Effects of a 16-hour light, 8-hour dark lighting schedule on behavioral traits and performance in male broiler chickens

Transcription

1 2010 Poultry Science Association, Inc. Effects of a 16-hour light, 8-hour dark lighting schedule on behavioral traits and performance in male broiler chickens A. Bayram and S. Özkan 1 Ege University, Faculty of Agriculture, Department of Animal Science, İzmir, 35100, Turkey Primary Audience: Flock Supervisors, Broiler Managers, Researchers, Extension Personnel SUMMARY The objective of this study was to examine the effects of a 16L:8D lighting schedule from d 2 of age through a 6-wk production trial on the performance and behavior of Cobb broiler males compared with a continuous 24-h light schedule (control). Behavioral tests included social reinstatement behavior measured in runway tests and fear response measured by tonic immobility. The 16L:8D group had BW and FCR similar to control birds at 6 wk. Based on behavior observations by scan sampling at 0900 to 1100 h and 1400 to 1600 h, the number of chicks eating, drinking, walking-standing, and pecking increased under the 16L:8D lighting schedule, whereas resting (sitting and sleeping) decreased (P 0.05). The 16L:8D group also exhibited comfort behaviors, such as preening and wing-shaking, more extensively than did control birds (P 0.05). Broilers in the 16L:8D group had a shorter duration of tonic immobility (P 0.05) at 11 and 35 d of age, which might represent less fearfulness. In runway tests at 13 and 30 d of age, the 16L:8D group exhibited shorter latencies to reach the zone of stimulus birds and spent more time near the stimulus birds (P 0.05). We therefore suggest that the birds in the 16L:8D group had a greater degree of sociality. Although we did not measure whole-day activity, we speculate that this greater sociality might have resulted from increased interactions among birds caused by increased activity during the observation periods. The16L:8D program tended to have more natural daytime behavioral patterns in broilers, to reduce fearfulness (and thus psychological stress), and to increase sociality. We conclude, based on the morning and afternoon measurements of this study, that birds in the 16L:8D group had a better welfare status than control birds without affecting 42-d broiler performance of the strain used. Key words: lighting, broiler performance, behavior, sociality, fearfulness 2010 J. Appl. Poult. Res. 19 : doi: /japr DESCRIPTION OF PROBLEM Constant genetic selection for growth and efficiency traits and improvements in nutrition and environmental conditions have resulted in today s fast-growing broilers [1]. Havenstein et al. [1] reported that 85 to 90% of the change in broiler growth rate from 1957 to 2001 was accounted for by genetic selection, whereas nutrition provided approximately 10 to 15% of the 1 Corresponding author: sezen.ozkan@ege.edu.tr

2 264 total change. However, fast growth attributable to both genetic selection and a high calorie intake has been associated with a higher incidence of ascites and skeletal deformities [2], which often reduces the performance and welfare of broilers. Lighting is one of the most important environmental factors affecting broiler performance and physical activity. Traditionally, broilers have been subjected to continuous or nearly continuous light to maximize growth and feed intake; however, shorter day lengths or alternative programs are now being considered because of welfare concerns and possible energy savings [3]. Photoperiodic regimens have the potential to reduce early growth and decrease susceptibility to metabolic diseases, such as pulmonary hypertension, sudden death syndrome, and tibial dyschondroplasia [4 6]. Therefore, photoperiodic lighting regimens, particularly those with an uninterrupted dark period, may have performance as well as welfare advantages, and may provide more natural cues for the synchronization of behavioral activities in broiler production compared with conventional continuous lighting [7 9]. Improvements in broiler health under longer periods of darkness have been attributed to a reduced early growth rate, which allows birds to become physiologically mature before maximal muscle accumulation in favor of increased leg strength [2, 4, 8], and to some functions in the body controlled by light and darkness, such as the diurnal rhythms of melatonin, which are involved in thermoregulation, feeding and digestion, and immune functions [10]. The fear reaction is an important component of welfare assessment because it can be indicative of overall stress [11]. Often, poor welfare is associated with chronic stress, which might coincide with fear and depression [12]. The duration of tonic immobility (TI) and the number of inductions needed per TI measurement are useful indicators of the fear reaction [11]. Sanotra et al. [8] suggested that light-dark schedules (16L:8D), either in increasing patterns or constantly from 2 d of age, may be useful in reducing chronic fear. Runway tests are commonly used to