Department of Physics, Chemistry and Biology
Master Thesis
Effects of the captive environment and
enrichment on the daily activity of European
bison (Bison bonasus)
Erika Godoy
LiTH-IFM- Ex--2142--SE
Supervisor: Mats Amundin, Linköpings universitet
Examiner: Matthias Laska, Linköpings universitet
Department of Physics, Chemistry and Biology Linköpings universitet
Avdelning, Institution Division, Department Avdelningen för biologi
Instutitionen för fysik och mätteknik
Datum Date 2009-06-05 Rapporttyp Report category Licentiatavhandling x Examensarbete C-uppsats x D-uppsats Övrig rapport _______________ Språk Language Svenska/Swedish x Engelska/English ________________ ISBN
LITH-IFM-A-EX--—09/2142—SE
__________________________________________________ ISRN
__________________________________________________
Serietitel och serienummer ISSN
Title of series, numbering
Handledare
Supervisor: Mats Amundin Ort
Location: Linköping
URL för elektronisk version
Titel
Title:
Effects of the captive environment and enrichment on the daily activity of European bison (Bison
bonasus)
Författare
Author: Erika Godoy
Sammanfattning
Abstract:
When breeding wild species in captivity, the animals may gradually become more adapted to captivity and therefore less suited for reintroduction which is the ultimate goal for some species. This study measured the activity budget of European bison (Bison bonasus) in six enclosures in Sweden with the aim to find out how the characteristics of the enclosures – with and without pasture - influenced the activity budget. The results show that there were significant differences in the activity budget, i.e. the activity was higher in the enclosures with pasture than in enclosures with barren ground. However, since barren enclosures were smaller than naturalistic, it was not possible to exclude the effect of size. Judged from observations of bison in the wild, there seems to be a direct correlation between food availability and ranging, indicating that enclosure characteristics affect activity more than size. The bison foraged differently in the two enclosure categories, but the total amount of time spent on feeding did not differ. A feeding enrichment experiment showed to have more positive effects in the barren enclosures than in the naturalistic ones, as the amount of time of inactivity decreased in the former. Since the genetic characteristics of all Swedish bison are very similar, the differences between the two enclosure groups indicate that the animals still have the ability to respond appropriately to improved environmental quality. Hence the next step in the assessment of the suitability of these bison for release would be to study them during an acclimatisation program.
Nyckelord
Keyword:
Content
1 Abstract……….………... 1
2 Introduction………..……… 1
3 Material and Methods……….……….……… 3
3.1 Observation sites and animals………. 3
3.1.1 Barren enclosures………... 3
3.1.2 Naturalistic enclosures………. 4
3.2 Observational methods……… 5
3.3 Data analysis……….……… 6
4 Results ………... 6
4.1 Enclosure design effects on the activity budget... 6
4.2 Group effects on the activity budget………... 8
4.3 Feeding enrichment effect on the activity budget……..…... 10
5 Discussion……… 12
5.1 Effects of enclosure design ……….. 12
5.2 Effects of social structure……….. 13
5.3 Effects of feeding enrichment………. 14
5.4 Conservation implications……… 15
5.5 Conclusion……… 16
6 Acknowledgements……….. 17
1 Abstract
When breeding wild species in captivity, the animals may gradually become more adapted to captivity and therefore less suited for reintroduction which is the ultimate goal for some species. This study measured the activity budget of European bison (Bison bonasus) in six enclosures in Sweden with the aim to find out how the characteristics of the enclosures – with and without pasture - influenced the activity budget. The results show that there were significant differences in the activity budget, i.e. the activity was higher in the enclosures with pasture than in enclosures with barren ground. However, since barren enclosures were smaller than naturalistic, it was not possible to exclude the effect of size. Judged from observations of bison in the wild, there seems to be a direct correlation between food availability and ranging, indicating that enclosure characteristics affect activity more than size. The bison foraged differently in the two enclosure categories, but the total amount of time spent on feeding did not differ. A feeding enrichment experiment showed to have more positive effects in the barren enclosures than in the naturalistic ones, as the amount of time of inactivity decreased in the former. Since the genetic characteristics of all Swedish bison are very similar, the differences between the two enclosure groups indicate that the animals still have the ability to respond appropriately to improved environmental quality. Hence the next step in the assessment of the suitability of these bison for release would be to study them during an acclimatisation program.
