Effect of Sight Barriers in Pens of Breeding Ring-Necked Pheasants (Phasianus Colchicus): I

Effect of sight barriers in pens of breeding ring-necked pheasants (Phasianus colchicus): I. Behaviour and welfare Charles Deeming, Jonathan Cooper, Holly Hodges

To cite this version:

Charles Deeming, Jonathan Cooper, Holly Hodges. Effect of sight barriers in pens of breeding ring- necked pheasants (Phasianus colchicus): I. Behaviour and welfare. British Poultry Science, Taylor & Francis, 2011, 52 (04), pp.403-414. 10.1080/00071668.2011.590796. hal-00732523

HAL Id: hal-00732523 https://hal.archives-ouvertes.fr/hal-00732523 Submitted on 15 Sep 2012

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. British Poultry Science

For Peer Review Only

Effect of sight barriers in pens of breeding ring-necked pheasants (Phasianus colchicus): I. Behaviour and welfare

Journal: British Poultry Science

Manuscript ID: CBPS-2010-256.R1

Manuscript Type: Original Manuscript

Date Submitted by the 02-Dec-2010 Author:

Complete List of Authors: Deeming, Charles; University of Lincoln, Biological Sciences cooper, jonathan; University of Lincoln, Biological Sciences Hodges, Holly; University of Lincoln, Biological Sciences

Keywords: Pheasant, Sight barriers, Behaviour, Welfare

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 1 of 36 British Poultry Science

Edited Hocking 1 1 29/04/2011 2 3 Effect of sight barriers in pens of breeding ring-necked pheasants (Phasianus 4 5 colchicus ): I. Behaviour and welfare 6 7 8 D. C. DEEMING, H. R. HODGES and J. J. COOPER 9 10 Department of Biological Sciences, University of Lincoln, Lincoln, UK. 11 12 13 14 Correspondence: D.C. Deeming, Department of Biological Sciences, University of Lincoln, 15 16 Riseholme Park,For Lincoln, PeerLN2 2LG, UK. Review E-mail: [email protected] Only 17 18 19 Running head: Sight barriers and behaviour 20 21 22 23 24 25 26 27 Accepted for publica ti on : 15 January 2011 28 29 30 31 32

33 34 35 36 37

38 39 Bibliography: 40 41 Delete Deeming and Wadland 2001 42 43 Taber should be 1949, not 1959 (Taber, not Tabor again) 44 Add: Cooper, J. & Hodges, H. (2010). Effect of tree cover on ranging behaviour of free range hens. 45 46 Proceedings of the 44th Congress of the International Society for Applied Ethology - Coping in large 47 48 groups. Lidfors, L., Blokhuis, H. and Keeling, L. (eds) Wageningen Academic Publishers, Holland, p.75 49

50 51 52

53 54 55

56 57 58

59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 2 of 36

2 1 2 3 4 5 6 7 8 9 10 11 12 Correspondence: D.C. Deeming, Department of Biological Sciences, University of Lincoln, 13 14 Riseholme Park, Lincoln, LN2 2LG, UK. E-mail: [email protected] 15 16 For Peer Review Only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 3 of 36 British Poultry Science

3 1 2 3 Abstract 1. The study investigated the effects of providing sight barriers in breeding pens on 4 5 pheasant mortality, feather damage and behaviour. 6 7 2. Data were collected from 11 conventional pens (control) and 11 pens with additional sight 8 9 barriers (barrier) over the course of a ten week breeding season. Each pen contained 8 males 10 11 and 56 females at the beginning of the season. 12 3. There was a higher rate of mortality in males (6.25%) than females (2.11%) that was 13 14 unaffected by treatment. 15 16 4. Feather damageFor increased Peer over the breedingReview season and bothOnly male and female pheasants 17 18 showed better feather condition in the pens with barrier at the end of the season. 19 5. The pheasants spent most of their time walking or standing. Providing barriers increased 20 21 perching, but reduced preening. 22 23 6. The provision of sight barriers had no effect on the incidence of courtship and mating, but 24 25 did reduce aggressive interactions such as pecking and chasing. 26 7. Thestudy provides baseline data on the behaviour of breeding pheasants under these 27 28 husbandry conditions, and suggests that barriers may improve pheasant welfare by reducing 29 30 potentially harmful aggressive interactions, without affecting activity patterns or reproductive 31 32 behaviour. 33 34 35 INTRODUCTION 36 37 38 39 The pheasant production and shooting industries provide the equivalent of 70,000 full time 40 41 jobs in the UK and generates £1.6 billion (PACEC, 2006). The industry is also important in 42 43 44 managing and conserving rural areas, due to the scrubland and copses typically used for 45 46 pheasant release (see Burger, 1966; Grahn, 1993). Pheasant rearing may involve up to 35 47 48 million birds being released into the British countryside every year (Farm Animal Welfare 49 50 51 Council, 2008). However, the conditions that pheasants experience in captive environments 52 53 (e.g . Pennycott, 2000) prior to release will differ significantly ( e.g . group size and availability 54 55 of cover) from that of their natural situation ( e.g. Taber 1949). Despite the large number of 56 57 58 birds used to produce stock for release into the countryside, there is relatively little research 59 60 directed to exploring the behavioural requirements of breeding and growing pheasants in

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 4 of 36

4 1 2 3 captivity and how this impacts on their welfare. By contrast, there has been considerable 4 5 6 interest in the breeding behaviour of wild ring-necked pheasants in terms of the function of 7 8 sexual displays ( e.g . Collias and Taber, 1951; Schenkel, 1956, 1958; Heinz and Gysel, 1970) 9 10 11 and more recently as a model species to test theories of sexual selection ( e.g. von Schantz et 12 13 al ., 1989; Mateos, 1998). 14 15 16 For Peer Review Only 17 18 There may be several welfare issues associated with current methods of pheasant 19 20 husbandry in general and breeding stock in particular. Birds are typically kept in semi- 21 22 intensive conditions in pens open to the elements with larger numbers of con-specifics at 23 24 25 higher stocking densities than would occur naturally (Ridley and Hill, 1987; Grahn et al ., 26 27 1993a; Leif, 2005). They are therefore exposed to extremes of weather and potential 28 29 infectious agents from wild birds and their droppings (Pennycott, 2000). There is 30 31 32 considerable variation in the management of these birds particularly during breeding (review 33 34 by Deeming, 2009) – group sizes vary from small (7 or 8 hens with a single cock) through to 35 36 37 hundreds of birds in a single pen (~300 birds, e.g. Deeming and Wadland, 2002). The 38 39 competence of stockmen may also vary from farm to farm, with potential costs for the birds’ 40 41 health and welfare (FAWC, 2008). 42 43 44 During the breeding season wild ring-necked pheasants are territorial, guarding harems 45 46 of females from potential rivals and engaging in threats, displays, such as calling and wing- 47 48 flapping, and aggressive acts, such as chasing, sparring and pecking (Taber, 1949; Ridley and 49 50 51 Hill, 1987; Mateos, 1998). Non-territorial males are also reported who do not display or fight 52 53 but nevertheless may mate with unaccompanied females (Burger, 1966). At the end of the 54 55 breeding season males cease territorial behaviour and mate guarding, and become more 56 57 58 tolerant of other males. However, in enclosed multi-male pens it may not be possible 59 60 establish stable harems or territories, leading to more male-male intimidation and/or fighting,