measure social reinstatement tendencies in young chickens and Japanese quail [13, 14]. This test incorporates exposure to a frightening event (a novel environment) and social stress (separation from the flock). Therefore, runway responses of either avoidance or affiliation reactions, measured as a tendency to stay close to accompanying birds, may show sensitivity to their sociality and fearfulness. However, the social reinstatement tendencies of broilers reared under a light-dark schedule, as opposed to continuous light, have not been studied. The purpose of this study was to use a constant light-dark schedule throughout the production period and to report the differences in behavior and performance compared with continuous lighting. Therefore, in this study, we investigated the effects of a constant cycle of 16 h of light and 8 h of darkness (16L:8D) beginning from d 2 compared with control (continuous) lighting consisting of 24 h of light (24L) from hatching to 42 d on growth performance and different forms of behavior, namely, sociality (as measured by a runway test) and the fear response (as measured by the TI reaction), in male broilers. Birds JAPR: Research Report MATERIALS AND METHODS One-day-old Cobb 500 male (feather-sexed) broiler chicks (n = 184 in total) obtained from a commercial hatchery (Köy-Tür Ege Entegre, İzmir, Turkey) were used in this experiment. The chicks were randomly distributed to replicate pens ( m), with a group size of 23, in 2 identical experimental rooms in the same house that varied by lighting program. Thus, each lighting treatment consisted of 4 replicate pens. After continuous 24-h lighting for the first 2 d, a 16L:8D lighting schedule was applied to chicks from d 2 onward in the treatment group. The lights were turned off between 2400 and 0800 h, and a dusk period of 30 min (approximately 4 to 5 lx) was provided. Daylight fluorescent lamps were used in the experiment, and average light intensity was 20 lx among pens at the bird level. Control birds were reared under continuous light (24L) over the course of the production period. The only environmental cue for the 24L birds was the human contact for daily care, which was similar to that of birds in the 16L:8D treatment.

3 Bayram and Özkan: Broiler lighting and behavior 265 Experimental rooms were equipped with ventilation and gas heating systems and were controlled by the same computer using similar set values. Standard brooding temperatures were applied to birds by gradually reducing the temperature from 32 to 25 C at the end of 3 wk of age. Thereafter, the temperature and humidity were between 22 and 25 C and 50 to 60%, respectively. The birds were fed a standard commercial starter diet in crumble form between 0 and 3 wk and a grower diet in pellet form between 3 and 6 wk. These diets were formulated to meet or exceed minimum NRC [15] standards for all ingredients. The starter diet contained 23.0% protein and 3,100 kcal/kg of ME from 0 to 3 wk, and 21.0% protein and 3,200 kcal/kg of ME from 3 to 6 wk. The animal care practices used in the experiment were in accordance with the principles of the Ege University Animal Research Ethics Committee and government law (No. 5199). Performance Measurements Body weight was individually measured at hatch, 3 wk, and 6 wk of age. Chicks were weighed to ±0.1 g at hatch and to 1 g thereafter. Daily BW gain was calculated for 0 to 3 wk and 3 to 6 wk. Feed consumption and the FCR were estimated for the entire period on a replicate pen basis (kg of feed/kg of BW gain). Mortality was recorded daily and the number of birds in the pens was adjusted from the point of death for feed consumption calculations. However, only 3 birds died during the experiment, 1 in the 16L:8D group and 2 in the 24L group; therefore, no statistical analyses were done for mortality. Behavior At 2, 4, and 6 wk of age, a scan sampling method at the pen level was used for behavioral data [16]. Observations were made in both the morning (0900 to 1100 h) and afternoon (1400 to 1600 h) sessions over 3 consecutive days of each week by a single observer. Behavior was assessed by recording the number of chicks involved in each predefined behavior at 1-min intervals during 10-min observation periods for each replicate pen. Therefore, a total of 60 min of data was available from each replicate pen during each observation week. In addition to 4 main behavioral activities, for which the scan sampling method gave accurate estimations, namely, feeding, drinking, walking-standing, and resting (sitting-lying), less frequently performed activities, namely, sleeping, pecking (inedible objects, litter, other birds), and comfort behaviors (vertical wing-shaking, wing-leg stretching, preening, and dust bathing), were also recorded whenever seen. Therefore, it was possible for the same bird to be counted for more than 1 behavior (e.g., 1 bird counted in the walking-standing category could also be counted as preening or pecking). TI Five birds from each replicate (20/treatment) were tested for TI