Keywords:
Activity budget, barren enclosure, captivity, feeding enrichment, naturalistic enclosure, reintroduction, social structure
2 Introduction
The reintroduction of European bison (Bison bonasus, Linnaeus 1758) into the wild is one of few successful mammalian ones using captive-bred animals (Wemmer and Derrickson 1987, cited by Kleiman 1989). This species was extirpated in the wild in the 1920s with only 54 individuals left in zoological gardens originating from 12 founders. As a result of intense captive breeding and reintroduction, today the total population is approximately 3,200 (Pucek 2004, Kleiman 1989) of which 1,400 live in captivity. As a result, the current population is highly inbred with a mean kinship of 24.9% in the free-living population (Pucek 2004). However, the captive population has a higher genetic diversity with a mean kinship of 16.1% (Pucek 2004) hence, the captive populations would serve as a valuable reserve of the species gene pool and as a source for further reintroductions or re-stocking of wild populations.
Success of captive breeding programs depends on many aspects, where maintaining genetic diversity is crucial. In addition to genetic diversity, survival of reintroduced animals depends largely on behaviour. Behavioural deficiencies in foraging, predator avoidance and habitat choice appear more frequently in animals that lack opportunities to associate with wild individuals in natural setting during critical learning periods and these behaviours are often critical for survival (Snyder et al. 1996; Beissinger 1997). Hence, despite comprehensive genetic management, populations may gradually become more adapted to captivity, as adaptations to barren, protected and unchallenging captive environments (Allendorf 1993). Studies have shown that rearing environments do influence behaviours (McPhee 2003; Kelley
et al. 2005; Håkansson and Jensen 2005) and affect the success of the release of captive
animals into the wild (Kleiman 1989; Biggins et al. 1999; Stoinski et al. 2003). Therefore it is important to know how the captive environment influences behaviour.
One way to evaluate the environment’s influence on the behaviour is to observe the animal’s activity budget in different environments (Burrell and Altman 2006). By studying the activity budgets of wild animals and comparing their schedule with those of their captive
counterparts, it is possible to determine whether the captive setting is conductive to the expression of species-typical behaviour (Maple and Perkins 1996). Activity budgets are related to important physiological and environmental factors such as energy balance, body size and food availability (Jaman and Huffman 2008). Furthermore, diet and habitat structure are known to affect time spent in activities because of trade-offs between obtaining food and the cost in energy of doing so. Hence, it is important to determine whether different types of captive habitats differentially affect the development of skills appropriate to living in the wild and if the habitat is dynamic enough to encourage flexibility of behaviour (Burrell and Altman 2006). Studies of different animals budget have shown that the amount of naturalistic elements influence the animal differently from animals housed in barren environments (Lee and Berejikian 2008; Beisner and Isbell 2008). Burnell and Altman (2006) studied cotton-top tamarins (Saguinus oedipus) and found that the environment most similar to their natural habitat (rainforest) elicited greater amount of activity compared to a plant-less and relatively barren environment. A group of mandrills (Mandrillus sphinx) were moved from a traditional, indoor exhibit to an “ecologically representative” exhibit (Chang et al. 1999). As a response, the amount of time feeding and locomotion increased, further, the amount of time stationary decreased after the move. Furthermore, these effects have also been shown in leopards (Panthera pardus) where the type of enclosure housing influenced their activity budget as the leopard in larger, more complex enclosures showed more active behaviour and less stereotypic pacing (Mallapur and Chellam 2002). These examples show how different elements in a captive environment influence the activity budget and that appropriate, “natural”
behaviours are most likely to occur in naturalistic surroundings (Maple and Perkins 1996). A way to provide environmental stimuli necessary for optimal psychological and
physiological wellbeing is to use environmental enrichment (Swaisgood and Shepherdson 2005). Often an emphasis is placed on the importance of providing enrichment that is appropriate to the biology of the species under consideration, i.e. promoting natural behaviour (Mellen and Macphee 2001; Newberry 1995). For animals in genetic conservation programs, enrichment should result from modifications to make the captive environment more similar to the environment where the animals are destined to be released (Newberry 1995). Different kinds of enrichments can be used depending on what kind of behaviour you want to stimulate and enrichment falls into four general categories; feeding, structural, sensory and social enrichment (Miller and Mench 2005). Feeding enrichment can be classified depending on the goal of the enrichment; increase of capture time (live prey), increase search time (scatter or hide) and increase handling time (browse, whole food, ice block with food). Studies on feeding enrichment have shown increased overall activity (Dishman et al. 2009) and increased time spent feeding (Kerridge 2005; Miller and Mench 2005). Many animals, as well as the European bison, spend a large portion of their daily activity budget in the search for and consumption of food. In captivity, however, food is generally provided at a single location and in sufficient quantities so that the time and energy spent in feeding are greatly reduced relative to what is experienced by their wild counterparts (Newberry 1995).