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 5 of 36 British Poultry Science

5 1 2 3 and greater disturbance of courtship between males and females. 4 5 6 Increased stocking density can lead to other undesirable responses including more 7 8 aggressive pecking and feather pecking. Aggressive pecking in pheasants is generally 9 10 11 directed at the head region and is of short duration (Hoffmeyer, 1966). Aggressive pecking 12 13 can be involved in the establishment of hierarchies and/or be associated with competition for 14 15 resources and can be elevated at higher stocking densities due to increased competition for 16 For Peer Review Only 17 18 resources, reduced inter-bird distances and difficulties in establishing stable hierarchies 19 20 (Hoffmeyer, 1966). By contrast, feather pecking involves gentler pecks to any area of the 21 22 body such as back, chest and wings, and can involve plucking and ingestion of feathers 23 24 25 (Hoffmeyer, 1966). The behaviour appears to be a form of redirected foraging behaviour 26 27 because providing foraging substrates reduces feather damage and deficiencies in specific 28 29 nutrients may increase the behaviour (Hoffmeyer, 1966). Increasing stocking density 30 31 32 increases feather pecking damage (Hoffmeyer, 1966; Cain et al., 1984; Kjaer, 2004) by 33 34 reducing opportunity to interact with natural forage substrates, whilst increasing the 35 36 37 opportunity to peck feathers due to smaller inter-bird distances. 38 39 Management practises, such as the fitting of “bits” and “specs”, may reduce the 40 41 incidence and/or effectiveness of both aggressive pecking and feather pecking and thus 42 43 44 improve overall condition of stock. The effectiveness of bits has recently been investigated in 45 46 young pheasants reared for release, but their impact on welfare proved equivocal (Butler and 47 48 Davis, 2010). Bits halved acts of bird-on-bird pecking, improved plumage condition and 49 50 51 damage to skin but increased headshaking, scratching, nostril inflammation and bill 52 53 deformities suggesting “bitting” can adversely affect the beak of individual birds. 54 55 The current study investigated the effect of sight barriers in pens of breeding pheasants. 56 57 58 Normal practise on many game farms is to use pens with little cover or refuges that could 59 60 allow birds to engage in undisturbed courtship or to escape unwanted agonistic attention from

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 6 of 36

6 1 2 3 other birds. Sight barriers have been used in large (50 x 25 m) pens for breeding pheasants 4 5 6 (Deeming and Wadland, 2002) but their impact on behaviour was not investigated. Visual 7 8 barriers have significant positive effects on the behaviour of the birds in pen housed laying 9 10 11 hens (Newberry and Shackleton, 1997) and in broiler chicken houses (Cornetto and Estevez, 12 13 2001). They have also been found to have positive benefits in reproductive outputs of broiler 14 15 breeders (Leone and Estevez 2008). We hypothesised that visual barriers would allow 16 For Peer Review Only 17 18 pheasants to seek refuge from aggressive interactions with pen-mates and have a positive 19 20 impact on bird mortality and feather damage. The presence of barriers might also provide a 21 22 refuge for courting birds and increase rates of effective copulation. The focus on measures of 23 24 25 reproductive performance as well as on behaviour and welfare was considered important 26 27 because should sight barriers prove beneficial for bird welfare, then adoption by the game 28 29 industry would be more likely if a positive impact on breeding performance could also be 30 31 32 demonstrated. 33 34 We carried out this study at a commercial game farm during the ten week breeding 35 36 37 season in spring 2010. Pens with sight barriers were established on a game farm with the 38 39 open pens typically used by the farmers acting as a control. We tested the hypothesis that 40 41 visual barriers would reduce aggressive behaviours and have positive impacts on bird welfare 42 43 44 as measured in terms of mortality, plumage scores and behaviour. In addition to behavioural 45 46 observations, data for egg production and quality, true fertility, embryonic mortality and 47 48 hatchability were also collected. These results are reported in a companion paper (Deeming et 49 50 51 al ., 2011). 52 53 54 55 MATERIALS AND METHODS 56 Experimental design 57 58 59 The trial was conducted at a commercial game farm near Newark, Nottinghamshire, UK. 60 Twenty-four pens were established to provide 12 control pens, and 12 experimental pens

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 7 of 36 British Poultry Science

7 1 2 3 that contained sight barriers (Figure 1). There were four rows of six pens and the barrier and 4 5 6 control pens alternated in each row. Early in the study birds “dug” a hole between one 7 8 control and one trial pen due to dust-bathing in adjacent locations. As a consequence birds 9 10 11 were able to mix between these two pens, which were therefore removed from study. 12 13 Subsequently, other enclosures were monitored during the study and any holes near pen 14 15 walls were blocked. 16 For Peer Review Only 17 18 Each pen measured 13.2 m x 13.2 m (174.3 m²) and initially contained 56 female and 19 20 8 male birds (2.73 m² per bird). All birds had been hatched in June 2009 and were in their 21 22 first breeding season. All pens comprised slatted wooden sides up to approximately 60 cm, 23 24 25 and wire mesh above this point to a total height of 1.6 m, and were roofed to a maximum of 26 27 2.5 m with soft nylon mesh. The floors were soil and grass, with sawdust introduced 28 29 periodically around the edges of the pens when it was particularly wet weather. All pens 30 31 32 held 6 feed hoppers, an open drinker-bowl, and a grit bowl, as well as 5 or 6 metal half 33 34 barrels to provide sheltered nesting sites (Figure 1). Commercially available pelleted feed 35 36 37 and water were available ad libitum , and medicated feed as a standard worming treatment 38 39 was used once a month. Prior to the study all birds had plastic spectacles fitted, as well as 40 41 brails on the wings to prevent birds flying around within the pen, and cock birds had their 42 43 44 spurs trimmed. 45 46 Control “open” pens used the typical arrangement of commercial pens at the farm 47 48 and only had nesting sites to obscure the view of the birds (Figure 1). Trial “barrier” pens 49 50 51 had ‘sight barriers’ to a height of 60 cm in the form of tin sheeting and straw bales placed in 52 53 the pen in the form of an elongated X (Figure 1), but were otherwise identical to control 54 55 pens. It was estimated that birds in the barrier pens could only see a maximum of a third of 56 57 58 the area of the pen at any time. 59 60 The experimental design was approved by the Ethics Committee of the Faculty of

