Selection of social traits in juvenile Japanese quail affects adults behaviour

Applied Animal Behaviour Science 112 (2008) 174 186 www.elsevier.com/locate/applanim Selection of social traits in juvenile Japanese quail affects adults behaviour Laureline Formanek *, Cécilia Houdelier, Sophie Lumineau, Aline Bertin, Guénaël Cabanès, Marie-Annick Richard-Yris UMR CNRS 6552, Ethologie Evolution Ecologie, Université de Rennes 1, Bât.25, Campus de Beaulieu, F-35042 Rennes, Cedex, France Accepted 4 July 2007 Available online 8 August 2007 Abstract The inability of animals to cope with their social environment in husbandry systems could be a major source of problems for both animal welfare management and economy. In this context, previous studies selected lines of domestic Japanese quail diverging for their level of social reinstatement, hypothesizing that quail with high levels of social motivation would adapt better to crowded social situations than quail with low levels of social motivation. However, these selections were based only on social motivation evaluated in young birds. As behavioural traits can vary with age, this study analysed the persistence of selected social traits into adulthood. With this in mind, we used several behavioural tests to estimate social motivation of adult females from selected lines for high social reinstatement (HSR, n = 24) and for low social reinstatement (LSR, n = 24). The tests were: open-field tests, emergence tests, separation tests, runway tests and confrontation tests. As birds behaviour in these tests can be influenced by the emotive characteristics of the subjects, we measured the emotional reactivity of quail not only during these tests but also in specific emotive tests: tonic immobility tests. Our results showed that the social motivation characteristics of birds persist in adulthood. Adult HSR females still presented a higher social motivation than adult LRS females: their response to social isolation was stronger and their motivation to reach conspecifics and to stay close to them was higher. However, the behavioural expression of social motivation appeared to be modified during development. Whereas vocal behaviour reflected the social motivation of chicks, in isolated contexts, this characteristic was not observed in adult quail. # 2007 Elsevier B.V. All rights reserved. Keywords: Poultry welfare; Japanese quail; Divergent selection; Social motivation; Influence of age * Corresponding author. Tel.: +33 223236855; fax: +33 223236927. E-mail address: laureline.formanek@univ-rennes1.fr (L. Formanek). 0168-1591/$ see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.applanim.2007.07.004

1. Introduction L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 175 The development of avian husbandry systems has allowed a rapid increase of animal production efficiency. At the same time, rearing conditions have changed greatly in terms of animal numbers and maintenance conditions (battery farming or ground rearing). These modifications are associated with observations of behavioural disorders indicating that the animals were not totally adapted to their rearing environment. This non-adaptation has negative consequences on production and has also been stressed by animal protection societies and certainly by consumers, more and more attached to animal welfare. Thus, improvement of the living conditions of farm animals appears necessary for productivity reasons as well as for ethic reasons. One approach to improve welfare is to increase the capacities of adaptation of individuals to husbandry systems by modifying some of the animals characteristics (Mills et al., 1994). In this context, two main categories of behaviour have been particularly studied because they imply many problems during rearing: fear reactivity and social behaviour (Mills et al., 1994). Although the adaptive value of fear is known in the wild, its expression in husbandry systems can impair welfare. Similarly, current rearing systems impose strong constraints on social organization (Hughes, 1975; Wood-gush, 1971; Jones and Harvey, 1987; Jones and Williams, 1992; Keeling, 1987; Vallortigara et al., 1990; Vallortigara, 1992): high densities, unisex groups or groups of males and females with a same age but not the same social motivation, frequent modifications of group composition that imply that birds must establish a new social order each time. The negative consequences of a prolonged expression of fear and social tension are important and influence the physiology of birds (immune system) (Gross and Siegel, 1983), their behaviours (panic reactions, agonistic behaviours, reproduction) (Jones, 1989) and their general state (growth is delay, bone or egg shell quality decreases) (Jones and Hugues, 1986; Satterlee and Roberts, 1990). So, to reduce these negative effects in poultry and to increase animal welfare, it appeared necessary to select quiet, not very emotive animals that tolerate their conspecifics. Selection on fear and social behaviour has been particularly developed in Japanese quail as these two characteristics have strong genetic components (Mills and Faure, 1990, 1991). This species is an interesting model especially for investigations of relationships between animal welfare and specifics social constraints. Indeed, contrary to most domestic poultry species that are