The overall purpose of this study was to investigate whether the different captive environments, differing from each other regarding their enclosure design and social environment, would affect the behaviour of the European bison kept in them. This could be of importance to know as the European bison in Sweden are part of the EEP, the European Endangered Species Programme. The EEP is a management programme which produces studbooks, plans for future management and makes recommendations as to which animals should breed and which individual should move from one zoo to another (EAZA 2008), and what animals should be selected for release. If the European bison kept in Sweden would be subject for re-stocking the wild population, it is of importance to see if/how their captive
environment might have an influence on their behaviour and if the activity budget in the captive population in Sweden differs largely from that in the wild.
As a result of the reintroduction programs, free-ranging and semi-free herds now live in Poland, Lithuania, Belarus, Russian federation, Ukraine and Slovakia. The daily activity rhythm of the wild living European bison is polyphasic: phases of foraging alternate with resting spent mostly on rumination (Pucek 2004). During natural conditions, the European bison spends approximately 60% of its time on feeding, 32% on resting and 8 % on walking. Walking usually occurs when the animals move from one feeding site to another, to watering places and in search for a resting site (Cabon-Raczynska et al. 1987). The European bison demonstrates habitat preferences in close relation to food preferences where deciduous and mixed forests are preferred. In the total diet, threes and shrubs constitute 33% while grasses, sedges and herbs feature at 67% (Pucek 2004). Habitat utilization depends on group size and structure and rotational exploitations of the environment (Pucek 2004, Cabon-Raczynska et
al. 1987, Daleszczyk et al. 2007). The European bison is a gregarious animal. Mixed groups
and bull groups are the basic units observed, where mixed groups (young males and cows) consist on average of 8–13 animals and bull groups with an average of two animals; bulls also often lead solitary lives (Pucek 2004). The cow groups are not family units as the size and structure change depending on season. Groups meet frequently, combine and then quickly split again, exchanging some of the individuals (Pucek 2004).
As an attempt to evaluate how the captive environments of European bison kept in Sweden influences their behaviour, my thesis aims at investigating the activity budget in these animals and relate this to the characteristics of their enclosures. One of the hypotheses was that the design of the enclosure has an effect on the activity budget and I predicted that the activity level would be higher in larger enclosures than in smaller enclosures. Furthermore, I predicted that the activity level would be higher in naturalistic enclosure than in barren enclosures and the time spent feeding would be higher in enclosures with natural grazing. Secondly, if the group composition has an effect on the activity budget I predicted that animals in large groups would have higher activity level than small groups and that cows would differ in activity compared to bulls. Finally, to assess the effect of feeding enrichment in the different enclosures, my hypothesis was that the activity budget would change if additional browse was positioned and scattered in the enclosures, due to increased activity level in foraging and walking.
3 Material and Methods
3.1 Observation sites and animals
This study took place from June to early October of 2008. A total of six locations in Sweden which hold European bison were included in the study, five zoos and one breeding centre. All adult animals on these locations were observed. Each site was visited twice; the data collected during the first visit was used as a baseline and during the second visit extra browse was added and scattered to the degree possible, i.e. a lighter form of enrichment.
3.1.1 Barren enclosures Avesta Visentpark
There are several enclosures within the park, ranging from pasture to more barren enclosures and with different group compositions. The park itself is not open to the public except for a tower outside the park from which the animals can be observed. The enclosure used in this study was a smaller pen (2500 m2) with a total of four animals, one bull and three cows (mean age = 12 years). The enclosure design was barren, with the ground consisting of gravel and a lack of natural elements such as trees. The enclosure seemed to having been two enclosures,
with the dividing wooden fence partly removed, enabling passage between the two. Each back end of the enclosure had a house where hay was offered ad libitum (unlimited access). Additionally, there were four food containers in which food was offered (e.g. potatoes, oat etc) twice a day and two containers on the front wooden fence which were daily filled with fresh cut grass. There were also three water cups.
Borås Zoo
This enclosure was largely barren with a rocky slope which also contained little grass but not enough to be classified as pasture. The total area was 3450 m2 divided in a smaller and a larger pen with free passing between the two except for once a day when the larger area was cleaned. The larger part of the enclosure held a large circular hay rack on a flat concrete block in which hay was provided ad libitum and two stone piles in which browse was offered. There were also a few trees, logs and a water cup. In the smaller area there was a house in which pellets were given during the cleaning of the larger area. In these enclosures there was a total of nine European bison but the study only observed the six adults, one male and five females (mean age = 8.8 years). The enclosure was surrounded by a wooden fence and a stone wall. Skansen
This was a barren enclosure where the ground consists of gravel. The enclosure had some natural elements such as several trees, both living and dead, as well as logs and rocks. There was also a house in which the animals were housed individually separated and given hay and pellets during the cleaning of the enclosure. By the house there were two water disposals and a pile with more or less decorticated logs which functioned as a “forest” when fresh browse was inserted daily. In the middle of the area there was an elevation and another pile of decorticated logs. The total area was 2668 m2 and held three animals, one bull and two cows (mean age = 7.33). The whole enclosure was surrounded by electric fences with a gap to the walls of the enclosure; these gaps were used by wild boars which were kept together with the bison.
