www.elsevier.comrlocaterapplanim
Early locomotor behaviour in genetic stocks of
chickens with different growth rates
D. Bizeray
), C. Leterrier, P. Constantin, M. Picard, J.M. Faure
Station de Recherches AÕicoles, Centre I.N.R.A. de Tours, 37380 Nouzilly, France
Accepted 19 January 2000
Abstract
Reduction in exercise increases the occurrence of lameness in meat-type chickens. Locomotor activity is dramatically reduced during the finishing period in chickens from fast-growing genetic types compared to slow-growing genetic types, but it is not known whether this difference is already present during the starting period and may be influenced by genetic factors. In order to define the effect of genetic origin on early locomotor behaviour, exercise was compared from 1 to
Ž .
22 days of age in two meat-type chicken stocks differing in growth rate: male broilers B which
Ž .
grow fast and are often lame, and male ‘‘label rouge’’ chickens L which grow slowly and are rarely lame.
Ž .
Time budget lying, standing, drinking, eating, walking was measured by scanning in six
Ž 2.
repetitions of five birds densitys2.5 birdsrm at 1, 8, 15 and 17 days of age. Standing bouts were analysed by focal sampling at 2–3, 6–7, 13–14 and 20–21 days of age.
Ž
B chicks spent less time standing than L chicks at 15 days of age Bs13"2%, Ls24"1%,
. Ž
P-0.01 and 17 days of age, and spent more time lying at 17 days of age Bs73"3%,
.
Ls60"4%, P-0.05 .
Ž .
The major part 74% of the total active time observed by focal sampling was linked to feeding activity. At 2 and 3 days, the activity of B chicks was half that of L chicks during standing bouts
Žduration of walking per bout: 19"4 s for B; 45"4 s for L, P-0.05 . The activity observed by.
focal sampling during non-feeding bouts at 20–21 days was significantly correlated with the corresponding data recorded at 2–3 days in the same chicks in the B stock but not in the L stock.
Ž .
We concluded that 1 both B and L genetic stocks have the same overall activity during the
Ž .
first 3 days of age scanning but they exhibit different organisation and composition of standing
Ž . Ž .
bouts focal sampling . 2 Genetic factors are probably involved in the expression of locomotor
Ž .
behaviour in very young chicks. 3 The correlations between the levels of activity at early and
)Corresponding author. Tel.:q33-24742-7945; fax:q33-24742-7778.
Ž .
E-mail address: [email protected] D. Bizeray .
0168-1591r00r$ - see front matterq2000 Elsevier Science B.V. All rights reserved. Ž .
later ages suggest that selection of young mobile broiler chicks might increase activity at a later age and might therefore reduce the occurrence of leg abnormalities.q2000 Elsevier Science B.V.
All rights reserved.
Keywords: Broiler chicken; Locomotion; Walking; Genetic
1. Introduction
Leg weakness often results from non-infectious disorders of the skeletal system in
Ž . Ž
broiler chickens Thorp, 1994 , with numerous bone abnormalities tibial
dyschon-. Ž
droplasia, tarsal angulation, femoral epiphysiolysis . Bone quality mass, shape, internal
.
architecture is known to be influenced by the mechanical stresses that are applied to the
Ž .
bones Lanyon, 1992 . Increased levels of activity strengthen bones and reduce bone
Ž .
deformities Reiter and Bessei, 1998 , whereas a lack of exercise can increase the
Ž .
incidence of leg abnormalities in broilers Haye and Simons, 1978 .
Locomotor activity is a component of many behavioural patterns but it may have lost part of its adaptive value for meat-type fowl in the present housing systems where heat, food and water are dispensed within easy reach. Selection for feed conversion has also probably reduced the need or desire for energy consuming behaviours such as running or long walks, just as the aggressive behaviour has been affected by selection for high
Ž .
growth rate Mench, 1988 . Better knowledge of locomotor behaviour is needed in broiler chickens in order to counterbalance this trend and its dramatic consequences on leg condition and welfare. This requires better understanding of how the behaviour develops, the motivations involved and which factors can enhance the pattern. Few papers describe the ontogeny of locomotor behaviour because it is often associated with
Ž
other patterns such as feeding, imprinting and social behaviour reviewed by Rogers,
.