known to derive from social species, the social organisation of wild quail is still unknown. Although young are known to live in family groups, the social interactions of adults are poorly documented. During the breeding season, males are territorial and are aggressive towards other males (Mills et al., 1997). Moreover, adult females can be very aggressive towards males when they are not sexually developed (Mills et al., 1997). Therefore, adults, during one phase of their biological cycle at least, do not seem to tolerate their conspecifics. So, in current intense poultry farming systems, quail are probably subjected to important social stress. In this context, selection of animals presenting higher social tolerance provides interesting insights for productivity as well as animal welfare. Genetic selection on social behaviour in Japanese quail has been based on the social reinstatement tendency of young birds measured by the distance that an individual can cover to reach conspecifics. Two divergent genetic lines have been obtained: a high social reinstatement tendency (HSR) line (these quail cover long distances to reach conspecifics) and a low social reinstatement tendency (LSR) line (these quail cover only short distances to join conspecifics) (Mills and Faure, 1990). This social characteristic appears associated with other social behaviours that, when they are considered all together, reflect the social motivation of the birds

176 L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 (social motivation is defined as the tendency of an isolated animal to establish contact with, to move towards and to remain close to a group of conspecifics (Le Masne, 1996). So, in social isolation, HSR quail emit more distress calls, utter their first distress call sooner after being isolated, and their locomotor activity is higher than do LSR quail (Launay et al., 1991; Mills et al., 1993). Thus, social separation appears to be more stressful for HSR than for LSR birds. When presented a social reinforcement (conspecifics) in runway or treadmill tests, HSR quail reach it faster (Launay et al., 1991; Jones et al., 1996). When HSR birds have reached conspecifics, they stay closer to them and for longer than do LSR quail (Launay et al., 1991; Jones et al., 1996; François et al., 1998, 1999). This genetic selection on social reinstatement tendency does not seem to have modified other behavioural characteristics (Launay, 1993). Selection on social reinstatement tendency was realized on 7 12-day old quail chicks. For economic reasons, it is preferable to evaluate birds when they are young so as to eliminate nonreproductive birds well before their sexual maturity (Mills et al., 1994). However, the social behaviour characteristics measured in young birds must be known to persist in adults. François et al. s (1999) paired open-field tests revealed no differences in interindividual distances between the two lines when quail were 6-weeks old. Moreover, HSR adult males appeared more aggressive towards other males than did LSR males (François, 1999). These previous reports suggested that social motivation decreased with age (François et al., 1998, 1999). No other data on the persistence of social motivation level during quail s life have been reported. Therefore, although this selection discriminates young quail very well, differences between the two lines when they become adult are poorly known, especially for females. The aim of this study was to analyse and to compare the social motivation of adult female quail (birds of more than 10-weeks old) from these two selected lines. As social behaviour is a complex occurrence, we used several ethological methods to characterize social motivation, by placing the animal either in social isolation or in social contact. We also evaluated and compared the levels of emotional reactivity of our different subjects, because emotional traits influence the animal responses in some context. 2. Materials and methods 2.1. Animals and rearing conditions Test subjects were female Japanese quail (Coturnix coturnix japonica) from the F38 generation of two lines genetically selected for high (HSR) or low (LSR) levels of social reinstatement behaviour (Mills and Faure, 1991; Faure and Mills, 1998). The HSR and LSR lines had been selected on the basis of their behaviour in a treadmill described by Mills and Faure (1990): 6 8-day old chicks were tested individually for 5 min in the treadmill, with five conspecifics placed in a target box in front of them. Latencies to vocalize and to move were recorded. When a chick moved towards the target box, the treadmill carpet was activated in the opposite direction to prevent the bird from reaching its conspecifics. Distance covered by the subject, measured in arbitrary units, was noted after 5 min, integrating the time past in the far zone. Chicks having covered long distances on the treadmill carpet were considered to have a high level of social motivation (HSR). On the contrary, chicks having covered short distances were considered to have a low social motivation level (LSR). This social behaviour was weighted for independence from fear (tonic immobility) responses (Mills and Faure, 1991). Mills and coauthors (Mills and Faure, 1991; Mills et al., 1994; Faure and Mills, 1998) have described these lines and their responses to selection.