3.1.2 Naturalistic enclosures Skåne Zoo
There were three enclosures which were open to each other at all times, allowing the animals to move freely between them. The larger pen contained pasture and several small areas of exposed soil where the animals could engage in wallowing activities. The middle pen also had pasture and areas of soil and several trees, some dead due to decorticating. In this part, browse was also presented each morning and occasionally also apples. The smallest pen was divided in two and was adjacent to the two feeding houses where the animals were fed hay and pellets. The enclosure was surrounded by a wooden fence, a wooden wall and a stone wall. The total area of all enclosures was 7100 m2. There was a total of six animals, five cows of different ages and one bull (mean age = 5 years).
Lycksele Zoo
This was a hilly enclosure with a total area of 24160 m2. The ground was pasturage with a large number of trees (pine, spruce and birch) of different sizes and several decorticated trees. A feeding trough, where hay and pellets were presented once a day, was situated in a smaller enclosed area where the ground was earthy. There was a total of three animals in this enclosure, but only two, one bull and one cow, were old enough to be a part of this study (mean age = 2 years). The enclosure was surrounded by a wooden fence and coarsely meshed netting.
Kolmården Zoo
These bisons were kept in two separate enclosures, a smaller one measuring 1400 m2 during the evening and night, and a larger one, measuring 4000m2, during the day. The smaller enclosure contained two houses, one with wooden chips on the ground for resting and one with a concrete floor, in which the supplementary food was given (hay, grass, and pellets). The ground consisted of grass with patches of soil, with a total of five living trees and three dead ones, as well as a few logs. There was a pile of stones in the middle, which served as a supporter for browse. There were two water cups. The larger enclosure had a smaller house with wooden chips on the ground. A small lake and a water cup provide water and several trees which were protected from decorticating as well as dead trees offered a naturalistic impression. In this enclosure the food was given on a rocky area in the southern part of the enclosure. A visitor’s bridge goes through the centre of the area, almost dividing it in two, with a rocky northern part and a grassier southern part. In these enclosures three animals were kept, one bull and two cows (mean age = 3 years). The larger enclosure also held two reindeers (one at the second visit).
3.2 Observational methods
The animals at each location were observed using a hand-held computer (Psion Workabout) and the behaviour study software Observer v 3.0 (Noldus Information Technology, The Netherlands). The behaviour states (Table 1) were recorded continuously for 72 hours for the cows and 24 hours for the bulls using rotational focal sampling and switching focal animal every fifth hour. The reason for observing the bulls for a shorter time than the cows was that there was only one bull in each enclosure. When the animals were out of sight they were recorded as “not available” (N/A) and this was excluded from the analysis. Binoculars (8x40) and night-vision binoculars were used when necessary. In order to test the response to feeding enrichment, extra browse was added and spread out in all enclosures and then the observations were repeated for 72 hours and 24 hours, respectively. The browse were recorded in the “feeding” state as it required “active eating” and handling time, i.e. it was not presented in a container. Worth noting is that in all enclosures browse was available as part of the routine, i.e. browse was not novel to the animals. At Skåne Zoo however, extra space and grazing became available as an extra field was opened in the hind part of the enclosure. At Kolmården Zoo the opposite occurred, due to an incident with the male reindeer in the large enclosure. As a result, the bisons were held in the small enclosure during the whole second observation bout.