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 8 of 36

8 1 2 3 Agriculture, Food and Animal Sciences, University of Lincoln. Pilot studies were conducted 4 5 6 prior to both the collection of behavioural data by scan sampling and the continuous 7 8 behavioural sampling of social interactions. These studies were used to assess the viability 9 10 11 of direct observation, as there were specific concerns that the presence of observers may 12 13 influence behaviour ( e.g. increasing vigilance and locomotor behaviour) and that the 14 15 presence of barriers may interfere with sampling from trial pens as more birds would be out 16 For Peer Review Only 17 18 of view. To test for observer influence all pens were each videoed for an hour, and there was 19 20 no difference in the behaviour of pheasants compared with direct observation. With regard 21 22 to barriers obscuring views of pheasants, the 60 cm high bales were low enough for a 23 24 25 standing observer to be able to account for the behaviour of all birds not inside the nest sites. 26 27 28 29 Measures of physical welfare 30 31 32 Bird welfare was assessed by recording bird mortality and plumage damage. The total 33 34 number of dead or terminally ill birds (which were subsequently humanely culled) that were 35 36 37 removed from each pen type was recorded for the course of the 10-week laying season. The 38 39 plumage of the birds in each pen was scored as an indicator of bird condition ( e.g. 40 41 Hoffmeyer 1966; Bright et al ., 2006) at the middle of the trial and at the end of the breeding 42 43 44 season. Plumage damage was scored by visual observation of selected animals from outside 45 46 the pen. Six males and six females per pen were scored, noting the amount of feathering on 47 48 their rump, chest, neck/head, wings and tail. The scoring system was carried out on a 0-4 49 50 51 number basis, with 0 being ‘best’ or most feathering, and 4 being ‘worst’ or least. The slight 52 53 differences in categorisation for scoring between the tail and the main body are described in 54 55 Table 1. All plumage scores were carried by a single observer (HH), who had previous 56 57 58 experience of feather scoring from studies of free ranging layer hens (Cooper and Hodges, 59 60 2010). Plumage damage score was analysed after summation of the scores for five

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 9 of 36 British Poultry Science

9 1 2 3 components so that the best possible score was 0 and the worst was 20. 4 5 6 7 8 Behavioural data 9 10 11 General Activity 12 13 Data collected during a pilot study prior to the breeding season were used to ascertain which 14 15 behaviours were most prevalent (Table 2). Data presented here were for subsequent samples 16 For Peer Review Only 17 18 recorded during weeks 1-2, weeks 5-6 and weeks 7-9 of the laying season. The extended 19 20 time periods for recording each set of observations were due to the time involved in taking 21 22 observations for the number of pens involved in the study. Observations typically began at 23 24 25 10.00h and finished no later than 16.00h, with 2 days necessary to observe each pen. Data 26 27 were collected from each pen along a row, before moving to the next row. 28 29 The observer stood quietly within 2 m of the side wall of the pen, and after a period of 30 31 32 2 minutes (which was sufficient time for birds to settle after any disturbances related to the 33 34 approach of the observer) and counting the number of birds engaged in each behaviour by 35 36 37 scanning the pen in a clockwise motion for every 2 minutes 20 minutes. The behaviour of 38 39 males and females was recorded separately. Data were expressed as a proportion of the total 40 41 number of birds that were recorded during each scan. 42 43 44 45 46 Social behaviour 47 48 Observations of social behaviours were made during three time observation periods during 49 50 51 weeks 3-5, 6-7 and 9-10. The observer stood quietly by the pen for 20 minutes, and 52 53 observed the pen continuously. If a social interaction occurred, the behaviour and the sexes 54 55 of the participants ( e.g . male-male chase, female-male peck) were noted. In addition to 56 57 58 recording the incidence of bouts of pecking, it was also possible to record the number of 59 60 individual pecks. For analysis, records of activities were collated into the following classes

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 10 of 36

10 1 2 3 of social behaviour: aggression, display, courtship, avoidance and mating (see Table 3 for 4 5 6 more details). 7 8 9 10 11 12 13 Statistical analysis 14 15 Statistical analyses were carried out using PSAW (SPSS Inc., version 17.0). The effect of 16 For Peer Review Only 17 18 the presence or absence of barriers on bird mortality and the plumage scores were tested 19 20 using Mann-Whitney U-tests. For behaviour data collected over the laying season a mean 21 22 value was determined for each pen that then was averaged to produce illustrative graphs but 23 24 25 statistical analyses used all of the raw data (one value per week per pen) using repeated 26 27 measures analysis of variance. Time budget data were expressed as proportions and arc sin 28 29 transformed before analysis (Fowler et al ., 1998). The variance of social behaviours was 30 31 32 typically greater than the means and so counts for social behaviours were log 10 transformed 33 34 prior to analysis (Fowler et al ., 1998). The effectiveness of this calculation was tested using 35 36 37 Kolmogorov-Smirnov tests. The effect of the pen type, and where appropriate gender, on all 38 39 variables was tested using repeated measures analysis of variance (ANOVA) using the 40 41 Greenhouse-Geisser test for sphericity. Means and standard errors are presented for 42 43 44 untransformed data but analyses were performed on transformed values. 45 46 47 48 RESULTS 49 50 51 Welfare data 52 53 Mortality 54 55 For the week prior to the start of the laying season and during the course of the 10 week 56 57 58 laying period more birds were removed from barrier pens than open pens (22 and 15 birds, 59 60 respectively) but this was not statistically significant (Mann Whitney: U = 45.5, N1 = 11, N 2

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 11 of 36 British Poultry Science

11 1 2 3 = 11, P > 0.05). Eleven out of 176 (or 6.25%) males were dead or culled during the study 4 5 6 compared with 26 out of 1,232 females (2.11%), suggesting that males were more likely to 7 8 be lost during study period ( χ² = 10.31, df = 1, P < 0.01), though with such small numbers 9 10 11 lost, there was little impact on the average sex ratio over the course of the study. The 12 13 numbers of birds removed were evenly spread over the time course of the study. 14 15 16 For Peer Review Only 17 18 Plumage damage score 19 20 Male birds had worse mean plumage damage scores than females at both the mid- and end 21 22 season in both types of pen (Figure 2). At the mid-season scoring, females in open pens had 23 24 25 significantly worse plumage damage scores than females in the barrier pens (Mann 26 27 Whitney: U = 20.0, N1 = 11, N 2 = 11, P < 0.01) whereas males exhibited no difference 28 29 (Figure 2). By the end of the laying season both male and female birds had significantly 30 31 32 worse plumage damage scores in the open pens than birds in the barrier pens (U = 18.0, N1 33 34 = 11, N 2 = 11, P < 0.01; and U = 20.5, N1 = 11, N 2 = 11, P < 0.01, respectively). 35 36 37 Comparisons of each gender for each pen type between the mid and late times of assessment 38 39 showed that plumage damage condition score became worse in all cases (Mann Whitney 40 41 tests, P < 0.001). In males this was associated with significant ( P < 0.05) increases in scores 42 43 44 for the neck and rump for barrier pens and for neck, rump and chest for males in the open 45 46 pens. Plumage damage scores increased significantly ( P < 0.05) for all parts of the female 47 48 birds except wings in barrier pens and wings and chest in open pens. 49 50 51 52 53 General behaviour 54 55 Mean time budgets of male and female pheasants averaged over the three time periods of 56 57 58 observation by type of pen are shown in Figure 3. Repeated measures ANOVA of all data 59 60 did not reveal any significant effect of either gender or pen type on the percentages of scans

