a relatively high and similar social rank. Therefore, cows having no farm mate might be subordinate, and cows having two to four farm mates might be dominant in the herd.4.3. Effect of the number of farm mates on maintenance behaÕiour
The mean duration of grazing behaviour tended to increase with the number of farm mates. In wild animals, the two most important adaptive values of living in a group may be the increase of protection from predators and the likelihood of finding and getting
Ž .
food Krebs and Davies, 1995 . The same effect of group size on feeding behaviour was
Ž .
farm mates escaped less. It was considered that social affiliative and social dominance may assure security in the lives of cows, which occupy high qualitative and quantitative grazing sites and comfortable resting sites.
Ž .
Sato et al. 1993 reported that the duration of allogrooming when housed was
Ž .
positively correlated with proximity during grazing. Bouissou and Hovels 1976 found
¨
that calves reared together from birth associated with each other during feeding and resting. The strength of the social bonds may influence maintenance behaviour.Ž . Ž . Ž .
Wood 1977 , Sato 1984 , and Sato et al. 1991b found that the amount of time
Ž .
spent allogrooming positively influenced animal productivity. Kondo et al. 1983 reported that the calves in groups of two had significantly less gain in body weight than calves in groups of six under crowding conditions. Consequently, it is considered that the number of farm mates influence animal productivity. In our study, it seems that cows having two to four farm mates are more socially stable than any other focal animals.
Ž .
Penning et al. 1993 found that sheep in smaller groups spent less time grazing than sheep in a larger group. They suggested that the minimum unit of replication for grazing experiments should be groups of four or five, and should not be less than three. Also,
Ž .
Kusunose et al. 1986 reported that the mean duration of grazing bouts prolonged linearly as the group size increased up to four horses. The Midwest Plan Service
Ž .
Committee 1985 also recommended that the number of weaned calves per pen be five to six.
Japanese Black cows derived from the same farm formed an affiliative group within their herd in communal pastures. The social behaviour tended to be more affiliative and dominant with, and the mean duration of maintenance behaviour tended to increase with the number of farm mates up to four cows. When the number of cows derived from the same farm was 17, social bonds among them were scarce. It is concluded that cows having two to four farm mates may be the most socially stable and the most secure in their lives in a communal pasture. Consequently, the optimal group size may be three to five for maximising animal productivity and the animal welfare level.
Acknowledgements
We thank Mr. Kiichi Seki in the Onikoube communal pasture, Mr. Makoto Watanabe in the Nishine Town office for offering us every convenience, and we particularly appreciate both farmers’ cooperation. Thanks are due to Ms. Tomoko Yokoyama, Ms. Asako Furuya, Mr. Tomoharu Inoue, and Dr. Shin-ichiro Ogura on helping our observations.
References
Arnone, M., Dantzer, R., 1980. Does frustration induce aggression in pigs? Appl. Anim. Ethol. 6, 351–362. Baldock, N.M., Sibly, R.M., 1990. Effects of handling and transportation on the heart rate and behaviour of
sheep. Appl. Anim. Behav. Sci. 28, 15–39.
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.
Bouissou, M.F., Hovels, J.H., 1976. Effet d’un contact precoce sur quelques aspects du comportement social¨ ´
des Bovins domestiques. Biol. du Comp. 1, 17–36.
Broom, D.M., Leaver, J.D., 1978. Effects of group-rearing or partial isolation on later social behaviour of calves. Anim. Behav. 26, 1255–1263.
Coe, C.L., Franklin, D., Smith, E.R., Levine, S., 1982. Hormonal responses accompanying fear and agitation in the squirrel monkey. Physiol. Behav. 29, 1051–1057.
Ewbank, R., 1967. Behavior of twin cattle. J. Dairy Sci. 50, 1510–1512.
Feh, C., De Mazieres, J., 1993. Grooming at a preferred site reduces heart rate in horses. Anim. Behav. 46, 1191–1194.
Fraser, A.F., Broom, D.M., 1997. In: Farm Animal Behaviour and Welfare. 3rd edn. CAB International, Wallingford, p. 131.
Gunnar, M.R., Gonzalez, C.A., Levine, S., 1980. The role of pairs in modifying behavioral distress and pituitary–adrenal response to a novel environment in year-old rhesus monkeys. Physiol. Behav. 25, 795–798.
Hayasaka, K., 1987. Seasonal changes in herding behaviour patterns and inter-animal relationship of Holstein heifers in a common pasture. Jpn. J. Zootech. Sci. 58, 687–697.