Table 1. The behavioural states and their definitions used in this study
Behavioural state Definition
Feeding Eating “actively” from natural sources such as grass and added browse
Feeding supplementary
Eating from supplementary food sources from a container such as hay and pellets
Walk Moving all four legs and at least one body length Run Trotting or galloping at least one body length
Standing Standing still without feeding or ruminating, but including grooming
Standing ruminating Standing still and ruminating
Lying Lying down or wallowing
Lying ruminating Lying down and ruminating
3.3 Data analysis
All analyses of activity budgets were performed on 24 hours observations. Comparisons were made on the activity budgets of bulls/cows, barren/naturalistic and before/after the feeding enrichment. A normality test showed that the data was not normally distributed. Therefore analyses comparing bulls with cows and barren enclosures with naturalistic ones were made with non-parametric Mann-Whitney U-test. Comparisons between group sizes were made using Kruskal-Wallis tests. Analyses comparing the activity budgets before and after the feeding enrichment in the different enclosures (paired data) were done using Wilcoxon signed ranks test. Spearman rank correlation test was used to estimate correlation between behaviours and enclosure sizes, and between age and behaviours. Mann- Whitney U-test and Kruskal-Wallis test were performed in Minitab 15 statistical software. Wilcoxon signed ranks test and Spearman rank correlation test were performed in SPSS for Windows v.17. All deviations from the mean values are given as standard error of the mean (S.E) and with a significance level of p≤0.05 or p≤0.01.
4 Results
4.1 Enclosure design effects on the activity budget
When comparing the enclosures with grazing possibilities (Skåne Zoo, Lycksele Zoo and Kolmården Zoo) to the enclosures with a barren ground (Avesta Visentpark, Borås Zoo and Skansen) there were several significant differences (Figure 1). Notably, the bison in the naturalistic enclosures walked more (U = 184, p<0.01), as well as ran more (U = 190, p<0.01) than the bison in the more barren enclosures, although the time spent on running was fairly small. There were significant distinctions in the different feeding categories as the bison in the barren environment fed more on the supplemented food (U = 68, p<0.01) and those in the naturalistic environment fed more from natural food sources (U = 187, p<0.01). The bison in the naturalistic enclosures also spent a larger amount of time lying down without ruminating (U = 179, p<0.01) than the bison in the barren enclosures. There was also a tendency for the bison in barren enclosures to stand and ruminate more than bison in naturalistic enclosures (U = 105.5, p=0.0548).
0 5 10 15 20 25 30 35 Feedi ng Feedi ng supl Standing Standi ng r um i Wal k Run Lying Lyin g rum i N/A A m oun t of time /da y ( % ) Natural Barren ** ** ** ** ** 0 5 10 15 20 25 30 35 Feedi ng Feedi ng supl Standing Standi ng r um i Wal k Run Lying Lyin g rum i N/A A m oun t of time /da y ( % ) Naturalistic Barren ** ** ** ** ** 0 5 10 15 20 25 30 35 Feedi ng Feedi ng supl Standing Standi ng r um i Wal k Run Lying Lyin g rum i N/A A m oun t of time /da y ( % ) Natural Barren ** ** ** ** ** 0 5 10 15 20 25 30 35 Feedi ng Feedi ng supl Standing Standi ng r um i Wal k Run Lying Lyin g rum i N/A A m oun t of time /da y ( % ) Naturalistic Barren ** ** ** ** **
Figure 1. Mean amount of time in the different behaviour states in the natural and barren enclosures. (±S.E showing variations between enclosures, ** p<0.01)
When adding the time spent feeding on both natural and supplementary food sources for both types of enclosures, the total time did not differ significantly from each other (Figure 2). For the barren enclosures the total combined mean time spent on feeding was 30.8% and for the naturalistic enclosures 28.0% of their total daily activity.
0 5 10 15 20 25 30 35 Naturalistic Barren A m oun t o f ti me /d a y ( % ) Feeding supl Feeding 0 5 10 15 20 25 30 35 Barren A m oun t o f ti me /d a y ( % ) Feeding supl Feeding 0 5 10 15 20 25 30 35 Naturalistic Barren A m oun t o f ti me /d a y ( % ) Feeding supl Feeding 0 5 10 15 20 25 30 35 Barren A m oun t o f ti me /d a y ( % ) Feeding supl Feeding
Figure 2. The total time spent feeding, both from natural food sources and supplementary feeding (feeding supl), in the different enclosure types.
The Spearman rank correlation test (Table 2) showed that for the behaviour “run” (rs = 0.886, d.f. = 4, p<0.01) there appeared to be an association between the amount of the behaviour and the enclosure size, that is, the larger the enclosure, the larger amount of time of running behaviour was performed by the animals. The same applies to the behaviour “feeding on natural food sources” (rs = 0.771 d.f. = 4, p<0.05) and “lying” (rs = 0.886, d.f. = 4, p<0.01). In “feeding on supplemental food sources” (rs = -0.943, d.f. = 4, p<0.01) and “standing ruminating” (rs = -0.771, d.f. = 4, p<0.05) the opposite applies as the larger the enclosure, the
smaller amount of time was spent on these behaviours. There was also a trend for “walking” (rs = 0.714, d.f. = 4, p=0.055) more in larger enclosures than in the smaller ones.