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 12 of 36

12 1 2 3 for walking, standing or sitting (Table 4). Significant gender effects were observed for 4 5 6 foraging and feeding with females more likely to be observed foraging or feeding but there 7 8 was no effect of the presence or absence of sight barriers (Table 4; Figure 3). Females were 9 10 11 also more likely to perch than males, and perching was more common in the barrier pens as 12 13 the straw bales that made up the sight barriers provided greater opportunities to perch 14 15 (Figure 3). Preening was more commonly seen in the open pens but was not affected by 16 For Peer Review Only 17 18 gender (Table 4). Males performed significantly more “other” behaviours but there was no 19 20 effect of pen type (Table 4). Mating was rarely recorded during scan sampling and was not 21 22 significantly affected by either gender or pen type (Table 4). 23 24 25 26 27 Social behaviours including aggressive and reproductive interactions 28 29 The mean values for the frequency of different behavioural categories are shown in Figure 30 31 32 4. There was a significant effect of the presence of the barriers on the mean numbers of 33 34 aggression behaviours observed with mean values for open pens being two times higher than 35 36 37 in barrier pens (repeated measures ANOVA: F 1,20 = 16.78, P < 0.001). Aggression was 38 39 categorised as chasing (male-male, male-female and female-female) and fighting (between 40 41 males) but around 75% of observations were of pecking. Discrete bouts of pecking by birds 42 43 44 on other birds were significantly less common in the barrier pens than in the open pens 45 46 (barrier pens: 3.4 ± 0.6; open pens: 8.2 ± 0.8; Repeated measures ANOVA: F1,20 = 13.44, P 47 48 < 0.001). The number of pecks delivered per bird in the pen (Figure 5) was higher in open 49 50 51 pens than barrier pens (F 1,40 = 13.44, P < 0.001), but the number of pecks did not differ 52 53 between the two sexes (F 1,40 = 1.63, P > 0.05) and there was no interaction between sex and 54 55 pen (F = 1.48, P > 0.05). 56 1,40 57 58 The presence of the barriers had no effect on the numbers of observed courtship 59 60 behaviours (Figure 4; repeated measures ANOVA: F 1,20 = 2.09, P > 0.05), or the number of

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 13 of 36 British Poultry Science

13 1 2 3 male display behaviours (repeated measures ANOVA: F = 2.18, P > 0.05). Neither 4 1,20 5 6 avoidance or mating behaviours were affected by the presence of the barriers (Figure 4; 7 8 repeated measures ANOVA: F 1,20 = 0.11, P > 0.05; and F 1,20 = 3.50, P > 0.05, respectively). 9 10 11 As these behaviours were associated with one gender or were performed equally by both 12 13 sexes ( e.g . mating) the effect of gender was not investigated. 14 15 16 For Peer Review Only 17 18 DISCUSSION 19 20 Although the behaviour of ring-necked pheasants during the breeding season has been 21 22 extensively studied by ecologists and behavioural ecologists, there have been comparatively 23 24 25 few studies of behaviour and social interactions of pheasants reared under commercial 26 27 conditions and these have tended to focus on the behaviour of growing pheasants and not the 28 29 breeding stock ( e.g . Hoffmeyer, 1966; Cain et al ., 1994; Kjaer, 2004; Butler and Davis, 30 31 32 2010). This presents challenges and opportunities for the discussion of the findings of this 33 34 study, as there is little by way of directly comparable data. The results will therefore be 35 36 37 discussed largely with respect to the larger body of work on the sexual behaviour of wild 38 39 pheasants, and the behaviour and welfare of growing pheasants. As a consequence it is 40 41 difficult to determine of our findings are typical of breeding flocks as data are only available 42 43 44 for production traits and crude measures of welfare, such as mortality. For these reasons we 45 46 shall cautiously speculate on the underlying reasons for any differences between the two 47 48 treatments as well as highlight other findings of note, and suggest potential areas for further 49 50 51 study. 52 53 Mortality during the study period was comparable to industry standards. Over the ten 54 55 weeks of study (and two weeks piloting prior to breeding season) 37 birds out of 1,408 died 56 57 58 or were culled, equivalent to 2.62% of the flock. Wise (1993) suggests that in well managed 59 60 flocks a mortality of around 5% should be expected over the entire breeding season which

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 14 of 36

14 1 2 3 lasts up to 20 weeks including establishment of flocks (4-6 weeks prior to the reproductive 4 5 6 period), and a few weeks prior to depopulation. Pennycott (2000) reported a mortality over 7 8 three seasons of about 3 % in a similar system of field pens with slightly larger groups of 8- 9 10 11 9 males and 60-62 females but in much larger pens at a stocking density of 4.6 m² per bird. 12 13 We did not establish the cause of death in our study, but approximately a quarter of 14 15 the birds examined at post-mortem by Pennycott (2000) had died as a result of trauma or 16 For Peer Review Only 17 18 injuries inflicted by other birds. Death by injury is more common in males than females, 19 20 largely due to head-pecking, whilst female injuries tended to be associated with treading 21 22 (during repeated mating) leading to feather loss, skin damage and cannibalism. Wise (1993) 23 24 25 suggests that without good management traumatic injuries can lead to the loss of up to 20% 26 27 of a breeding flock. It is noteworthy that mortality in wild populations can be higher still 28 29 over similar periods. For example, Grahn (1993) estimated that about a quarter of males and 30 31 32 females died in a free ranging population during the breeding season, with male mortality 33 34 tending to peak early in the season, possibly associated with the costs of mating or 35 36 37 defending territories, whilst females tended to die later in the season. In most cases dead 38 39 birds were either killed (or scavenged) by predators though other causes, such as disease or 40 41 injury, may have made them easier prey. In our flocks the trimming of males’ spurs and the 42 43 44 use of spectacles (Butler and Davis, 2010) may have mitigated the injuries through fighting, 45 46 feather pecking and mating in both open and barrier pens, and predators were excluded from 47 48 pens. 49 50 51 Plumage damage and plumage condition scores have been used extensively in studies 52 53 of hen welfare ( e.g . Webster and Hurnik, 1990; Bilcik and Keeling, 1999; Kjaer et al., 2001) 54 55 and variations of these measures have been used for pheasant chicks (e.g. Kjaer, 2004). 56 57 58 These detailed approaches rely on observers being in close proximity to birds for visual 59 60 inspection or even handling birds, which was not practical in this study due to the potential

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 15 of 36 British Poultry Science

15 1 2 3 effects of disturbance on behaviour and reproduction. Consequently we used a feather 4 5 6 damage scoring system based on those used in hen welfare, which we had previously 7 8 employed in feather scoring free-ranging hens, where we had also been seeking to minimise 9 10 11 disturbance (Cooper and Hodges, 2010). 12 13 Although there was no effect of barriers on mortality there was an effect on plumage 14 15 condition which became more pronounced with time. Plumage condition was generally 16 For Peer Review Only 17 18 good; a score of 0 signifies no feather damage, and few birds exceeded scores of 10 even by 19 20 the end of the study. Feather pecking was rarely observed in our study, which may be a 21 22 consequence of the availability of grass and other ground cover (Hoffmeyer, 1966). 23 24 25 Aggressive pecks were frequent enough to be detected by behavioural sampling at a 26 27 frequency of approximately 10 pecks per 20 minutes, equivalent to each bird per pen 28 29 receiving an aggressive peck every 2 hours. We did not identify individual birds so it is not 30 31 32 possible to determine if the receipt of aggressive pecks was evenly distributed across all 33 34 birds, or if particular birds were the victims of the majority of pecks. 35 36 37 During scan sampling behavioural observations, it was noted that social interactions 38 39 such as aggressive or reproductive behaviours were rarely detected, which is not unusual if 40 41 activities are rare and/or of short duration. Therefore an alternative sampling technique 42 43 44 involving continuous observation was piloted in week 2. During the pilot study it was found 45 46 that an observer could record a wide range of social interactions by direct observation in 47 48 both open and barrier pens. This information was used to construct a ‘social behaviour 49 50 51 ethogram’ as well as defining a sequence of events that comprised each activity. 52 53 Throughout the study pecking interactions between pheasants were described as 54 55 aggressive pecks, implying the underlying motivation was competition between individuals. 56 57 58 In other species, particularly in the laying hen, a clear distinction is made between 59 60 aggressive pecks related to social competition and so-called gentle feather pecks ( e.g. Bilcik