responses at 11 and 35 d of age, with different birds used at each age. To induce immobility, the observer restrained the bird on its back in a U-shaped wooden cradle for 15 s. If the bird stayed immobile 10 s after the restraint was ended, the TI duration was recorded until the bird righted itself, with a 600-s test ceiling. Otherwise, the restraining stage of 15 s each was repeated, up to 5 induction trials. The TI duration was divided by the induction number to calculate the TI duration per induction unit (TI/I). Runway Tests For the reinstatement responses of broilers in runway tests of sociality, 5 birds from each replicate that were not used for TI measurement were tested at 13 and 30 d of age in a runway apparatus in a separate room (20 birds/treatment per age). During the runway test, 2 additional birds from the same pen were used as familiar stimulus birds in the goal box. At 13 d, the runway test apparatus described by Marin et al. [13], consisting of an unpainted wooden corridor measuring 2.0 m long 0.4 m wide 0.5 m high, was modified by increasing the width to 0.5 m. The corridor was divided into 3 compartments using removable wire-mesh partitions. The 0.2-m-long compartments at the opposite ends of the runway were used either as a start box or a goal box. Therefore, the actual runway was 1.60 m. The test chick was placed in the start box and left for an acclimatization period of 2 min, during which it could see the stimulus birds in the goal box. The wire mesh door

4 266 JAPR: Research Report of the start box was then raised and latencies to leave the start box (emerging) and to enter the stimulus bird zone (0.2-m zone next to the goal box) and total time spent in the stimulus bird zone were measured by an unobtrusive observer during the 10-min test. At 30 d of age, the total length of the apparatus was increased to 2.5 m; therefore, the actual runway was 2.0 m and the stimulus bird zone was 0.25 m [17]. Statistics One-way ANOVA was performed for continuous data that were normally distributed, namely, BW, BW gain, and FCR [18]. The TI duration per bird, number of inductions per bird, and TI/I were subjected to ANOVA, with lighting group effects transformed to the log scale. However, the actual means are presented in the tables. The runway test data were analyzed using nonparametric tests [19]. The logistic regression procedure (PROC GENMOD) was fitted to behavior data by generalized estimating equations using SAS statistical software [18, 20]. Statements of statistical significance are based on P RESULTS AND DISCUSSION Performance Parameters The BW, BW gain, and FCR of broilers reared under different lighting regimens, and at different ages, are shown in Table 1. Total feed consumption was 3.44 and 3.31 kg/bird for birds in the 16L:8D and control group, respectively, and there was no difference between groups. Birds in the 16L:8D photoperiod treatment weighed less than those in the 24L treatment at 3 wk of age (P 0.05) but were similar again by 6 wk of age. In this study, a constant photoperiod (16L:8D) beginning at 2 d of age reduced early growth, leading to a decrease in BW at 3 wk and in BW gain between 0 and 3 wk compared with broilers reared under the 24L photoperiod. The mean daily BW gain of broilers from 3 to 6 wk was 89.7 g/d in the 16L:8D group and 86.6 g/d in the 24L group. Body weights of broilers in the 16L:8D group reached those of broilers in the 24L group at slaughter age. Therefore, final BW and FCR were not affected by lighting treatment under the experimental conditions. These results are in accordance with many previous reports indicating that moderate day lengths result in acceptable performance levels, with similar final BW compared with continuous lighting [21 23]. A better FCR was one of the expected benefits of the short photoperiod, as has been reported previously [3, 7, 21, 22]. Increased melatonin secretion during the dark phases has been shown to coincide with a reduced energy expenditure for physical activity in broilers [10]. In contrast to these findings, feed consumption and FCR were not influenced by lighting in our study. However, reports that a 16L:8D lighting schedule resulted in similar FCR values as continuous lighting [23, 24] are in accordance with the findings in this experiment. Different environmental factors, such as feed wastage, temperature differences, and stocking density, may affect FCR values. In both lighting groups, iden- Table 1. Mean BW, BW gain, 1 and FCR of male broilers reared under 2 different lighting schedules 2 BW, g BW gain, g/d FCR, kg/kg Item Hatch 3 wk 6 wk 0 to 3 wk 3 to 6 wk 0 to 3 wk 3 to 6 wk Lighting 16L:8D b 2, b Control a 2, a SEM P-value Source of variation Lighting < < a,b Values with different superscripts within a column differ significantly (P 0.05). 1 n values for BW and BW gain means are between 90 and 91 for birds in the control and 16L:8D treatments, respectively, per age. 2 16L:8D = 16 h of light and 8 h of dark daily from 2 to 42 d; control = 24 h of light daily from 0 to 42 d.