1995 . In contrast, several authors have studied the factors modifying locomotor activity,
Ž .
especially during the finishing period 3–6 weeks . Using environmental manipulation to enhance activity in chickens has been widely examined, including increased floor area
Ž
and modified stocking density Rodenhoff and Dammrich, 1971; Lewis and Hurnik,
¨
. Ž
1990 , increased distance between feeder and drinker Simons, 1986; Reiter and Bessei,
. Ž .
1996 , use of perches Hughes and Elson, 1977; Newberry et al., 1995 , light colour and
Ž . Ž
intensity Praytino et al., 1997a,b , forced displacements using fan movements Lei and
.
Van Beek, 1997 .
Genetic variability of locomotor activity has been much less studied. Differences in
Ž .
growth rate lead to modifications in feeding Boa-Amponsem et al., 1991 and in
Ž .
locomotion Reiter and Bessei, 1997 , but it has not been established whether these modifications are induced by genetic factors or by differences in body weight. In order to clarify this point, the behaviour of chicks was studied in the present experiment during early life, when body weights are very close in genetic stocks with different growth rates. This starting period is very interesting because during this period the rates
Ž .
differences in activity during the starting period can be used as behavioural markers to select more mobile chickens.
2. Materials and methods
2.1. Birds
The study comprised two experimental groups: 30 male broiler chicks from a fast
Ž .
growing genetic stock B, IJV 915, Institut de Selection Animale, Lyon, France and 30
´
Ž .
male ‘‘label rouge’’ chicks L, T551, SASSO, Sabres, France from a slow growing genetic stock. Chicks were provided by the same commercial hatchery. There were six replicates for each treatment. Five birds from each stock were randomly assigned to each replicate. They were wing banded at hatching and weighed. All birds were reared from hatching to 22 days of age. Very low stocking densities were used because total potential activity needs to be described without overcrowding effects. Low densities enabled us to identify all individuals in order to study interindividual correlations and facilitate behavioural observations. Each group was kept in a 2 m2 floor pen covered
Ž 2.
with wood shavings densitys2.5 birdsrm . Each pen contained one drinker and a 1-m-long feeder. To avoid spillage the food was covered with a wire mesh until the chicks were 10 days old. The lighting schedule was 24 L. Chicks were fed a standard
Ž
diet metabolizable energys12.98 kJ, crude proteins22%, Cas0.15%, available
.
Ps0.42% . Food was provided in the form of mash during the first 3 days of life. From
Ž .
3 to 7 days of age, mash and pellets diameters2.5 mm were mixed. Only pellets were provided from day 8. Food and water were available ad libitum. At 17 days of age, every chick was individually observed in the pen in order to detect early lameness. At
Ž .
the end of the experiment day 22 , each individual was weighed and abnormal tarsal
Ž . Ž .
angulations were detected varus or valgus , according to Leterrier and Nys’s 1992 classification.
2.2. ObserÕations
Four birds were chosen at random from each pen at the beginning of the experiment
Ž .
and identified by colours red or blue marks on the head or the body , the fifth was not marked. Birds were marked again when necessary throughout the rearing period. The observer sat 2 m above the birds and 50 cm back from pens during observation. Chicks
Ž .
were observed by scan sampling and focal sampling Martin and Bateson, 1986 . All observations were made by the same observer.
2.2.1. Scan sampling
At 1, 8, 15 and 17 days of age, each floor pen was scanned for 90 min. The number
Ž
of individuals which expressed each category of behaviour standing, eating, drinking,
.
Ž .
had its beak in water or stood in front of the drinker with raised head swallowing . The number of ‘‘active immobile’’ chicks was calculated as the sum of the birds standing or eating or drinking.
2.2.2. Focal sampling
Ž .
Since chickens lie for 80% to 90% of their time Bessei, 1992 , only behavioural patterns expressed during the ‘‘standing’’ bouts were recorded, i.e. behaviours expressed from the moment when a bird stood up until it lay down. The observations were sampled
Ž .
during four periods of 2 days 2–3, 6–7, 13–14, 20–21 days after hatch . About 10 behavioural bouts were recorded per pen and per day of observation. Behaviours were
Ž .
recorded using the Observer 3.0 Noldus Information Technology, Netherlands . This
Ž
software records duration of the different states walking, running, standing, eating,
. Ž
drinking, preening and exploring litter and occurrence of the different events each step
.
and interaction with a congener . A bird was considered to be walking only when it was mobile ; when it was walking step by step, short periods of immobility were recorded as standing. Eating and drinking were considered to have stopped as soon as the bird stood inactive, even if it was in front of a feeder or a drinker. Exploring litter included pecking and scratching the floor. Preening contained feather licking and scratching itself. The
Ž .
term interaction was used for behaviours antagonistic or not which involved contact between the interacting pair during the standing bouts: pecking, leaping, sparring, preening each other and stepping on another bird.