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 16 of 36

16 1 2 3 and Keeling, 1999). These gentle feather pecks in hens are commonly associated with 4 5 6 redirection of substrate manipulation, for example where there are few environmental 7 8 resources available for foraging activities. Whilst there is no absolutely clear cut distinction 9 10 11 between the two types of pecking they typically differ in their form and location from so- 12 13 called aggressive pecks and take the form of gentle pecks at feathers, rather than more 14 15 forcefully pecks through feathers to skin, and are being applied over a wider part of the body 16 For Peer Review Only 17 18 rather than concentrating on the head region. In our study the pecks delivered by pheasants 19 20 bore close resemblance to the aggressive pecks seen in other captive galliforms, and little 21 22 resemblance to gentle feather pecking. Pecks appeared to be delivered forcefully, occurred 23 24 25 in short bouts (typically a single peck) and were generally avoided by their recipients. 26 27 Whilst it is not possible to determine the underlying motivation for these pecks without 28 29 further study, the conditions that have been associated with gentle feather pecking in hens, 30 31 32 and may be associated with redirected foraging in pheasants (Hoffmeyer, 1966), were absent 33 34 in the breeding pens we studied in that stocking densities were relatively low, and all pens 35 36 37 had good access to ground cover in the form of grass and other vegetation that would have 38 39 facilitated natural patterns of foraging. 40 41 In the absence of related studies it is also difficult to determine if the levels of 42 43 44 aggression and any associated loss of feather condition are comparable to other captive 45 46 breeding flocks or indeed wild pheasants. In a study of aggressive interactions between wild 47 48 male pheasants (Burger 1966), the incidence of aggressive acts between males and 49 50 51 significant deteriorations in tail and wing plumage were noted in some males as the breeding 52 53 season progressed but were not systematically scored. Nevertheless, in our study there was 54 55 good evidence of individuals with severe feather loss, associated with aggressive pecking in 56 57 58 males and treading by males in females, and these were less severe in the pens with barriers. 59 60 A reduction in apparently aggressive interactions, particularly male-male aggression,

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 17 of 36 British Poultry Science

17 1 2 3 was one of the clearest effects of barriers in this study. There are a number of potential 4 5 6 explanations for this finding, including formation of mating territories, more even 7 8 distribution of birds over the enclosure, or simply reducing visual horizons within the 9 10 11 enclosure. It may be that males were better able to form or defend stable territories, with 12 13 barriers dividing up the enclosure or providing clearer landmarks and 14 15 boundaries. Consequently, there may have been fewer territorial disputes but there was little 16 For Peer Review Only 17 18 evidence of territorial behaviour in the study. Alternatively, it is possible that merely 19 20 providing a form of shelter or a foraging substrate drew more birds to the centre of the 21 22 enclosure, resulting in a more even distribution of birds and less social competition. The 23 24 25 pheasants tended to aggregate around the edges of the enclosure in both treatments, 26 27 suggesting strong wall-seeking tendencies (thigmotaxis). Newberry and Shackleton (1997) 28 29 found that layer pullets tended to aggregate around the visual barriers they introduced into 30 31 32 pens. However in our work, although we recorded pheasants perching on the barriers, and 33 34 noted birds sitting and standing beside barriers, we did not systematically record bird 35 36 37 location and cannot determine if the barriers encouraged more even spacing compared with 38 39 open pens. The final potential effect of the visual barriers was to limit the birds' field of 40 41 view, in that birds in non-barrier pens had unrestricted views over the entire pen, whereas in 42 43 44 barrier pens the view was restricted to about a third of the pen from most locations. It was 45 46 noted that cock pheasants would actively disrupt mating acts, so these may not have been as 47 48 visible in a barrier pen. Conversely, barriers provide greater opportunity for the recipients of 49 50 51 aggressive encounters to remove themselves from the field of view of aggressive pen mates, 52 53 which may also have a role in moderating aggressive interactions. Whilst this latter 54 55 explanation seems to concur with the evidence of this study, it would be worthwhile to 56 57 58 include more detailed analysis of social interactions and the utilisation of the enclosure by 59 60 individual birds to determine the role of each of these factors.

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 18 of 36

18 1 2 3 In the absence of comparable data on breeding flocks of pheasants, it is not possible 4 5 6 to determine if our findings are typical of the behaviour of breeding pheasants in captivity. 7 8 Nevertheless some findings are worth discussing and highlight areas for further research. 9 10 11 Firstly, there was very little evidence of the territorial behaviour described in wild pheasants 12 13 (Ridley and Hill, 1987; Grahn et al ., 1993a, 1993b). Although some male-male aggression 14 15 was seen, this was largely aggressive pecking; aggressive displays or bouts of sparring 16 For Peer Review Only 17 18 involving spurs were absent or very rarely observed. There was some evidence of other 19 20 forms of sexual display such as crowing and wing-flapping described in wild pheasants ( e.g. 21 22 Heinz and Gysel, 1970), harassment of females and interruption of mating attempts, but 23 24 25 there was little overt evidence of the courtship displays and harem formation described in 26 27 wild populations (Ridley and Hill 1987; Mateos 1998). It may be that these activities are 28 29 genuinely rare even in wild populations, or that they are more frequent at different periods 30 31 32 of the breeding cycle, or at different times of day than our sampling period. However, it is 33 34 also possible the pheasants adopted very different behavioural strategies in the high density, 35 36 37 female-biased study population. 38 39 In the wild, although the occurrence of dominant territorial harem forming male 40 41 pheasants are well known, subordinate non-territorial males are also common, and these also 42 43 44 engage in copulation with females albeit less frequently than a successful territorial male 45 46 (Burger, 1966; Ridley and Hill, 1987; Göransson et al. , 1990). Competing for females can 47 48 be an expensive occupation involving time and energy, and risking injury and death; males 49 50 51 will therefore be involved in a trade off between these costs and the benefits of mating 52 53 (Grahn, 1993; Mateos and Carranza, 1997). It may be that at high population densities, and 54 55 with a large number of captive hens, it is more cost effective for males to minimise 56 57 58 territorial confrontations. Furthermore, males may not need to engage in courtship displays 59 60 to attract mates if the opportunities of females to escape or express mate choice are limited

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 19 of 36 British Poultry Science