5 Bayram and Özkan: Broiler lighting and behavior 267 Table 2. Probabilities from the generalized estimating equation analyses for the effect of lighting group, 1 age, and pen at each observation time (morning: 0900 to 1100 h; afternoon: 1400 to 1600 h) on the behavioral traits of male broilers Time of observation, P-value Morning Afternoon Behavior Lighting Week of age Pen Lighting Week of age Pen Eating Drinking Walking-standing < Resting < Sleeping < < Pecking < Preening Stretching Wing-shaking < Dust bathing L:8D = 16 h of light and 8 h of dark daily from 2 to 42 d; control = 24 h of light daily from 0 to 42 d. tical pen sizes were used from the first day of the experiment, with a stocking density of 14 birds/ m 2, which was lower than commercial conditions, where the density commonly ranges from 16 to 18 birds/m 2. In fact, greater activity, as was observed during the morning and afternoon behavioral scans, attributable to the 16L:8D program may have resulted in no beneficial effect of the 16L:8D program on the FCR in this study. However, this conclusion needs further support by increasing the scan sampling of behaviors to a greater proportion of the day. Behavior Table 2 contains the probabilities obtained from the generalized estimating equations based on the logistic regression procedure for behavioral traits during morning and afternoon observations. Based on odds ratio (OR) comparisons, drinking, walking-standing, pecking, preening, and wing-shaking behaviors tended to be higher under the 16L:8D photoschedule compared with the 24L photoschedule (P 0.05) at both observation times (Table 3). However, the 16L:8D program decreased (P 0.05) the tendency of broilers to perform resting and sleeping behaviors during both the morning and afternoon observations (Table 3). Because our behavioral observations did not encompass the whole day, we can only speculate that this difference is probably due to the broilers under the 16L:8D photoschedule sleeping for more time at night because this lighting program would have caused more synchronized behavior. The mean proportions of broilers performing different behaviors in the morning and afternoon observations are presented in Figure 1. The frequencies of feeding behavior were similar in both lighting groups in the afternoon, but broilers from the 16L:8D group exhibited increased feeding activity during the morning (Table 3 and Figure 1), which is in accordance with the diurnal rhythm of feeding activities in domestic fowl [25, 26]. Management techniques other than lighting (i.e., sequential feeding, diet composition [27], or their combination [28]) have been reported to improve leg health and increase activity. Increasing environmental complexity has also been suggested as an effective management tool to improve the activity of broilers [29]. Broilers in the 16L:8D group showed more stretching behavior than broilers in the 24L group during the afternoon observation; however, stretching behavior was observed less in the 16L:8D group in the morning as compared with the 24L group (Table 3). Dust-bathing behavior did not differ with lighting in either observation period. Broilers in the 16L:8D photoperiod treatment exhibited a greater tendency to peck during each observation time than did those reared in the control photoperiod. The pecking behavior records in our study included foraging, which is an activity-related behavior, and activity-relat-

6 268 ed behaviors have been shown to be related to lameness in broilers [30]. We did not evaluate the walking ability of broiler chickens in this experiment, but because we reported quite a small number of broilers with poor walking ability on d 42 in an adjoining study [31], we can speculate that lameness may have had only a small role in the behaviors. Recently, Brickett et al. [28] confirmed that reduced growth rate by light or nutrient modifications resulted in improved walking ability of broilers. The effect of age (wk) was found to vary with the behavior of broilers (Table 2). The 2-wk-old chicks generally had a greater incidence of eating, drinking, and walking-standing behaviors as compared with the 6-wk-old chicks during the morning (OR = 2.03, 1.38, and 1.80) and afternoon (OR = 2.90, 1.73, and 1.60) observation periods for eating, drinking, and walking-standing, respectively (data not shown in tables). This represents