Table 2. Results of Spearman rank correlation on enclosure size and mean amount of time spent in each behaviour class.
Behaviour Correlation coefficient (rs) Significance
Feeding 0.771 0.036 Feeding supplemental -0.943 0.002 Standing -0.257 0.311 Standing ruminating -0.771 0.036 Walk 0.714 0.055 Run 0.886 0.009 Lying 0.886 0.009 Lying ruminating -0.314 0.272
4.2 Group effects on the activity budget
Significant differences were found between the group sizes in the behaviour state “standing ruminating” (H = 10.05, d.f. = 3, p<0.05; Figure 3) but only a trend was found in “standing” (H = 6.74, d.f. = 3, p = 0,081). Both “lying” and “lying ruminating” were significantly different between the group sizes (H = 12.64, d.f. = 3, p<0.01, H = 12.11, d.f. = 3, p<0.01), as were “run” (H = 8.15, d.f. = 3, p<0.05). For the two eating states (“feeding” and “feeding supplementary”) and “walking” no significant differences were found between the group sizes. 0 5 10 15 20 25 30 35 40 Feeding Feeding supl Standing Standing rumi
Walk Run Lying Lying
rumi N/A A m o u n t of tim e /d a y ( % ) 2 3 4 6 0 5 10 15 20 25 30 35 40 Feeding Feeding supl Standing Standing rumi
Walk Run Lying Lying
rumi N/A A m o u n t of tim e /d a y ( % ) 2 3 4 6
Figure 3. Mean amount of time per day (±S.E.) spent in behavioural states by each group size consisting of 2, 3, 4 or 6 animals.
When comparing cows to bulls (Figure 4a), no significant differences were found when pooling all cows and bulls from the different enclosures. However, in the barren enclosures (Figure 4b) there was a significant gender difference in the behaviour “run” (U = 70, p<0.05), otherwise no significant differences between cows and bulls were found in the barren enclosures or in the naturalistic enclosures (Figure 4c).
0 5 10 15 20 25 30 35 Fee ding Fee ding supl Sta ndi ng Sta ndi ng ru mi Wal k Run Lying Lyin g rum i N/A A m oun t of time/d a y ( % ) Female Male a) 0 5 10 15 20 25 30 35 Fee ding Fee ding supl Sta ndi ng Sta ndi ng ru mi Wal k Run Lying Lyin g rum i N/A A m oun t of time/d a y ( % ) Cows Bulls a) 0 5 10 15 20 25 30 35 Fee ding Fee ding supl Sta ndi ng Sta ndi ng ru mi Wal k Run Lying Lyin g rum i N/A A m oun t of time/d a y ( % ) Female Male a) 0 5 10 15 20 25 30 35 Fee ding Fee ding supl Sta ndi ng Sta ndi ng ru mi Wal k Run Lying Lyin g rum i N/A A m oun t of time/d a y ( % ) Cows Bulls a) 0 5 10 15 20 25 30 35 Feedi ng Feed ing supl Stan ding Stan ding rum i Wal k Run Lying Lying rumi N/ A A m ou nt of time/d a y ( % ) Female Male * b) 0 5 10 15 20 25 30 35 Feedi ng Feed ing supl Stan ding Stan ding rum i Wal k Run Lying Lying rumi N/ A A m ou nt of time/d a y ( % ) Cows Bulls * b) 0 5 10 15 20 25 30 35 Feedi ng Feed ing supl Stan ding Stan ding rum i Wal k Run Lying Lying rumi N/ A A m ou nt of time/d a y ( % ) Female Male * b) 0 5 10 15 20 25 30 35 Feedi ng Feed ing supl Stan ding Stan ding rum i Wal k Run Lying Lying rumi N/ A A m ou nt of time/d a y ( % ) Cows Bulls * b)
0 5 10 15 20 25 30 35 Fee ding Feedi ng s upl Stan ding Stan ding rumi Wa lk Run Lying Lyin g ru mi N/A A m oun t of time/d a y ( % ) Female Male c) 0 5 10 15 20 25 30 35 Fee ding Feedi ng s upl Stan ding Stan ding rumi Wa lk Run Lying Lyin g ru mi N/A A m oun t of time/d a y ( % ) Cows Bulls c) 0 5 10 15 20 25 30 35 Fee ding Feedi ng s upl Stan ding Stan ding rumi Wa lk Run Lying Lyin g ru mi N/A A m oun t of time/d a y ( % ) Female Male c) 0 5 10 15 20 25 30 35 Fee ding Feedi ng s upl Stan ding Stan ding rumi Wa lk Run Lying Lyin g ru mi N/A A m oun t of time/d a y ( % ) Cows Bulls c)
Figure 4. Comparisons between; a) cows and bulls in all enclosures, b) cows and bulls kept in a barren environment c) cows and bulls kept in a naturalistic environment (±S.E., *p<0.05)
Additionally, Spearman rank correlation test was also performed on mean age (Table 3) and amount of time spent in each behaviour class. Only ”run” showed a significant correlation with age (rs = -0.829, d.f. = 4, p<0.05) where younger bison ran for a longer period of time than older bison.