19 1 2 3 by the confines of the enclosure and other husbandry practises. For example, male 4 5 6 secondary traits, such as spur length, tail condition, and wattle colour are thought to be 7 8 involved in female choice (von Schantz et al ., 1989; Göransson et al ., 1990; Mateos, 1998) 9 10 11 but these can be modified in the captive environment. Spurs are shortened to reduce injuries, 12 13 whilst other secondary traits that may advertise male quality in the wild could be negatively 14 15 affected by high bird density or positively affected by medication and diet. This, coupled 16 For Peer Review Only 17 18 with the inability to escape the pen, may limit a female’s ability to operate choice of mates 19 20 and result in a mating system based on coercion by males. This mating system has been 21 22 described by von Schatz et al . (1994) in association with non-territorial males in wild 23 24 25 populations, and contrasts with systems based on male contests or female choice (Mateos, 26 27 1998). A better understanding of mating systems in captivity may inform better management 28 29 of breeding stock and enhance productivity, and it would be of interest to further investigate 30 31 32 the body condition, behaviour and mating success of individual males over the breeding 33 34 season, by marking individual males and genetically tracing their offspring (von Schantz et 35 36 37 al ., 1994). It is nevertheless noteworthy that egg fertility in the study population was high in 38 39 both barrier and control pens (Deeming 2011), suggesting that whatever mating system was 40 41 in operation did not impinge greatly on productivity. 42 43 44 Another unexpected result was that pheasants spent much of their observation time 45 46 engaged in locomotion. Initially this led to a concern regarding observer influence as this 47 48 might be a reflection of avoidance of the observer, but video observations ( i.e. no human 49 50 51 observer in sight) also found a high incidence of locomotion. Video observation had been 52 53 considered for all data collection, but with the equipment available it was not possible to 54 55 achieve the same visual acuity, depth and width of field as a human observer. It is possible 56 57 58 be that this is the normal level of activity for adult pheasants during the breeding season, 59 60 although lack of cover and harassment by males may have led to elevated levels of

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 20 of 36

20 1 2 3 locomotion. Pheasants in the wild tend to seek cover for periods of inactivity ( e.g. Ridley 4 5 6 and Hill, 1987) whereas breeding pens provide little cover or refuges for resting and 7 8 sheltering behaviour. Although bird location was not systematically recorded, it was clear 9 10 11 that pheasants tended to be found along the walls of the breeding pens. There was some 12 13 evidence in this study of less locomotion in the barrier pens, which may have provided a 14 15 degree of additional cover, though this was not significant, and it may be that providing 16 For Peer Review Only 17 18 additional roosting or sheltering sites could allow more resting behaviour. 19 20 The second factor increasing locomotion in both males and females may relate to the 21 22 lack of harem formation or mate guarding. Ridley and Hill (1987) noted that females that 23 24 25 were guarded by a male spent less time in locomotor activity and less time feeding than 26 27 unguarded females. They speculated that females may benefit in terms of reduced predator 28 29 risk (due to vigilance of males) and reduced harassment from non-territorial males, and 30 31 32 noted that their most common response to a non-territorial male was to flee. Walking, 33 34 running or even short bursts of attempted flight significantly increase energy expenditure at 35 36 37 a time when energy demands are high in females for egg formation (hence the more time 38 39 spent feeding by females than males in this study). Whilst breeding pheasants can be fed ad 40 41 libitum to compensate for these additional energy demands they will still be constrained by 42 43 44 digestion rate. It would, therefore, be worthwhile investigating if other means of 45 46 environmental enrichment, such as providing cover to allow greater opportunity for resting, 47 48 improves productivity by, for example, reducing feed costs or even increasing the rate of 49 50 51 lay. 52 53 In summary, the study determined the effect of a cheap modification to breeding pens, 54 55 i.e . the provision of a sight barrier using resources that would be at the disposal of many 56 57 58 farmers, on the behaviour and welfare of breeding pheasants in a captive environment. The 59 60 presence of sight barriers had no impact on bird mortality but birds in the barrier pens had

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 21 of 36 British Poultry Science

21 1 2 3 significantly better plumage damage scores, which appeared to be related to a lower 4 5 6 incidence of aggressive interactions, as indicated by the lower number of bouts of 7 8 aggressive pecking in pens with the barriers. In addition to reducing aggressive interactions 9 10 11 and improving condition, the provision barriers in this study had positive consequences for 12 13 productivity (Deeming et al. , 2011) with significantly higher fertility that persisted for 14 15 longer during the breeding season and significantly higher hatchability. The provision of 16 For Peer Review Only 17 18 sight barriers in breeding pheasant pens could, therefore, provide a cost effective means of 19 20 improving welfare and productivity. 21 22 23 24 25 ACKNOWLEDGEMENTS 26 27 The study was funded in part by the Marsden’s Game Feeds, UK Game Farmers’ 28 29 Association, the Universities’ Federation for Animal Welfare, and the Faculty of 30 31 32 Agriculture, Food and Animals Sciences, University of Lincoln. Many thanks to Bernard 33 34 and Wayne Voce of Shelford Pheasantries for their enthusiasm for this project, for providing 35 36 37 facilities, and for their invaluable assistance in establishing treatment pens. Many thanks go 38 39 to John Flynn for statistical advice and to Oliver Burman and the anonymous referee who 40 41 provided constructive criticism of a previous version of the manuscript. 42 43 44 45 46 REFERENCES 47 48 BILCIK, B. & KEELING, L.J. (1999) Changes on feather condition in relation to feather pecking 49 50 51 and aggressive behaviours in laying hens. British Poultry Science , 40: 444–451. 52 53 BRIGHT, A., JONES, T.A. & DAWKINS, M.S. (2006) A non-intrusive method of assessing 54 55 plumage condition in commercial flocks of laying hens. Animal Welfare , 15: 113-118. 56 57 58 BURGER, G.V. (1966) Observations on aggressive behaviour of male ring-necked pheasants in 59 60 Wisconsin. Journal of Wildlife Management , 30: 57-64.

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 22 of 36

22 1 2 3 BUTLER, D.A. & DAVIS, C. (2010) Effects of plastic bits on the condition and behaviour of 4 5 6 captive-reared pheasants. Veterinary Record , 166: 398-401. 7 8 CAIN, J.R., WEBER, J.M., LOCKAMY, T.A. & CREGER, C.R. (1984) Grower diets and bird 9 10 11 density effects on growth and cannibalism in ring-necked pheasants. Poultry Science , 63: 12 13 450-457. 14 15 COLLIAS, N.E. & TABER, R.D. (1951) A field study of some grouping and dominance relations 16 For Peer Review Only 17 18 in ring-necked pheasants. Condor , 53: 265-275. 19 20 COOPER, J. & HODGES, H. (2010). Effect of tree cover on ranging behaviour of free range 21 22 hens. Proceedings of the 44th Congress of the International Society for Applied Ethology - 23 24 25 Coping in large groups. Lidfors, L., Blokhuis, H. and Keeling, L. (eds) Wageningen 26 27 Academic Publishers, Holland, p.75. 28 29 CORNETTO, T. & ESTÉVEZ, I. (2001) Behavior of the domestic fowl in the presence of vertical 30 31 32 panels. Poultry Science , 80: 1455-1462. 33 34 DEEMING, D.C. (2009) Ratites, game birds and minor poultry species. In: Biology of Breeding 35 36 37 Poultry , HOCKING, P.M. (ed.), pp. 284-304 (Wallingford, CAB International). 38 39 DEEMING, D.C., HODGES, H.R. & COOPER, J.J. (2011). Effect of sight barriers in pens of 40 41 breeding pheasants: II. Reproductive parameters. British Poultry Science , submitted. 42 43 44 DEEMING, D.C. & WADLAND, D. (2002) Influence of mating sex ratio in commercial pheasant 45 46 flocks on bird health and the production, fertility, and hatchability of eggs. British Poultry 47 48 Science , 43: 16–23. 49 50 51 FARM ANIMAL WELFARE COUNCIL (2008) Opinion on the Welfare of Farmed Gamebirds. 52 53 FAWC, London 54 55 FOWLER, J., COHEN. L. & JARVIS, P. (1998). Practical Statistics for Field Biologists , pp. 56 57 58 150-152 (Chichester, Wiley). 59 60 GÖRANSSON, G., VON SCHANTZ, T., FRÖBERG, I., HELGÉE, A. & WITTZELL, H. (1990)