For analysis, standing bouts were sorted into two types: feeding bouts in which
Ž .
chicks drank or ate and other bouts non-feeding bouts . These two types of bout were analysed separately because the motivation for activities in each kind of bout was believed to be different.
2.3. Data analysis
For all data analysed, the pen was the experimental unit. Pen means were therefore used to examine the effects of age and genetic strain. Because data were not normally distributed, they were analysed with nonparametric statistical tests using Statview 4.05. Genetic effects were tested with the Mann–Whitney U test, and age effects for each strain using the Friedman test. Feeding and non-feeding bouts were compared using Wilcoxon test. The correlation coefficients between body weight and the different behaviours at different ages were computed using Spearman’s rank correlation. For focal sampling, we chose to eliminate from analysis the birds which were lame at day 17
Žns5 in order to avoid any possible effect on general activity Weeks and Kestin,. Ž .
1997 .
3. Results
3.1. Bird growth
Ž .
Body weight was significantly lower at hatching in B chicks 36.3"0.3 g than in L
Ž . Ž .
ŽZs6.65, P-0.001 : 707.3. "16.2 g in B chicks versus 413.7"6.7 g in L chicks. Body weight at hatching was significantly related to 3-week body weight in B birds
ŽRs0.595, ns30, P-0.01 , but not in L birds. Mortality rate was zero in both.
groups.
3.2. Leg problems
No L birds became lame. Four B chicks were identified as lame at 17 days of age and had abnormal tarsal angulation at day 22. One B chicken was lame at day 17 but did not present any tarsal defect at day 22. Another B chicken had a valgus angulation at day 22 but was not identified as being lame at day 17.
3.3. Time budget
The behaviour data converted to proportions of time spent in each activity are
Ž . Ž .
presented as means and standard errors SE Fig. 1 . Most of the birds were lying during observations. When all groups and ages were mixed, chicks spent 67% of the
Ž .
time lying, 28% of the time active immobile eatingqdrinkingqstanding and only 5% of the time walking.
Walking time did not significantly differ according to age or to genetic factor but every other behavioural pattern was significantly affected by age: in the B group, level
Ž
of activity fluctuated with age numbers of birds lying, drinking, eating and standing, respectively, Hs8.6, Hs8.34, Hs7.8, Hs7.6, dfs3, P-0.05 using Friedman
. Ž
test and activity of L chicks increased gradually with time numbers of eating and
.
standing birds, respectively Hs8.49, Hs9.81, dfs3, P-0.05, using Friedman test .
Ž .
Fig. 1. Time budget in both genetic types at different ages six repetitions of five birds per treatment .
U UU
Ž .
B birds spent significantly less time standing and actively immobile than L birds at
Ž . Ž .
15 and 17 days Fig. 1 . B chicks also spent more time lying at 17 days Fig. 1 .
3.4. Standing bouts
During the four 2-day periods of focal sampling, 836 standing bouts were recorded. The mean standing bout duration was 55.1"15.4 s. Feeding and non-feeding bouts differed widely in general characteristics: mean durations of feeding bouts differed
Ž
significantly from those of non-feeding bouts 212.6"83.3 s versus 19.6"8.8 s,
.
Zs3.05, P-0.001, using Wilcoxon test, Fig. 2 . Birds walked significantly more in
Ž
feeding bouts than in non-feeding bouts 33.7"4.1 s versus 3.4"0.52 s, Zs3.05,
. P-0.001 using Wilcoxon test .
Ž .
Fig. 2. Mean total duration of the standing bouts feeding and non-feeding bouts at different ages. Number of standing bouts registered is noted into columns. Pen of birds was the statistical unit after removing the five
U
Ž .
3.4.1. Feeding bouts
Feeding bouts represented 21% of the recorded bouts and 74% of the total time observed. Durations of feeding bouts in B chicks were half as long of those of L chicks
Ž .
at days 2–3 and at days 20–21 Fig. 2 .
Duration and occurrence of the behavioural patterns were little affected by age: only duration of eating and number of interactions decreased in L chicks at 6–7 and 13–14
Ž .
days Table 1 .