We tested 24 HSR and 24 LSR adult females that had been reared in group of same line in avian center (Station de Recherche Avicole SRA France). All birds were tested at young age with treadmill and presented extreme social reinstatement behaviour. Females were 12-weeks old at the beginning of this study. They were not sexually developed during the study but they had had a reproductive phase before the experiment. They were wing-banded and housed, in single-line groups of 12 individuals, in four boxes (length 225 width 185 height 280 cm) with wood shavings on the floor. Ambient temperature was 20 2 8C and the non-stimulant photoperiod was LD 10:14 h. Food and water were available ad libitum. The physiological state of the females was controlled by measuring regularly their weight and the length of their cloacal vent (Siopes and Wilson, 1975). HSR birds were always lighter than LSR birds and length of the cloacal vent of HSR were smaller than LSR. 2.2. Behavioural tests L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 177 A time schedule with an overview of all tests realised is presented in Table 1. 2.2.1. Tests in social isolation 2.2.1.1. Tonic immobility (TI) test. Our protocol was similar to that described by Jones (1986). Animals were caught and carried individually into a new room. Quail were placed on their back in a U-shaped apparatus and restrained for 5 s prior to release. If more than 10 s lapsed between the release of the subject and its escape, TI duration was noted (the minimum score for TI duration was then 11 s). If not, the experimenter immediately reattempted to induce TI again. If TI had not been induced after five induction attempts, a TI duration score of 0 s was noted. If the quail had not yet attempted to stand up after 5 min, the test was stopped, and a maximum score of 300 s was allocated. TI duration is positively correlated to animal fear level (Mills et al., 1994). Birds were tested when they were 3-months old (HSR n = 24, LSR n = 24). 2.2.1.2. Open-field test. This test consists in transferring an animal from its familiar rearing place to a new environment (Launay, 1993). In our test, quail were placed individually in the middle of a wire-netting cylinder (diameter 120 height 62 cm) on a linoleum floor, for 5 min. The cylinder was completely empty. This apparatus was placed in the centre of an unfamiliar box similar to their rearing box (shape, surface, colour, and light), but without food, water or wood shavings on the floor. Hidden behind a black curtain with an observation window, the experimenter noted latency of first step, numbers of steps, jumps and the subject s observations (high or low postures). Low posture observations Table 1 Time schedule of realised tests and measured parameters Age Tests Measured parameters 3 months Tonic immobility Emotional reactivity Open-field Social motivation 5 months Runway Social motivation + emotional reactivity 7 months Emergence Social motivation + emotional reactivity Separation Social motivation 8.5 months Confrontation Social motivation

178 L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 correspond to observations when animals are lying, crouching or standing upright (in a position identical to the walking position). High posture observations correspond to typical observations when animals stand on tiptoe with their bodies very straight. An occurrence of an observation starts and ends when the bird changes its posture. Latency of first distress call and number of distress calls were also recorded. Quail were tested when they were 3-months old (HSR n = 24, LSR n = 24). 2.2.1.3. Emergence test. This test followed a protocol similar to that described by Jones et al. (1991). Quail were placed individually in an opaque transport box (18 cm 18 cm 18 cm). This box was placed at the entrance of a larger well-lighted experimental box (length 43 depth 40 height 48 cm) equipped with an observation window. The transport box was kept closed for 1min and then was opened for 3 min. Latencies of first distress call and number of distress calls, before and after the transport box was opened, were recorded. The experimenter also noted latencies of emergence of head, feet and whole body from the transport box into the experimental box. If a quail had not emerged after 3 min, a maximum score of 180 s was recorded. When a quail emerged from the box, the experimenter noted numbers of movements (walks), jumps, and observations (high or low postures). Quail were tested when they were 7-months old (HSR n = 24, LSR n = 23). 2.2.1.4. Separation test. Quail were placed, individually, for 3 min in a box similar to their rearing box (shape, surface, colour, light, wood shavings on the floor, but no water nor food), but without any conspecifics. Using a recording system (video + audio), the experimenter recorded numbers of steps and jumps. Latency of first distress call and number of distress calls were also recorded. Subjects were tested when they were 7-months old (HSR n = 24, LSR n = 23). 2.2.2. Tests in the presence of conspecifics 2.2.2.1. Runway test. This test was based on a similar principle than the treadmill test (Mills and Faure, 1990), but contrary to the first test, our procedure permits the bird to reach conspecifics. The runway was a 120 cm-long straight wire netting tunnel with a cage (length 36 width 22 height 36 cm) at one end, containing two quail (Fig. 1). This tunnel was divided into three zones of equal sizes: zones A (beginning of the tunnel), B (middle of the tunnel), and C (end of the tunnel). Zone C included a proximal area situated less than 1 bird long from the end cage. This apparatus was placed in the centre of a box similar to the rearing box (shape, surface, colour, and light), but without food, water or wood shavings on the floor. An opaque transport box, with the test subject inside, was placed at the entrance of the tunnel and kept closed. For Fig. 1. Runway test set-up.