Kenny, F.J., Tarrant, P.V., 1987. The reaction of young bulls to short-haul road transport. Appl. Anim. Behav. Sci. 17, 209–227.
Kondo, S., 1987. Behavioral adaptation in group of cattle. Memoirs of the Faculty of Agriculture, Hokkaido
Ž .
University 15, 192–233, in Japanese, with English abstract .
Kondo, S., Ono, K., Nishino, S., 1983. Differences in spatial and social behaviour between calves in groups of two different sizes. J. Coll. Dairy. 10, 63–71.
Krebs, J.R., Davies, N.D., 1995. In: An Introduction to Behavioural Ecology. 3rd edn. Blackwell, Oxford, pp. 120–146.
Kusunose, R., Hatakeyama, H., Ichikawa, F., Kubo, K., Kiguchi, A., Asai, Y., Ito, K., 1986. Behavioral studies on yearling horses in field environments: 2. Effects of the group size on the behavior of horses.
Ž .
Bull. Equine Res. Inst. 23, 1–6, in Japanese, with English abstract .
Lagory, K.E., 1986. Habitat, group sizes, and the behaviour of white-tailed deer. Behaviour 98, 168–179. Martin, P., Bateson, P., 1986. In: Measuring Behaviour. Cambridge Univ. Press, London, pp. 135–136. Midwest Plan Service Committee, 1985. In: Dairy Housing and Equipment Handbook. 4th edn. Iowa State
Ž .
University, Ames, p. 8, in Japanese .
Murphey, R.M., 1990. Social aggregations in cattle: I. Segregation by breed in free-ranging herds. Behavior Genetics 20, 341–354.
Penning, P.D., Parsons, A.J., Newman, R.J., Orr, R.J., Harvey, A., 1993. The effects of group size on grazing time in sheep. Appl. Anim. Behav. Sci. 37, 101–109.
Pollard, J.C., Littlejohn, R.P., Suttie, J.M., 1993. Effects of isolation and mixing of social groups on heart rate and behaviour of red deer stags. Appl. Anim. Behav. Sci. 38, 311–322.
Risenhoover, K.L., Bailey, J.A., 1985. Relationships between group size, feeding time, and agonistic behavior of mountain goats. Can. J. Zool. 63, 2501–2506.
Sato, S., 1984. Social licking pattern and its relationships to social dominance and live weight gain in weaned calves. Appl. Anim. Behav. Sci. 12, 25–32.
Sato, S., Sassa, H., Sonoda, T., 1991a. The change in social behaviour of newly introduced heifers with
Ž .
original group cows. Anim. Sci. Technol. Jpn. 62, 83–92.
Ž .
Sato, S., Tarumizu, K., 1993. Heart rates before, during and after allo-grooming in cattle Bos taurus . J.
Ethol. 11, 149–150.
Sato, S., Hatae, K., Tarumizu, K., 1993. The influence of social factors on allogrooming in cows. Appl. Anim. Behav. Sci. 38, 235–244.
Sato, S., Sonoda, T., Tarumizu, K., 1991b. Social, behavioural and physiological functions of allo-grooming in
Ž .
cattle. In: Appleby, M.C., Horrell, R.I., Petherick, J.C., Rutter, S.M. Eds. , Applied Animal Behaviour: Past, Present and Future. University Federation for Animal Welfare, Herts, pp. 77–78.
Sato, S., Wood-Gush, D.G.M., Wetherill, G., 1987. Observations on creche behaviour in suckler calves.´
Sato, S., Yasue, K., Mitamura, T., 1998. Subgrouping and homerange of cattle in agroforest without fencing in
Ž .
Northern Japan. In: Veissier, I., Boissy, A. Eds. , Proc. of 32nd. Congr. of ISAE, Clermont-Ferrand, France. p. 212.
Schino, G., Scucchi, S., Maestripieri, D., Turillazzi, P.G., 1988. Allogrooming as a tension-reduction mechanism: a behavioral approach. Am. J. Primatol. 16, 43–50.
Susaki, H., Hirota, H., 1976. Effect of the sociability on the liveweight gains in raising Holstein heifers on
Ž .
pasture. J. Jpn. Grassl. Sci. 22, 301–306, in Japanese, with English abstract .
Warnick, V.D., Arave, C.W., Mickelsen, C.H., 1977. Effects of group, individual, and isolated rearing of calves on weight gain and behavior. J. Dairy Sci. 60, 947–953.
Winfield, C.G., Syme, G.J., Pearson, A.J., 1981. Effect of familiarity with each other and breed on the spatial behaviour of sheep in an open field. Appl. Anim. Ethol. 7, 67–75.