Locomotor behaviour was not significantly related to age but to genetic type, except running, which was very rare: L chicks were more active than B chicks at days 2–3
Žwalking, standing, numbers of steps, respectively, Zs2.56, Zs2.19, Zs2.74, P
-.
0.05 using Mann–Whitney’s U test , although they were heavier. The same differences
Table 1
Ž .
Duration s of different states and occurrence of different events during feeding bouts recorded on focal samplinga
Ž .
Age days 2–3 6–7 13–14 20–21 Age effect
Genetic type mean"SE mean"SE mean"SE mean"SE
Locomotor behaÕiour Ž .
Walking s L 44.8"4.4 42.4"7.0 35.9"9.7 42.2"8.7 ns
B 19.4"4.0 67.2"37.8 31.0"13.5 18.7"2.8 ns
U U U
UU UUUUU
Type effect ns ns
Number of steps L 34.9"2.5 35.5"6.0 30.1"8.6 35.0"8.2 ns
B 12.2"2.9 53.9"27.4 26.1"11.8 16.0"2.3 ns
U
Running s L 2.2"0.7 0.6"0.3 0.7"0.5 1.0"0.4 ns
B 0.7"0.3 1.1"0.1 1.1"0.9 0.3"0.2 ns
Type effect ns ns ns ns
Ž .
Standing s L 100.2"6.5 59.6"14.6 62.5"9.6 65.1"17.2 ns
B 48.8"15.1 49.3"17.4 46.2"19.8 30.0"5.7 ns
U
Drinking s L 7.2"2.9 10.5"3.6 11.8"2.3 13.1"3.6 ns
B 9.7"3.8 7.5"3.9 13.2"5.7 13.5"4.1 ns
Type effect ns ns ns ns
Ž . Ž .
Eating s L 180.1"42.8 116.6"21.0 44.0"6.1 104.5"34.2 0.05 B 70.0"23.1 158.7"44.3 78.7"46.8 46.2"11.6 ns
U
Preening s L 9.4"2.9 3.5"1.0 8.3"4.3 8.3"2.7 ns
B 5.9"2.9 4.1"2.4 2.0"0.7 3.5"1.1 ns
Type effect ns ns ns ns
Ž .
Exploring litter s L 16.8"3.5 28.7"9.8 30.8"12.6 24.6"10.5 ns
B 6.1"1.8 9.0"2.6 6.5"2.9 2.0"0.7 ns
U Number of L 5.3"0.8 3.5"1.5 1.7"0.5 4.7"1.0 interactions B 1.6"0.6 3.2"1.3 2.4"1.5 0.6"0.2 ns
U U U
UU UUUUUUUUUU
with congener Type effect ns ns
a
Ž .
were again significant at 3 weeks of age when B birds were twice as heavy as L birds. Eating duration was greater for L than B birds as early as 2–3 days of age. Preening did not differ between stocks. There were more interactions between L birds than between B birds at 2–3 and 20–21 days of age. L chicks spent more time exploring litter than B
Ž .
birds at least three times longer, Zs2.88, P-0.01, using Mann–Whitney’s U test .
3.4.2. Non-feeding bouts
Ž . Ž .
Duration Fig. 2b and composition of the non-feeding bouts data not shown were not affected by age or by genetic type. Only preening duration at 13–14 days of age
Ž
differed between the two crossbreeds L: 3.9"0.7 s, B: 0.8"0.2 s, Zs2.40, P-0.05
.
using Mann–Whitney’s U test .
3.4.3. Correlations between actiÕities
Only L activity during feeding bouts was related to body weight: at 3 weeks of age, L
Ž
chicks were less active while they were heavier walking: Rs y0.505, ns22,
.
P-0.05; standing: Rs y0.388, ns22, Ps0.07 . Time spent standing at days 2–3
Ž
was negatively related to body weight at day 22 in L birds Rs y0.639, ns15,
. P-0.05 .
During non-feeding bouts, there was a negative correlation between body weight at
Ž
hatching and duration of walking at 3 weeks of age L: Rs y0.486, P-0.05; B:
.
Rs y0.459, P-0.05, ns30 . The more active B birds at days 2–3 were the same at
Ž
20–21 days of age bout total duration: Rs0.412, P-0.05, ns27 ; standing:
. Rs0.409, P-0.05, ns27 .