1 min, the experimenter recorded the latency of first distress call and number of distress calls. Then the transport box was opened. For 5 min, hidden behind a black curtain with an observation window, the experimenter recorded the time the subject took to emerge entirely from its transport box, latency of first distress call, number of distress calls as well as the following parameters: time to reach conspecifics, exploration of the cage containing conspecifics (number of pecks on the cage), aggressions (number of pecks against conspecifics), fear attitudes (freezing, crouching on the floor), jumps. Total time spent in each zone was recorded. Birds were tested when they were 5-months old (HSR n = 24, LSR n = 23). Each quail was tested twice: once with familiar same-line conspecifics (same rearing box) and second with nonfamiliar same-line conspecifics (different rearing boxes). Half the birds were tested with familiar conspecifics first; the other half were tested with non-familiar conspecifics first. All test birds were also used twice as familiar and twice as non-familiar conspecifics. 2.2.2.2. Intra-line confrontation test. Subjects were 8.5-months old non-familiar quail (females maintained in different boxes) (HSR n = 2 12, LSR n = 2 10). The experimental device was a rectangular confrontation cage with a transparent wall and a floor covered with wood shavings. The cage could be divided lengthwise into two parts (a small part: length 17 depth 45 height 30 cm and a large part: length 55 depth 45 height 30 cm) with removable glass or opaque partitions inserted from the cage top. All tests were video and audio recorded. At the beginning of a trial, one quail was placed in each part without any possible visual contact with the conspecific (both opaque and glass partitions lowered). After a 1 min familiarization period, the opaque partition between the two parts was removed, allowing visual (glass partition lowered), but not direct, contact. After 3 min visual contact, the experimenter withdrew the glass partition so the birds could then make direct contact. All fear postures (freezing, crouching on the floor) and agonistic behaviours (attacks, pecks, pursuits, running away, and avoidances) were then recorded for 5 min. For ethical reasons and animal respect, the test would have been stopped if birds had injured severe aggression especially in the case of pecks in head (eyes) part. 2.3. Statistical analyses Non-parametric statistical tests evaluated differences using R statistic software (http:// www.r-project.org/). 3. Results 3.1. Tests in social isolation L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 179 Only two HSR and one LSR birds could not be induced after 5 attempts in the TI test. Numbers of induction attempts did not differ significantly between HSR birds (1.9 0.3 inductions) and LSR birds (1.5 0.2 inductions) (Mann Whitney, p > 0.1). TI durations did not differ significantly between induced HSR (104.9 17.3 s) and LSR (85.1 16.3 s) females (Mann Whitney, p > 0.1). Latencies of first step in the open-field test did not differ significantly between the two lines (Mann Whitney, p > 0.1). Similarly, no differences in the emission of vocalisations (latency of first call and number of distress calls) during the test could be evidenced between the two lines (Mann Whitney, p > 0.1).