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 23 of 36 British Poultry Science

23 1 2 3 Male characteristics, viability, and harem size in the pheasant Phasianus colchicus. Animal 4 5 6 Behaviour 40: 89-104. 7 8 GRAHN, M. (1993) Mortality in the pheasant Phasianus colchicus during the breeding season. 9 10 11 Behavioral Ecology & Sociobiology , 32: 95-101. 12 13 GRAHN, M., GÖRANSSON, G. & VON SCHANTZ, T. (1993a) Spacing behaviour of male 14 15 pheasants. Phasianus colchicus, in relation to dominance and mate acquisition. Animal 16 For Peer Review Only 17 18 Behaviour , 45: 93-103. 19 20 GRAHN, M., GÖRANSSON, G. & VON SCHANTZ, T. (1993b) Territory acquisition and 21 22 mating success in pheasants Phasianus colchicus : An experiment. Animal Behaviour , 46: 23 24 25 721-730. 26 27 HEINZ, G.H. & GYSEL, L.W. (1970) Vocalization behaviour of the ring-necked pheasant. Auk , 28 29 87: 279-295. 30 31 32 HOFFMEYER, I. (1966) Feather pecking in pheasants – an ethological approach to the problem. 33 34 Danish Review of Game Biology , 6: 2-36. 35 36 37 KJAER, J.B. (2004) Effects of stocking density and group size on the condition of the skin and 38 39 feathers of pheasant chicks. The Veterinary Record , 154: 556-558. 40 41 KJAER, J.B., SORENSEN, P. & SU, G. (2001) Divergent selection on feather pecking behaviour 42 43 44 in laying hens ( Gallus gallus domesticus ). Applied Animal Behaviour Science, 71: 229–239. 45 46 LEIF, A.P. (2005) Spatial ecology and habitat selection of breeding males pheasants. Wildlife 47 48 Society Bulletin , 33: 130-141. 49 50 51 LEONE, E. H. AND ESTÉVEZ, I. (2008) Economic and welfare benefits of environmental 52 53 enrichment for broiler breeders. Poultry Science , 87: 14–21 54 55 MATEOS, C. (1998) Sexual selection in the ring-necked pheasant: a review. Ethology, Ecology & 56 57 58 Evolution , 10: 313-332. 59 60 MATEOS, C. & CARRANZA, J. (1997) Signals in intra-sexual competition between ring-necked

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 24 of 36

24 1 2 3 pheasant males. Animal Behaviour , 53: 471-485. 4 5 6 NEWBERRY, R.C. & SHACKLETON, D.M. (1997) Use of visual cover by domestic fowl: a 7 8 Venetian blind effect? Animal Behaviour , 54: 387-395. 9 10 11 PACEC (2006) The Economic and Environmental Impact of Sporting Shooting . (Cambridge, 12 13 London, Public and Corporate Economic Consultants). http://www.shootingfacts.co.uk/ 14 15 PENNYCOTT, T.W. (2000) Causes of mortality and culling in adult pheasants. Veterinary 16 For Peer Review Only 17 18 Record , 146: 273-278. 19 20 RIDLEY, M.W. & HILL, D.A. (1987) Social organisation in the pheasant (Phasianus colchicus ): 21 22 harem formation, mate selection and the role of mate guarding. Journal of Zoology, London , 23 24 25 211: 619-630. 26 27 SCHENKEL, R. (1956) Zur deutung der balzleistungen einiger phasianiden und tetraoniden. Der 28 29 Ornithologische Beobachter , 53: 182-201. 30 31 32 SCHENKEL, R. (1958) Zur deutung der balzleistungen einiger phasianiden und tetraoniden. 33 34 Zweiter teil. Der Ornithologische Beobachter , 55: 65-95. 35 36 37 TABER, R.D. (1949) Observations on the breeding behaviour of the ring-necked pheasant. 38 39 Condor , 51: 153-175. 40 41 VON SCHANTZ, T., GÖRANSSON, G., ANDERSSON, G., FRÖBERG, I., GRAHN, M., 42 43 44 HELGÉE, A. & WITTZELL, H. (1989) Female choice selects for a viability-based male 45 46 trait in pheasants. Nature , 337: 166-169. 47 48 VON SCHANTZ, T. GRAHN, M. & GÖRANSSON, G. (1994) Intersexual selection and 49 50 51 reproductive success in the pheasant Phasianus colchicus. American Naturalist , 144: 510- 52 53 527. 54 55 WEBSTER, A.B. & HURNIK, J.F. (1990) Behavior, production, and well-being of the laying 56 57 58 hen. 1. Effects of movable roosts, relationship of cage mates, and genetic stock. Poultry 59 60 Science , 69: 2118–2121.

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 25 of 36 British Poultry Science

25 1 2 3 WISE, D.R. (1993) Pheasant Health and Welfare (Cambridge, UK, Piggott Printers). 4 5 6 7 8 9 10 11 12 13 14 15 16 For Peer Review Only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 26 of 36

26 1 2 3 Figure legends. 4 5 6 Figure 1. Diagrammatic plan (not to scale) of the open pen layout (left) and barrier pen 7 8 layout (right) 9 10 11 Figure 2. Mean (+SE) total plumage damage score in barrier and open pens for each gender 12 measured at the mid-point and end of the laying season. Higher scores indicate greater 13 14 damage to feathers. 15 16 For Peer Review Only 17 18 Figure 3. Time budget for male and female pheasants in barrier and open pens. Data are 19 mean values (+SE) calculated from three time periods over the laying season (n=11). 20 21 22 23 Figure 4. Incidence of different social behaviours of male and female pheasants in barrier 24 25 and open pens. Data are mean values (+SE) calculated from three time periods over the 26 laying season (n = 11). 27 28 29 30 Figure 5. Number of pecks delivered per bird for male and female pheasants combined in 31 barrier and open pens. Data are mean values (+SE) calculated from three time periods over 32 33 the laying season (n=11). 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 27 of 36 British Poultry Science