4. Discussion
4.1. Low locomotor actiÕity
Before attempting to analyse the factors modifying locomotion in the present study, it must be emphasised that little time is allocated to locomotor behaviour in chicks, even when they are young. Chicks, especially broilers, were lying for about 70% of the time and they walked for only 5% of the time during the starting period. Our results were very close to those obtained in some commercial flocks, where broilers spent 65% of
Ž .
time lying Murphy and Preston, 1988 . This low level of activity could result from several environmental factors such as diet and the housing system. A good balance in
Ž .
the composition of the diet energy, protein content and proportions of amino acids will
Ž .
lead to a low level of activity Rovee-Collier et al., 1993 . Furthermore, the housing system usually used for broilers seems not enriched enough to stimulate a high level of
Ž .
exploratory activity Newberry, 1999 . The low level of activity, especially of locomo-tion, may also result from selection over several decades for higher growth rate and more efficient rate of feed conversion. This artificial selection has probably favoured
Ž
birds which were able to reduce subordinate energy expenditure walking, exploring
.
litter, running, etc. because productivity traits sometimes conflict with behavioural traits
Žreviewed by Muir and Craig, 1998 . The short duration of non-feeding bouts may.
reduction in locomotor behaviour is not complete and walking is not only restricted to indispensable transitory behaviour between two activities because the average distance travelled by broilers per day exceeds the minimum locomotion required to visit feeders
Ž .
and water dispensers Lewis and Hurnik, 1990 .
Study of locomotor activity is supposed to focus only on walking and running. Nevertheless, walking, and especially running, which is a very rare behavioural pattern, seem too restrictive if we wish to study all mechanical demands on the locomotor system. As a consequence, we must also take into account all the time spent standing without moving because the standing posture also induces stress on bones and joints
ŽAbourachid, 1993 ..
4.2. Significance of early locomotor behaÕiour
A detailed study of the first week of life was helpful to understand how locomotor behaviour develops. We found negative correlations between body weight at hatching and walking activity at 3 weeks of age in both genetic stocks while activity at 2–3 days of age was positively related to activity at 3 weeks of age in B chicks. Gordon and
Ž .
Tucker 1993 also found that individual birds that walked more at an early age were also more likely to walk more at a later age, even when they observed broilers only from 17 days of age. However, the greatest value of further investigations during the first days of life would be to determine whether there is a critical period beyond which it is not possible to change future locomotor behaviour of chicks. In this case, one solution to limit the occurrence of leg abnormalities in intensive rearing would be to stimulate activity during the starting period, when birds are not constricted by overcrowding on the floor or by obesity. An increase in activity only during this early period would limit deterioration of feed conversion, because at this early age body weight gain is limited. Moreover, action during this period would be particularly profitable since it is an
Ž .
essential phase for bone development Rose et al., 1996 .
4.3. InÕolÕement of genetic components
At 2–3 days of age, the duration of walking, standing and the number of steps in B birds were half of those of L chicks when feeding bouts were compared, suggesting a genetic stock effect. However, we emphasise that the duration of several behavioural patterns were reduced in B birds at this age and that the modifications in locomotor behaviour may be linked to changes in general activity. These differences have not been reported to date in other published experiments and need to be confirmed: the breeder’s characteristics, the actual date of birth andror the stimuli perceived during the first hours of life may have led to differences in behaviour and in body weight between the two genetic stocks. These early differences between genetic types cannot be explained by the effect of body weight per se because B chicks were lighter than L chicks at hatching and their body weight were similar to L chicks at 2–3 days. We cannot therefore reject the hypothesis of a genetic effect on locomotor behaviour for young
Ž
chicks. Many studies have detected a genetic behaviour component for birds reviewed
.
Ž . Ž
estimated in quails Saleh and Bessei, 1981 and in laying hens Jezierski and Bessei,
.
1978 .
The two genetic stocks had different organisation of feeding bouts. For both groups,
Ž
most of the activity when not resting was devoted to feeding duration of the feeding bouts represented three quarters of the non-lying time, as reported by Gordon and
.
Tucker, 1993 but L birds spent more time eating than B birds, whereas they grew twice
Ž
as slowly. This can be explained by the fact that many pecks are exploratory two pecks
.
in three according to Yo et al., 1997 . The number of effective pecks in relation to total number of pecks may vary with genetic origin: chickens selected for high body weight spent less time in front of the feeder but ingested more than chickens selected for lower
Ž .
body weight Barbato et al., 1980 .