180 L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 However, HSR females performed significantly more steps than LSR females (Mann Whitney, p = 0.032) (Fig. 2A). Although a few HSR females (2/24) but no LSR females (0/24) jumped during a test, this difference was not significant (Fisher, p > 0.1). Finally, HSR females expressed nearly twice as many high posture observations in the open-field arena as LSR birds (Mann Whitney, p = 0.016) (Fig. 2B). Neither emission of vocalisations (latency of first call and number of distress calls) nor emergence latencies recorded in the emergence test differed significantly between the two lines (Mann Whitney, p > 0.1). Six HSR and four LSR quail did not emerge from the transport box. In the larger experimental box, mean numbers of moving did not differ significantly between the two lines (Mann Whitney, p > 0.1). Nevertheless, as in the open-field tests, although a few HSR females (4/24) but no LSR females (0/23) jumped during a test, this difference was not significant (Fisher, p = 0.055). Again, HSR females expressed significantly more high posture observations (4.4 1.5 observations) than did LSR females (0.4 0.3 observation) (Mann Whitney, p = 0.009) (Fig. 3). In the separation test, locomotor activity of HSR females (235.2 43.1 steps) was significantly higher than that of LSR females (77.6 13.1 steps) (Mann Whitney, p = 0.002) (Fig. 4). During this test, 1/24 HSR versus 0/23 females jumped. The difference was not significant (Fisher, p > 0.1). Fig. 2. Open-field test. (A) Steps (mean numbers S.E.) and (B) high posture observations (mean numbers S.E.) by HSR (black bars) and LSR (white bars) females; * p < 0.05.

L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 181 Fig. 3. Emergence tests. High posture observations (mean numbers S.E.) by HSR (black bars) and LSR (white bars) females; ** p < 0.01. No differences for vocalisation parameters (latency of first call and number of distress calls) were evidenced between the two lines during this test (Mann Whitney, p > 0.1). 3.2. Tests in the presence of conspecifics The runway test revealed no significant differences between the two lines for emergence latencies or for vocalisation parameters (Mann Whitney, p > 0.1). In the runway, similar differences between the two lines were observed whatever the conspecifics presented (familiar or non-familiar). HSR females reached their conspecifics significantly faster (Mann Whitney, p = 0.016 familiar; p = 0.001 non-familiar) and stayed longer near them (proximity zone) (Mann Whitney, p = 0.017 familiar; p = 0.003 non-familiar) than LSR females (Fig. 5A). Conversely, LSR females tended to remain longer in the non-social zones (zones A + B) than HSR females (Mann Whitney, p = 0.017 familiar; p = 0.084 nonfamiliar). Finally, HSR females explored the target box significantly more often than LSR females did (Mann Whitney, p = 0.034 familiar; p = 0.003 non-familiar) (Fig. 5B). On the contrary, LSR females spent significantly more time near familiar than near nonfamiliar conspecifics (Wilcoxon, p = 0.033) (Fig. 6A) and they tended to reach their conspecifics Fig. 4. Separation tests. Steps (mean numbers S.E.) by HSR (black bars) and LSR (white bars) females; ** p < 0.01.

182 L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 Fig. 5. Runway tests. (A) Latencies (in s) to reach conspecifics (mean times S.E.), and time (in s) spent in the proximal area (mean S.E.) and (B) exploration of conspecifics cage (mean S.E.) by HSR (black bars) and LSR (white bars) females; * p < 0.05; ** p < 0.01. faster when they were familiar than when they were not (Wilcoxon, p = 0.099) (Fig. 6B). These differences were not observed in HSR females (Wilcoxon, p > 0.1). But we found that HSR females left the proximal area in zone C more often to return to zones B (Wilcoxon, p = 0.015) and A (Wilcoxon, p = 0.047) when conspecifics were non-familiar than when they were familiar. LSR quail did not express this difference in relation to familiarity of conspecifics (Wilcoxon, p > 0.1). Levels of agonistic interactions (attacks, pecks, pursuits, running away, and avoidances) did not differ significantly between the two lines in the confrontation test (Mann Whitney, p > 0.1). In fact, adult females expressed few agonistic interactions during this test. Number of fear postures (freezing, crouching on the floor) did not differ significantly between the HSR and LSR lines (Mann Whitney, p > 0.1). Emissions of vocalisations (latency and number of calls) did not differ significantly between HSR and LSR females (Mann Whitney, p > 0.1). 4. Discussion The aim of our study was to investigate the persistence in adults of social traits selected in juveniles. Therefore, ethological tests evaluated and compared adult quail from the LRS and