Table 3. Results of Spearman rank correlation on mean age and mean amount of time spent in each behaviour class
Behaviour Correlation coefficient (rs) Significance
Feeding -0.771 0.072 Feeding supplemental 0.771 0.072 Standing 0.257 0.632 Standing ruminating 0.657 0.156 Walk -0.657 0.156 Run -0.829 0.042 Lying -0.600 0.208 Lying ruminating -0.143 0.787
4.3 Feeding enrichment effects on the activity budget Barren enclosures
When pooling the barren enclosures and comparing them before and after the feeding enrichment (Figure 5a), the amount of “feeding on natural food sources” increased significantly (z = -3.059, p<0.01) and correspondingly reduced their “feeding on supplemented food sources” (z = -2.197, p<0.05). Additionally, the time spent on “standing” were significantly lower after the addition of browse (z = -2.275, p<0.05).
Figure 5b shows the mean amount of time active (walk, run) and inactive (lying, standing and ruminating) as well as the feeding categories. The total time inactive were significantly less after the enrichment (z = -2.275, p<0.05) but no changer in total time active.
0 5 10 15 20 25 30 35 40 Feedi ng Feedi ng s upl Sta nding Stand ing rum i Wal k Run Lying Lying rum i N/A A m oun t of time /da y ( % ) Before After ** * * 0 5 10 15 20 25 30 35 40 Feedi ng Feedi ng s upl Sta nding Stand ing rum i Wal k Run Lying Lying rum i N/A A m oun t of time /da y ( % ) Before After ** * *
a)
0 5 10 15 20 25 30 35 40 Feedi ng Feedi ng s upl Sta nding Stand ing rum i Wal k Run Lying Lying rum i N/A A m oun t of time /da y ( % ) Before After ** * * 0 5 10 15 20 25 30 35 40 Feedi ng Feedi ng s upl Sta nding Stand ing rum i Wal k Run Lying Lying rum i N/A A m oun t of time /da y ( % ) Before After ** * *a)
0 20 40 60 80Feeding Feeding supl
**
*
*
0 20 40 60 80Feeding Feeding supl Inactive Active
A m ou nt o f t ime /d ay ( % ) After Before After
*
*
b)
0 20 40 60 80Feeding Feeding supl
**
*
*
0 20 40 60 80Feeding Feeding supl Inactive Active
A m ou nt o f t ime /d ay ( % ) After Before AfterAfter Before After
*
*
b)
Figure 5. Mean amount of time spent per day (±S.E.) on the different behaviour states before and after addition and scattering of browse, all barren enclosures combined. a) shows each behaviour state, b) shows mean time active and inactive (**p<0.01, *p<0.05).
Naturalistic enclosures
When pooling the three naturalistic enclosures and comparing them before and after the addition of browse (Figure 6), the amount of time “standing” decreased significantly (z = -2.293, p<0.05). Additionally, there was a significant decrease in “walking” (z = -1.886, p<0.05).
Figure 6b shows the mean amount of time active (walk, run) and inactive (lying, standing and ruminating) as well as the feeding categories. The total time active decreased significantly (z = -2.293, p<0.05), however, the other categories did not change significantly after the enrichment.
0 5 10 15 20 25 30 35 40 Feeding Feeding s upl Standing Standing rum i Wa lk Run Lying Lyin g rum i N/A A m oun t of time /da y ( % ) Before After * * 0 5 10 15 20 25 30 35 40 Feeding Feeding s upl Standing Standing rum i Wa lk Run Lying Lyin g rum i N/A A m oun t of time /da y ( % ) Before After * *
a)
0 5 10 15 20 25 30 35 40 Feeding Feeding s upl Standing Standing rum i Wa lk Run Lying Lyin g rum i N/A A m oun t of time /da y ( % ) Before After * * 0 5 10 15 20 25 30 35 40 Feeding Feeding s upl Standing Standing rum i Wa lk Run Lying Lyin g rum i N/A A m oun t of time /da y ( % ) Before After * *a)
After Before After 0 20 40 60 80 Feeding Feeding supl 0 20 40 60 80 Feeding Feeding supl Inactive Active A m o u n t o f t ime /d ay (% )*
b)
After Before After 0 20 40 60 80 Feeding Feeding supl 0 20 40 60 80 Feeding Feeding supl Inactive Active A m o u n t o f t ime /d ay (% )*
b)
Figure 6. Mean amount of time spent per day (±S.E.) on the different behaviour states before and after addition and scattering of browse, all naturalistic enclosures combined. a) shows each behaviour state, b) shows mean time active and inactive (*p<0.05).