There were also differences between both genetic stocks in response to changes in life conditions. For instance, L chicks explored the litter for three times longer than B birds during feeding bouts. Their better knowledge of the environment may make them more flexible when something changes in their living conditions. For example, when food appearance radically changed at day 7 from a mix of mash and pellets to only pellets, without any change of composition, B birds decreased time spent feeding and increased time spent lying, drinking and walking just after the change. In contrast, L birds steadily increased the time spent feeding and were apparently not disturbed by the
Ž .
food change. This is in agreement with Jones 1986 conclusion that light hybrid chickens are less reluctant to feed on unfamiliar food compared to medium hybrid chickens and that the difference can be related to enhanced ability to adapt to novelty. Our results suggest that using genetic variability could help to increase bird activity. Manipulating locomotor activity by genetic means needs further studies on
interindivid-Ž
ual variability of activity in the same genetic strain, as has been done in quails Saleh
.
and Bessei, 1981 . Another interesting use of the genetic differences would be to breed L chicks as ‘‘guides’’ for B birds: the great exploring and playing activities of L birds
Žexploring litter, numbers of interactions with a congener could be used to stimulate.
locomotor behaviour in broilers. This should be explored because change in feeding
Ž .
behaviour, for example, can be stimulated by intermingling species Savory, 1982 or
Ž .
breeds Mahagna et al., 1994 .
In conclusion, the differences in feeding-bout duration and composition between 2-day-old L and B chicks suggest that locomotor activity in young chicks is influenced by genetic effects, but further genetic studies of individual variability are needed to confirm this point. Manipulation of activity by genetic or environmental means during the first week of life may be one way of preventing leg disorders at slaughter if the relationship between early and late activity is confirmed by further experiments.
Acknowledgements
We are grateful to O. Callut for animal care, to C. Bouchot for assistance with video-recording and to D. Raine for revision of the English language of the manuscript. Funding was provided by the Ministere de l’Agriculture et de la Peche, Direction
`
ˆ
Ž .
References
Abourachid, A., 1993. Mechanics of standing in birds: functional explanation of lameness problems in giant turkeys. Br. Poult. Sci. 34, 887–898.
Barbato, G.F., Cherry, J.A., Siegel, P.B., Van Krey, H.P., 1980. Quantitative analysis of the feeding behavior of four populations of chickens. Physiol. Behav. 25, 885–891.
Bessei, W., 1992. Das Verhalten von Broilern unter intensiven Haltungsbedingungen. Arch. Geflugelkd. 56,¨
1–7.
Boa-Amponsem, K., Dunnington, E.A., Siegel, P.B., 1991. Genotype, feeding regimen, and diet interactions in meat chickens: 2. Feeding behavior. Poult. Sci. 70, 689–696.
Craig, J., Muir, W., 1998. Genetics and the behavior of chickens: welfare and productivity. In: Grandin, T.
ŽEd. , Genetics and the Behavior of Domestic Animals. Academic Press, San Diego, USA, pp. 265–297.. Ž .
Gordon, S.H., Tucker, S.A., 1993. Broiler walking behaviour. In: Savory, C.J., Hugues, B.O. Eds. , 4th European Symposium on Poultry Welfare, Universities Federation for Animal Welfare. p. 291.
Haye, U., Simons, P.C.M., 1978. Twisted legs in broilers. Br. Poult. Sci. 9, 549–557.
Hughes, B.O., Elson, H.A., 1977. The use of perches by broilers in floor pens. Br. Poult. Sci. 18, 715–722. Jezierski, T., Bessei, W., 1978. Der Einfluß Genotyp und Umvelt auf die locomotorische Aktivitat von¨
Legehennen in Kafigen. Arch. Geflugelk. 42, 159–166.¨ ¨
Jones, R.B., 1986. Responses of domestic chicks to novel food as a function of sex, strain and previous experience. Behav. Processes 12, 261–271.
Lanyon, L.E., 1992. Functional load-bearing as a controlling influence for fracture resistance in the skeleton.
Ž .
In: Whitehead, C.C. Ed. , Bone Biology and Skeletal Disorders in Poultry. Carfax Publishing Company, Abingdon, UK, pp. 61–66.
Lei, S., Van Beek, G., 1997. Influence of activity and dietary energy on broiler performance, carcass yield and sensory quality. Br. Poult. Sci. 38, 183–189.