L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 183 Fig. 6. Runway tests. (A) Time (in s) in the proximal area (mean S.E.) and (B) time (in s) to reach conspecifics (mean S.E.) for HSR (black bars) and LSR (white bars) females; plain bars: familiar subjects; striped bars: non-familiar subjects; * p < 0.05; # p < 0.1. HRS lines. We evidenced, for the first time, the persistence into adulthood of behavioural characteristics selected on young birds. However, we also evidenced that behaviour changed with age, as not all the behavioural characteristics observed in young birds persisted into adulthood. Selection of HSR and LSR lines were weighted for independence from fear (tonic immobility) (Mills and Faure, 1991). This independence from fear responses persisted in adulthood: tonic immobility tests revealed no significant differences between HSR and LSR adult females. Similarly, latencies of emergence from a sheltered area into an open and unfamiliar one (emergence test, runway test) revealed no differences between lines. These parameters are known to be good estimates of emotional reactivity: fearful animals take longer to emerge (Archer, 1976; Mills and Faure, 1986; Jones, 1987). Our results confirm and extend previous reports for 10-day old to 10-weeks old quail from these lines (Mills and Faure, 1991; Launay, 1993). Like HSR and LSR chicks, our adult females presented important differences in their levels of social motivation. In the isolation tests, HSR quail expressed higher locomotor activity than LSR birds. Moreover, HSR females were the only ones that jumped when they were alone although we never found any significant difference with LSR females. These differences were also observed

184 L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 in 6 10-day old juveniles (Launay et al., 1991). Jumping (Faure et al., 1983) and high locomotor activity (Jones and Merry, 1988) indicate active search for conspecifics and strong motivation to re-establish contact with conspecifics. Our HSR females presented more high posture observations than LSR birds did. This behaviour has not been described in previous reports on social motivation. Although this behaviour is often considered as a temporary increase in attention level following a change in the environment, it is also performed simultaneously with the emission of distress calls (Guyomarc h, 1974). Thus, the many high posture observations of HSR females could reflect, as jumping and locomotor activity, active search for visual contact with conspecifics. HSR chicks, in isolation, emit distress calls earlier and in greater quantities than LSR chicks (Launay, 1993). Frequencies of distress calls appear to be positively correlated to social motivation level (Jones and Merry, 1988). We did not find that vocal production of adult females differed significantly between the two lines. Isolated adult birds, whatever their line, produced very few distress calls. This result, which contrasts with previous reports, could be related to the age, sex, and physiological state of our birds. Indeed, Collias (1952) showed that the frequency of distress calls uttered by domestic chicks in response to social isolation decreased with age. Moreover, non-laying adult female quail emit less vocalisations than males (Guyomarc h, 1974). Although young birds express distress during social separation mainly by emitting distress calls, this no longer seems to be the case for adult females. Tests with conspecifics revealed differences in social motivation of females between the two lines. During the runway test, HSR females ran quicker towards their conspecifics and spent longer near them than did LSR females. These females tended to remain relatively far from their conspecifics. This suggests that, as in chicks (as this test is based on the same principle as the treadmill test) (Mills and Faure, 1990), adult HSR females have a higher motivation to remain in contact with conspecifics than do LSR. In addition, behaviour of quail from these two lines differed in presence of familiar or of nonfamiliar conspecifics. LSR females joined quicker familiar quail than non-familiar conspecifics and remained longer near familiar than near non-familiar conspecifics. This difference was not found in HSR quail. Nevertheless, HSR quail left the proximal area for the other tunnel zones more often when non-familiar conspecifics were present in the target box. This difference was not found in LSR quail. The capacity to modulate their behaviour according to their familiarity with conspecifics presented also exists in 7-day old HSR and LSR chicks. Indeed, young chicks, whatever their social line, approached quicker familiar conspecifics than non-familiar conspecifics. They remained longer in contact with them and maintained shorter interindividual distances (Mills and Faure, 1991; Jones et al., 1996). So, it appears that juveniles of both lines modulate their behaviour according to their partners status. Our study tends to confirm this trend, but adult HSR and LSR quail no longer seem to express familiar/non-familiar discrimination in the same way. Although the two lines differ in their motivation to join and to stay near conspecifics, no differences were observed in their interactions in the confrontation tests. In particular, levels of aggressiveness did not differ significantly between the two lines. Our females were never very aggressive. However, HSR males are more aggressive than LSR males (François, 1999). This discrepancy between the two reports could be related to sex. Until now, no study has concerned females. On the other hand, immature males appear less aggressive than sexually mature males (Guyomarc h and Guyomarc h, 1986). The low level of aggressiveness of our females may be related to the level of their physiological development: they were not sexually developed.