a decreasing frequency for these types of behaviors with an increase in age of the broilers. Sleeping (OR = 1.32) and pecking (OR = 1.16) also showed a decreasing pattern by age in the morning observations. During the afternoon observations, broilers showed less tendency to perform preening (OR = 1.38) and dust-bathing (OR = 2.06) behaviors with increasing age. Resting was the only behavioral pattern that increased in broiler chickens of ages 2 to 6 wk (OR = 0.91) during the afternoon observations. JAPR: Research Report Although our behavioral data are limited to observation periods in the late morning and midafternoon, these results are consistent with those in other reports for broilers with regard to changes in behavioral patterns by age [8, 32 34]. Nielsen et al. [32] reported a gradual increase in both the proportion of chickens feeding and the overall activity in broilers during the first 3 wk of life and a gradual decline thereafter. A decrease in the activity of broilers by age has been related to their increased BW [33]. A decrease in comfort behaviors such as pecking, stretching, wing-shaking, and dust bathing with an increase in age of the broiler might also be related to an increased stocking density [8] and the quality of the litter, which gets worse by age [34]. TI The mean durations of TI and number of inductions per bird, and the TI/I in broilers at 11 and 35 d of age for the different lighting schedules are presented in Table 4. Continuous lighting was associated with a longer duration of TI and TI/I compared with 16L:8D (P 0.05). The mean TI duration of birds in the 16L:8D treatment was 60% of that observed in control birds on d 11 and was 50% of that observed in control birds on d 35. For the TI/I measurements, the magnitude of the difference between lighting groups was larger in the 16L:8D group, being 48% of that observed in the control group at 11 d Table 3. Odds ratios (OR), CI, and probabilities for the behavioral traits of male broilers reared on a 16L:8D lighting schedule and using the control treatment as a reference group 1 Time of observation 2 Morning Afternoon Behavior OR 95% CI P-value OR 95% CI P-value Eating to 1.77 < to Drinking to 1.97 < to 1.80 < Walking-standing to 2.76 < to 1.80 < Resting to 0.78 < to 0.90 < Sleeping to 0.30 < to 0.47 < Pecking to 1.96 < to 1.61 < Preening to 1.40 < to 1.32 < Stretching to to Wing-shaking to 7.13 < to Dust bathing to to L:8D = 16 h of light and 8 h of dark daily from 2 to 42 d; control = 24 h of light daily from 0 to 42 d. 2 Morning: 0900 to 1100 h; afternoon: 1400 to 1600 h.

7 Bayram and Özkan: Broiler lighting and behavior 269 Figure 1. Mean values and error bars representing SEM of lighting groups for proportions of chicks performing different behaviors during the observation periods. A) Morning (0900 to 1100 h) observation periods; B) afternoon (1400 to 1600 h) observation periods. An asterisk (*) indicates that the tendency to perform a specific behavior differed significantly between lighting groups according to odds ratio comparisons (P 0.05). 116L:8D = 16 h of light and 8 h of dark daily from 2 to 42 d; control = 24 h of light daily from 0 to 42 d.

8 270 JAPR: Research Report Table 4. Effect of lighting treatment 1 on the duration of tonic immobility (TI) per bird and per induction (TI/I), and number of inductions (I) at d 11 and 35 2 d 11 d 35 Item TI, s TI/I, s I, no. TI, s TI/I, s I, no. Lighting 16L:8D b b b b 1.75 Control a a a a 1.50 SEM P-value Source of variation Lighting a,b Values with different superscripts within a column differ significantly (P 0.05). 1 16L:8D = 16 h of light and 8 h of dark daily from 2 to 42 d; control = 24 h of light daily from 0 to 42 d. 2 n = 20 for the 16L:8D and control treatments at both ages. of age and 43% of that in the control group at 35 d of age. Based on TI duration and TI/I, we suggest that the 16L:8D schedule reduced the fear state of broilers, in agreement with other reports for broilers [8, 35]. Because fear is considered an important component of stress [11] and stress is often negatively associated with welfare [12], low fear levels in broilers under the 16L:8D photoschedule may be related to a better welfare status. However, because the mean induction number required to achieve the TI reaction of birds did not differ significantly between lighting groups, we conclude that the