27 1 2 3 4 5 6 Table 1 . Criteria for scoring plumage condition in male and female breeding pheasants. 7 8 9 10 Body Score 0 Perfect feathering, no obvious damage/displacement of feathers 11 12 1 Slight damage to feathers 13 14 2 More extensive displacement of feathers, some loss 15 16 3For Small-medium Peer baldReview patches, surrounding Only feathers ruffled 17 18 19 4 Large bald patches or total baldness of the area 20 21 Tail score 0 All feathers intact and unruffled 22 23 1 Some feathers slightly out of place 24 25 2 Some lost feathers 26 27 28 3 Most feathers lost or damaged 29 30 4 No medium/long tail feathers remaining 31 32 . 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 28 of 36

28 1 2 3 Table 2. Description of recorded behaviours in the time budget of breeding pheasants. 4 5 6 7 8 Behaviour Description of actions 9 10 11 Walk Locomotion at a steady pace. 12 13 Stand Motionless legs, upright position. 14 15 Forage Pecking or scratching at the ground, or other objects (such as straw 16 Forbales). Peer Review Only 17 18 19 Feed/drink Feeding or drinking from hoppers, water or grit bowls. 20 21 Sit Legs folded under the body, resting on the undercarriage. 22 23 Perch Standing or sitting on sight barriers or nest-tins. 24 25 Preen Self-grooming, either with beak or feet. 26 27 28 Mate Reproduction between a male and female bird. 29 30 Other Behaviours rarely observed and combined for analysis 31 32 Enter or exit Male or female birds walking under or out from a tin. 33 nestbox 34 35 36 Run Locomotion at a faster speed than walking. 37 38 Dust bathe Sitting, usually in loose soil, and shuffling earth into feathers and wings. 39 40 Wing flap Moving wings out from the body and flapping. 41 42 Wing stretch Extending wing from the body. 43 44 Defecate Pass waste. 45 46 47 Egg peck Pecking at a whole egg. 48 49 Body shake Rapid quivering of whole body, accompanied by ‘fluffing’ of feathers. 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 29 of 36 British Poultry Science

29 1 2 3 Table 3 . Descriptions of social interactions and behaviours within specific categories in the 4 5 ethogram for social behaviour in breeding pheasants. 6 7 8 9 Behaviour Sequence of actions 10 11 Aggression 12 13 Peck Aimed by either gender at a member of the same, or opposite gender. 14 15 Lunging at any part of another bird with the beak, sometimes 16 Forplucking Peer out feathers. Review Only 17 18 Chase Deliberately following another bird at speeds faster than a walk. 19 20 Fight Males initially face one another and attempt to peck or spar. Can 21 22 involve jumping and attempts to strike with spurs. 23 24 Courtship 25 26 Male to Female Male approaches female and enters lateral display (wings and tail 27 display feathers splayed and tilted towards female) and either circles or 28 29 stands in front of female. Female moves away. 30 31 Squat Female bends legs, either momentary ‘buckling’ and then 32 33 straightening, or remaining in squatting position. Usually occurs as 34 male comes into close proximity. 35 36 Display 37 38 Male : Male display Same motion as male to female display, male displayed to moves 39 40 away. 41 42 Vocal and wing flap Male extends neck and stands tall, emits a guttural call and flaps 43 wings. Usually accompanied by another male emitting a quieter call 44 45 during the performance, but it was not always possible to distinguish 46 47 if this had occurred or not. 48 49 Avoidance 50 Avoidance Deliberately moving out of the path of another, oncoming bird, either 51 52 at a walk, run or by jumping/flapping. The oncoming bird must be 53 54 the obvious reason for movement to be recorded as avoidance. 55 56 Mating 57 58 Successful mating Intercourse between male and female birds, where the cloaca 59 appeared to meet. 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 30 of 36

30 1 2 3 4 Interrupted mating Intercourse between male and female birds, during which another 5 male bird approaches the pair and the mating male either moves 6 7 away during approach, or when the other male reaches the pair. 8 9 Attempted mating Female squats and is apparently receptive to male, but copulation 10 11 fails due to M not mounting after treading, or attempting to mate 12 with the side or front of a female. 13 14 Attempted mating Male approaches female and attempts to tread her back, female 15 16 (female reject)For moves Peer away either Review during treading or asOnly male mounts. 17 18 Out of sight mating Copulation of unknown type occurs out of sight of observer ( e.g. 19 only tails/heads are visible). 20 21 Forced mating Male holds female down by pecking, sometimes having chased 22 23 female or mounted forcibly whilst the hen is sitting. Female attempts 24 25 to escape but is prevented. 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 31 of 36 British Poultry Science

31 1 2 3 4 5 6 Table 4. Results of repeated measures ANOVA tests to examine the effects of gender (male 7 and female) and pen type (barrier and open) for the main behaviours in the time budget of 8 9 breeding pheasants. Values are F 1,38 ratios with significance level in parentheses (NS = not 10 11 significant). 12 13 14 15 Behaviour Gender Type Interaction 16 For Peer Review Only 17 18 Walk 0.11 (NS) 0.61 (NS) 0.03 (NS) 19 20 Stand 3.06 (NS) 0.85 (NS) 1.03 (NS) 21 22 Forage 27.53 (P < 0.001) 2.98 (NS) 2.42 (NS) 23 Feed/drink 30.61 (P < 0.001) 0.00 (NS) 0.00 (NS) 24 25 Sit 2.87 (NS 2.16 (NS) 0.33 (NS) 26 27 Perch 5.67 (P < 0.05) 90.36 (P < 0.001) 0.24 (NS) 28 29 Preen 2.31 (NS) 10.60 (P < 0.01) 1.43 (NS) 30 31 Mate 1.25 (NS) 0.15 (NS) 0.19 (NS) 32 Other 6.33 (P < 0.05) 0.19 (NS) 0.03 (NS) 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 32 of 36

32 1 2 3 Figure 1. Diagrammatic plan (not to scale) of the open pen layout (left) and barrier pen 4 5 6 layout (right) 7 8 9 10 11 12 13 14 15 16 For Peer Review Only 17 18 19 20 21 22

23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 33 of 36 British Poultry Science

33 1 2 3 Figure 2. Mean (+SE) total plumage damage score in barrier and open pens for each gender 4 5 measured at the mid-point and end of the laying season. Higher scores indicate greater 6 7 damage to feathers. 8 9 10 11 12 13 14 15 16 For Peer Review Only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 34 of 36

34 1 2 3 Figure 3. Time budget for male and female pheasants in barrier and open pens. Data are 4 5 mean values (+SE) calculated from three time periods over the laying season (n=11). 6 7 8 9 10 11 12 13 14 15 16 For Peer Review Only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps Page 35 of 36 British Poultry Science

35 1 2 3 Figure 4. Incidence of different social behaviours of male and female pheasants in barrier 4 5 and open pens. Data are mean values (+SE) calculated from three time periods over the 6 7 laying season (n = 11). 8 9 10 11 12 13 14 15 16 For Peer Review Only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps British Poultry Science Page 36 of 36

36 1 2 3 Figure 5. Number of pecks delivered per bird for male and female pheasants combined in 4 5 barrier and open pens. Data are mean values (+SE) calculated from three time periods over 6 the laying season (n=11). 7 8 9 10 11 12 13 14 15 16 For Peer Review Only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

E-mail: [email protected] URL: http://mc.manuscriptcentral.com/cbps