L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 185 In conclusion, our study shows that selection on social motivation at young age persists in adults: social motivation of HSR females is higher than that of LSR females, in isolation as well as in the presence of conspecifics. This persistence of a selected behavioural trait is very important for husbandry that has to keep animals in different age classes: when adults, HSR birds are still more attracted by conspecifics than are LSR ones, so we suppose that adult HSR quail, as HSR chicks, are better adapted to husbandry condition. This good correlation between behavioural phenotypes of young and adults has been demonstrated in lines of quail selected on their emotive reactivity: fear level of an individual persists all through its life (Launay et al., 1993). However, contrary to these authors, we evidenced some modifications of social motivation, with disappearance of some juvenile characteristics (vocalisations). Acknowledgment We are grateful to Dr. Ann Cloarec for helpful comments and proofreading of this manuscript. References Archer, J., 1976. The organization of aggression and fear in vertebrates. In: Bateson, P.P.G., Klopfer, P.H. (Eds.), Perspectives in Ethology. Plenium Publishing Corporation, New York, pp. 231 298. Collias, N.E., 1952. The development of social behavior in birds. The Auk. 69 (2), 127 159. Faure, J.M., Jones, R.B., Bessei, W., 1983. Fear and social motivation as factor in open field behaviour of the domestic chick. A theoretical consideration. Biol. Behav. 8, 103 116. Faure, J.M., Mills, A.D., 1998. Improving the adaptability of animals by selection. In: Grandin, T. (Ed.), Genetics and the Behaviour of Domestic Animals. Academic press, London, pp. 235 264. François, N., 1999. Influence d une sélection pour la motivation sociale sur le comportement social des cailles domestiques (Coturnix japonica). Thèse, Université de Rennes I. François, N., Mills, A.D., Faure, J.M., 1998. Place preferences of Japanese quail given a permanent choice between a social or a non-social but enriched situation. Behav. Process. 43, 163 170. François, N., Mills, A.D., Faure, J.M., 1999. Inter-individual distances during open-field tests in Japanese quail (Coturnix japonica) selected for high or low levels of social reinstatement behaviour. Behav. Process. 47, 73 80. Guyomarc h, J.C., 1974. Les vocalisations des gallinacés - structure des sons et des répertoires. Ontogenèse motrice et acquisition de leur sémantique. Thèse, Université de Rennes I. Gross, W.B., Siegel, H.S., 1983. Evaluation of the heterophil/lymphocyte ratio as a measure of stress in chickens. Avian Dis. 27, 972 979. Guyomarc h, J.C., Guyomarc h, C., 1986. Etude morphologique et comportementale de la sexualité chez la caille japonaise : développement des mâles. Cah. Ethol. Appl. 6, 185 208. Hughes, B.O., 1975. The concept of optimum stocking density and its selection for egg production. In: Freeman, B.M., Boorman, K.N. (Eds.), Economics Factors Affecting Eggs Production. British Poultry Science, Edinburgh, pp. 271 298. Jones, R.B., 1986. The tonic immobility reaction of the domestic fowl: A review. World s Poult. Sci. J. 42, 82 97. Jones, R.B., 1987. The assessment of fear in the domestic fowl. In: Zahan, 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 on related effects of environmental novelty: an annoted bibliography, 1960 1988. Psychol. Rec. 39, 397 420. Jones, R.B., Harvey, S., 1987. Behavioural and adrenocortical responses of domestic chicks to systematic reductions in group size and to sequential disturbance of companions by the experimenter. Behav. Process. 14, 291 303. Jones, R.B., Hugues, B.O., 1986. Feafullness and abnormalities in the chicken s egg shell Is there a link? In: Czako, J. (Ed.), Proc. Inter. Symp. Appl. Ethol. Farm Anim., Hunharian Academy of Sciences, pp. 171 175. Jones, R.B., Merry, J., 1988. Individual or pairs exposure of domestic chicks to an open field: some behavioural and adrenocortical consequences. Behav. Process. 16, 75 86. Jones, R.B., Mills, A.D., Faure, J.M., 1991. Genetic and experimental manipulation of fear-related behavior in Japanese quail chicks (Coturnix coturnix japonica). J. Comp. Psychol. 105, 15 24.