overall susceptibility of birds was similar. Runway Test of Sociality In runway tests, the number of birds emerging from the start box was higher for birds in the 16L:8D than the 24L photoschedule (P 0.05). At 13 and 30 d of age, all birds from the 16L:8D room emerged from the start box. However, at 13 and 30 d of age, only 4 (20%) and 7 (35%) birds from the 24L group, respectively, emerged from the start box (Table 5). The greater proportion of broilers emerging from the start box in the 16L:8D group coincides with the shorter TI durations in this group and might be regarded as further evidence of less fear anxiety among broilers in the 16L:8D group. The level of fear is also known to affect exploratory and locomotor activities [36]. However, Bizeray et al. [37] demonstrated a poor relationship between TI and activity level in the home pen of broilers. Therefore, less fearfulness and a greater number of activity-related behaviors in broilers under the 16L:8D photoperiod, during the observation periods in the study, may provide additional evidence for the benefits of this program on broiler behaviors [8]. The latency to emerge from the start box did not differ between lighting groups (Table 6). Broilers reared under the 16L:8D Table 5. Effect of lighting treatment 1 on the percentages of birds emerging and not emerging from the start box during the runway test d 13 d 30 Item Emerging, % Not emerging, % Emerging, % Not emerging, % Lighting 16L:8D 100 (n = 20/20) (n = 0/20) 100 (n = 20/20) (n = 0/20) Control 20 (n = 4/20) 80 (n = 16/20) 35 (n = 7/20) 65 (n = 13/20) Statistic χ P-value < < L:8D = 16 h of light and 8 h of dark daily from 2 to 42 d; control = 24 h of light daily from 0 to 42 d.

9 Bayram and Özkan: Broiler lighting and behavior 271 Table 6. Means ± SE for latency to emerge from the start box (s), latency to enter the stimulus bird zone (s), and time spent in the stimulus bird zone (s) for broilers from 2 lighting groups 1 tested on the runway at d 13 and 30 Latency to emerge from start box, s Latency to enter stimulus bird zone, s Time spent in stimulus bird zone, s Item d 13 2 d 30 3 d 13 d 30 d 13 d 30 Lighting 16L:8D 57.4 ± ± ± ± ± ± 50.8 Control 79.5 ± ± ± ± ± ± 90.1 Statistic χ P-value L:8D = 16 h of light and 8 h of dark daily from 2 to 42 d; control = 24 h of light daily from 0 to 42 d. 2 n = 20 for the 16L:8D treatment; n = 4 for the control treatment. 3 n = 20 for the 16L:8D treatment; n = 7 for the control treatment. schedule had shorter latency values for reaching the stimulus bird zone and spent more time in the stimulus bird zone (P 0.05). We suggest, based on the runway test results, that the sociality of broilers, as reflected by the tendency to form social groups, was increased when reared under the 16L:8D photoperiod compared with the 24L photoperiod. Although our behavioral observations were based on 2 small portions of the photoperiod (late morning and midafternoon), we can speculate that an underlying factor in this increased sociality may partly be related to the heightened interactions among birds attributable to the increased intensity of activity-related behaviors, such as eating, drinking, walking-standing, pecking, preening, and wing-shaking, during the photoperiod. Social motivation (sociality) directly affects the behavioral and physiological responses of birds to social encounters [38], and increased sociality could be indicative of a positive mood among birds, which we can interpret as an improvement in welfare [12]. Marin et al. [13] suggested that the social reinstatement motivation becomes more pronounced in broilers if they had been exposed to a frightening event (i.e., 5-min mechanical restraint) before the runway test. However, Jones et al. [14] suggested that quail lines selected for a low stress response exhibited greater sociality, which represents a negative relationship between stress and sociality. In our study, a special effort was made before the runway tests for birds not to be exposed to any other stressful manipulations, such as additional noises or distractions, other than daily care actions by the same caretaker. Therefore, we can be more confident that the greater social reinstatement responses of broilers reared under the 16L:8D schedule reflect the effects of the lighting treatment. As a result, the 16L:8D photoschedule increased activity and comfort