Leterrier, C., Nys, Y., 1992. Clinical and anatomical differences in varus and valgus deformities of chick limb suggest different aetio-pathogenies. Avian Pathol. 21, 343–356.
Lewis, N.J., Hurnik, J.F., 1990. Locomotion of broiler chickens in floor pens. Poult. Sci. 69, 1087–1093. Mahagna, M., Nir, I., Nitsan, Z., 1994. Influence of the presence of 3-day-old-chickens on the behaviour of
meat and egg-type posthatch counterparts. Appl. Anim. Behav. Sci. 40, 143–152.
Martin, P., Bateson, P., 1986. Measuring behaviour. In: An Introductory Guide. Cambridge Univ. Press, Cambridge, UK, pp. 48–68.
Mench, J.A., 1988. The development of aggressive behavior in male broiler chicks: a comparison with laying-type males and the effects of feed restriction. Appl. Anim. Behav. Sci. 21, 233–242.
Muir, W.M., Craig, J.V., 1998. Improving animal well-being through genetic selection. Poult. Sci. 77, 1781–1788.
Murphy, L.B., Preston, A.P., 1988. Time-budgeting in meat chickens grown commercially. Br. Poult. Sci. 29, 571–580.
Newberry, R.C., 1999. Exploratory behaviour of young domestic fowl. Appl. Anim. Behav. Sci. 63, 311–321. Newberry, R.C., Blair, R., Gill, S., Knott, M., 1995. Health of broilers reared under two patterns of increasing
Ž .
light and with access to perches. Poult. Sci. 74 Suppl. 1 , 84.
Praytino, D.S., Phillips, C.J.C., Omed, H., 1997a. The effects of color of lighting on the behavior and production of meat type chickens. Poult. Sci. 76, 452–457.
Praytino, D.S., Phillips, C.J.C., Stokes, D.K., 1997b. The effects of color and intensity on behavior and leg disorders in broiler chickens. Poult. Sci. 76, 1674–1681.
Reiter, K., Bessei, W., 1996. Effect of the distance between feeder and drinker on behaviour and leg disorders of broilers. In: Proceedings of the 30th International Congress of Applied Ethology, Guelph, Canada. pp. 131–132.
Reiter, K., Bessei, W., 1997. Gait analysis in laying hens and broilers with and without leg disorders. Equine Vet. J. Suppl. 23, 110–112.
Reiter, K., Bessei, W., 1998. Einfluß des Laufaktivitat auf die Knochenentwicklung und Beinschaden bei¨ ¨
Broilern. Arch. Geflugelk. 62, 247–253.¨ ¨
Rodenhoff, G., Dammrich, K., 1971. Uber die Beeinflussung des Skeletes der Masthahnchen durch Haltung¨ ¨
Rogers, L.J., 1995. In: The development of brain and behaviour in the chicken. CAB International, Wallingford, UK, pp. 72–118.
Rose, N., Constantin, P., Leterrier, C., 1996. Sex differences in bone growth of broiler chickens. Growth Dev. Aging 60, 49–59.
Rovee-Collier, C., Collier, G., Egert, K., Jackson, D., 1993. Developmental consequences of diet and activity. Physiol. Behav. 53, 353–359.
Saleh, K., Bessei, W., 1981. Eine Untersuchung zur genetischen Veranlagung der Laufaktivitat von Wachteln.¨
Arch. Geflugelkd. 1, 23–26.¨
Savory, C.J., 1982. Effects of broiler companions on early performance of turkeys. Br. Poult. Sci. 23, 81–88. e
`me Simons, P.C.M., 1986. The incidence of leg problems in broilers as influenced by management. In: 7
Conference Europeenne d’Aviculture, 24–28´ ´ r08r1986, Paris, France. pp. 289–297, Vol. 1. Thorp, B.H., 1994. Skeletal disorders in the fowl: a review. Avian Pathol. 23, 203–236.
Weeks, C.A., Kestin, S.C., 1997. The effect of leg weakness on the behaviour of broilers. In: Koene, P.,
Ž .
Blockhuis, H.J Eds. , 5th European Symposium on Poultry Welfare, Wageningen, Nederlands, 7– 10r06r1997. pp. 117–118.
Yo, T., Vilarino, M., Faure, J.M., Picard, M., 1997. Feed pecking in young chickens: new techniques of˜