186 L. Formanek et al. / Applied Animal Behaviour Science 112 (2008) 174 186 Jones, R.B., Mills, A.D., Faure, J.M., 1996. Social discrimination in Japanese quail Coturnix japonica chicks genetically selected for low or high social reinstatement motivation. Behav. Process. 36, 117 124. Jones, R.B., Williams, J.B., 1992. Responses of pair-housed male and female domestic chicks to the removal of a companion. Appl. Anim. Behav. Sci. 32, 375 380. Keeling, L.J., 1987. Social Spacing in Domestic Fowl. Ph.D. Thesis, Edinburgh University, p. 225. Launay, F., 1993. Conséquences comportementales et physiologiques de sélections pour l émotivité et l attraction sociale chez la caille japonaise (Coturnix japonica). Thèse, Université de Rennes I. 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. Process. 24, 95 110. Launay, F., Mills, A.D., Faure, J.M., 1993. Effects of age, line and sex on tonic immobility responses and social reinstatement behaviour in Japanese quail Coturnix japonica. Behav. Process. 29, 1 16. Le Masne, G., 1996. Sociétés animales. Encyclopaedia Universalis. Corpus 21. Mills, A.D., Faure, J.M., 1986. The estimation of fear in domestic quail: correlations between various methods and measures. Biol. Behav. 11, 235 243. Mills, A.D., Faure, J.M., 1990. The treadmill test for the measurement of social motivation in Phasianidae chicks. Med. Sci. Resc. 18, 178 180. Mills, A.D., Faure, J.M., 1991. Divergent selection for duration of tonic immobility and social reinstatement behavior in Japanese quail (Coturnix coturnix japonica) chicks. J. Comp. Psychol. 105 (1), 25 38. Mills, A.D., Crawford, L.L., Domjan, M., Faure, J.M., 1997. The behavior of the Japanese or domestic quail Coturnix japonica. Neurosci. Biobehav. R. 21, 261 281. Mills, A.D., Jones, R.B., Faure, J.M., Williams, J.B., 1993. Responses to isolation in Japanese quail genetically selected for high or low sociality. Physiol. Behav. 53, 183 189. Mills, A.D., Launay, F., Turro, I., Jones, R.B., William, J.B., Faure, J.M., 1994. Sélection divergente sur la peur et la sociabilité chez les cailles japonaises Coturnix japonica. Réponses et conséquences. Dans : Comportement et bienêtre animal (coord. : Picard M., Porter R.H., Signoret J.P.) INRA Editions. Satterlee, D., Roberts, E.D., 1990. The influence of stress treatment on femur cortical bone porosity and medullary bone status in Japanese quail selected for high or low blood corticosterone response to stress. Comp. Biochem. Physiol. 95, 401 405. Siopes, T.D., Wilson, W.O., 1975. The cloacal gland: an external indicator of testicular development in Coturnix. Poult. Sci. 54, 1225 1229. Vallortigara, G., 1992. Affiliation and aggression as related to gender in domestic chicks (Gallus gallus). J. Comp. Psychol. 106, 53 57. Vallortigara, G., Cailotto, M., Zanforlin, M., 1990. Sex differences in social reinstatement motivation of the domestic chick (Gallus gallus) revealed by runway tests with social and non social reinforcement. J. Comp. Psychol. 104, 361 367. Wood-gush, D.M.G., 1971. The Behaviour of the Domestic Fowl. Heinemann Educational Books ltd, London, p.147.