behaviors, and decreased resting and sleeping during the observations made in the morning and afternoon scan sampling times. Although we did not measure broiler behavior throughout the whole day, the greater number of activity behaviors (eating, drinking, walking-standing) and smaller number of resting and sleeping behaviors in the 16L:8D group were more apparent during the morning observations, and this could be representative of more natural daily rhythms. Behavioral synchrony may affect the stress response of birds and can give a good indication of bird welfare [39]. A reduced level of fear, as indicated by a shorter TI duration and increased sociality, also was evident in our runway test with broilers in the 16L:8D group. CONCLUSIONS AND APPLICATIONS 1. The 16L:8D lighting schedule from d 2 during a 6-wk production period significantly reduced BW and BW gain of male broilers in the first 3 wk. However, lightrestricted male broilers had compensated for the BW gain deficiency by marketing age at 6 wk, with similar BW, feed consumption, and FCR as control birds. 2. The 16L:8D lighting schedule led to more synchronized behavioral patterns in broilers in this group as compared

10 272 with control birds, namely, reduced fearfulness, and thus psychological stress, but a greater degree of sociality, which may have decreased their susceptibility to social stress. REFERENCES AND NOTES 1. Havenstein, G. B., P. R. Ferket, and M. A. Qureshi Growth, livability and feed conversion of 1957 versus 2001 broilers when fed representative 1957 and 2001 broiler diets. Poult. Sci. 82: Julian, R. J Rapid growth problems: Ascites and skeletal deformities in broilers. Poult. Sci. 77: Lewis, P., and T. Morris Lighting for broilers. Pages in Poultry Lighting: The Theory and Practice. Northcot, Hampshire, UK. 4. Classen, H. L., and C. L. Riddel Photoperiodic effects on performance and leg abnormalities in broiler chickens. Poult. Sci. 68: Renden, J. A., S. F. Bilgili, R. J. Lien, and S. A. Kincaid Live performance and yields of broilers provided various lighting schedules. Poult. Sci. 70: Renden, J. A., E. T. Moran Jr., and S. A. Kincaid Lighting programs for broilers that reduce leg problems without loss of performance or yields. Poult. Sci. 75: Gordon, S. H Effect of daylength and increasing daylength programs on broiler welfare and performance. World s Poult. Sci. J. 50: Sanotra, G. S., J. Damkjer Lund, and K. S. Vestergaard Influence of light-dark schedules and stocking density on behavior, risk of leg problems and occurrence of chronic fear in broilers. Br. Poult. Sci. 43: Prescott, N. B., H. H. Kristensen, and C. M. Wathes Light. Pages in Measuring and Auditing Broiler Welfare. C. Weeks, ed. Butterworths/CABI Publishing, Cambridge, MA. 10. Apeldoorn, E. J., J. W. Schrama, M. M. Mashaly, and H. K. Parmentier Effect of melatonin and lighting schedule on energy metabolism in broiler chickens. Poult. Sci. 78: Jones, R. B Fear and adaptability in poultry: Insights, implications and imperatives. World s Poult. Sci. J. 52: Veissier, I., and A. Boissy Stress and welfare: Two complementary concepts that are intrinsically related to the animal s point of view. Physiol. Behav. 92: Marin, R. H., P. Freytes, D. Guzman, and R. B. Jones Effects of an acute stressor on fear and on the social reinstatement responses of domestic chicks to cagemates and strangers. Appl. Anim. Behav. Sci. 71: Jones, R. B., R. H. Marin, D. Satterlee, and G. G. Cadd Sociality in Japanese quail (Coturnix japonica) genetically selected for contrasting adrenocortical responsiveness. Appl. Anim. Behav. Sci. 75: NRC Nutrient Requirements of Poultry. 9th ed. Natl. Acad. Sci. Washington, DC. 16. Lehner, P. N Sampling methods in behavior research. Poult. Sci. 71: JAPR: Research Report 17. Hocking, P. M., C. E. Channing, D. Waddington, and R. B. Jones Age-related changes in fear, sociality and pecking behaviors in two strains of laying hen. Br. Poult. Sci. 42: SAS Institute SAS User s Guide: Version 8 Edition. SAS Inst. Inc., Cary, NC. 19. The runway test data were handled in 2 steps. First, a binary coding of birds, either emerging from the test box or not, was done and these data were subjected to chi-squared contingency analysis to compare lighting group effects. The second step included data from birds that had emerged from the start box. Latency to emerge from the start box, latency to reach to the stimulus bird zone, and time spent at the stimulus bird zone were not normally distributed, even when subjected to transformation. Therefore, these data were analyzed by Wilcoxon Kruskal-Wallis tests. 20. In the primary analyses of behavior data, pen (replicates), age (observation weeks), and time of day of the observation (morning or afternoon) were described as repeated measurements. Because the effects of age (wk) and time of day on behavior varied, behavioral data from morning and afternoon measurements were analyzed separately with a model including the age (wk) and pen effect as repeated measurements. The lighting programs were separated using contrast statements and the calculated OR (Petrie, A., and P. Watson Statistics for Veterinary and Animal Science. Blackwell Science Inc., Malden, MA) at each observation time. The tendency to perform each form of behavior under the 16L:8D schedule was compared with continuous lighting (control) as a reference group. 21. Blair, R., R. C. Newberry, and E. E. Gardiner Effects of lighting pattern and dietary tryptophan supplementation on growth and mortality in broilers. Poult. Sci. 72: Olanrewaju, H. A., J. P. Thaxton, W. A. Dozier III, J. Purswell, W. B. Roush, and S. L. Branton A review of lighting programs for broiler production. Int. J. Poult. Sci. 4: Renden, J. A., S. F. Bilgili, and S. A. Kincaid Comparison of restricted and increasing light programs for male broiler performance and carcass yield. Poult. Sci. 72: Özkan, S., Ö. Altan, and S. Yalçın Effects of restricted lighting schedules on broiler performance. In Proc. XXI World s Poult. Congr., Montreal, Canada. Canadian Branch of the World s Poult. Sci. Assoc. [CD-ROM] 25. Xin, H., I. L. Berry, T. L. Barton, and G. T. Tabler Feeding and drinking patterns of broilers subjected to different feeding and lighting programs. J. Appl. Poult. Res. 2: Appleby, M. C., J. A. Mench, and B. O. Hughes Maintenance. Pages in Poultry Behavior and Welfare. CABI Publishing, Cambridge, MA. 27. Bizeray, D., C. Leterrier, P. Constantin, M. Picard, and J. M. Faure Sequential feeding can increase activity and improve gait score in meat type chickens. Poult. Sci. 81: Brickett, K. E., J. P. Dahiya, H. L. Classen, C. B. Annett, and S. Gomis The impact of nutrient density, feed form, and photoperiod on the walking ability and skeletal quality of broiler chickens. Poult. Sci. 86: Bizeray, D., I. Estevez, C. Leterrier, and J. M. Faure Influence of increased environmental complexity on leg condition, performance, and level of fearfulness in broilers. Poult. Sci. 81:

11 Bayram and Özkan: Broiler lighting and behavior Weeks, C. A., T. D. Danbury, H. C. Davies, P. Hunt, and S. C. Kestin The behavior of broiler chickens and its modification by lameness. Appl. Anim. Behav. Sci. 67: Özkan, S., S. Yalçın, Y. Akbaş, F. Kırkpınar, Y. Gevrekçi, and L. Türkmut Effects of short day (16L:8D) length on broilers: Some physiological and welfare indices. World s Poult. Sci. J. 62(Suppl.):584. (Abstr.) 32. Nielsen, B. L., M. Litherland, and F. Noddegaard Effects of qualitative and quantitative feed restriction on the activity of broiler chickens. Appl. Anim. Behav. Sci. 83: Bokkers, E., and P. Koene Behavior of fastand slow-growing broilers to 12 weeks of age and the physical consequences. Appl. Anim. Behav. Sci. 81: Shields, S. J., J. P. Garner, and J. A. Mench Effect of sand and wood-shavings bedding on the behavior of broiler chickens. Poult. Sci. 84: Zulkifli, I., A. Raseded, O. H. Syaadah, and M. T. C. Morma Daylength effects on stress and fear responses in broiler chickens. Asian-australas. J. Anim. Sci. 11: Jones, R. B The assessment of fear in adult laying hens: Correlational analysis of methods and measures. Br. Poult. Sci. 28: Bizeray, D., C. Leterrier, P. Constantin, G. Le Pape, and J. M. Faure Typology of activity bouts and effect of fearfulness on behavior in meat type chickens. Behav. Processes 58: Jones, R. B., and A. D. Mills Divergent selection for social reinstatement behavior in Japanese quail: Effects on sociality and social discrimination. Avian Poult. Biol. Rev. 10: Duncan, I. J. H Behavior and behavioral needs. Poult. Sci. 77: Acknowledgments The presented results were a part of the master of science study of A. Bayram and were funded by the Scientific Research Projects Committee at Ege University (2005-ZRF- 035).