5 Discussion
5.1 Effects of enclosure design
The results show that the enclosure design did have effects on the activity budget. When comparing the natural enclosure to the barren, there were some highly significant differences. In the natural enclosures there was more movement, both walking and running than in the more barren environment. As European bison movements are correlated mainly with feeding activity and to ensure optimum use of food supply (Pucek 2004), this would be expected, as the bison in the natural enclosure had pasture and those in barren enclosures were offered food in a way that did not stimulate movement or exploratory behaviour, i.e. in the same way and in a container. This is also noted by Price (1984), who stated that feeding animals a concentrated diet once or twice a day leaves the animal with much “free” time which
otherwise would have been spent on foraging. However, even though the time spent “feeding on supplemental food sources” was significantly higher in the barren enclosures, the total time spent feeding did not differ between the two enclosure types (Fig 2). D’Eath et al. (2009) reviewed the question of quantitative restriction (restricting the quantity of food) versus qualitative restriction (food quality reduced but offered ad libitum) and by comparing food restriction in broiler chickens and in dry sows. They found that the meal patterns seemed more normal over 24 hours and the time spent eating longer and more natural when food quality was restricted. Additionally, the broilers were also more able to initiate and terminate meals. As the feeding routines for the bison in the different enclosures, in addition to offering concentrates (quality food), also offered ad libitum accessibility to hay (quantitative food), the latter appears to fill the “free” time and possibly satisfy the natural, polyphasic feeding behaviour in these bison in both types of enclosures.
There was also a significant difference in the amount of time lying down per day; however, the amount of lying down in natural enclosures could have been increased as the natural enclosures were also larger in size which made the observations during the night troublesome, as the bison sometimes were lying far away and with the night-vision binoculars is was not distinguishable if the bisons were ruminating or not when lying down.
As mentioned above, the result of enclosure design in the different feeding behaviour states is not surprising as the larger enclosures had pasture and the smaller enclosures were barren. Leone’s and Estevez’ (2008) hypothesis that enclosure size, group size and group density have a unique impact on movement and space use in domestic fowl (Gallus gallus
domesticus), was supported by their results. They found that movement patterns in the
domestic fowl are primarily determined by enclosure size i.e. the birds utilized the space given to them. The results of Leone and Estevez (2008) to one extent agree with those of this study, i.e. even if no significant correlation between walking and enclosure size was found, a trend could be distinguished in that bison in larger enclosures walked more. My hypothesis that design has an effect on the activity budget is supported by the results. Additionally, there was a higher activity level (i.e. “walk” and “run”) in natural enclosures as well as in larger enclosures. However, as the natural enclosures also were the largest it was not possible to distinguish whether the result reflected the impact of the size or the nature of each enclosure. The availability of food and water is one of the factors that determine how much space is required by wild animals and the size of home ranges can vary on a seasonal and annual basis (Price 1999). Animals with large home-ranges in the wild are often assumed to need large space in captivity. This may not be entirely true if the size is determined by feeding behaviour and not by an innate need for movement stimulation, although the animals may be motivated to explore (Price 1999) which could be the case of European bison. Van den Brink (1980) made a study in a large enclosure (22000 m2), where he investigated if European bison behaved naturally and secondly to which extent the available space was used. Van den Brink concluded that the size of the enclosure seemed to be somewhat less important, although he pointed out the importance of trees. From that and with the background of what is known from the wild, e.g. movement is mainly connected to feeding and that the activity budget is different during snow conditions in the winter compared to snow-free periods, as a result of supplementary feeding (Cabon-Raczynska et al. 1983), my conclusion is that the activity budget of bison mainly depends on the enclosure design an less on enclosure size.
5.2 Effects of social structure
There are many factors that influence the group size; these factors can be divided into two subgroups: predation risk and resource need such as availability, competition and diet (Pollard and Blumstein 2008) of which resource need is of relevance to this study. In a study by Jørgensen et al. (2009), of ewes housed in two different group sizes (9 and 24 individuals), it
was found that ewes spent more time eating when in large groups. However, in this study no significant difference was found in either of the two eating states between groups of different size. Jørgensen et al. (2008) suggested that the continuous change from small to larger groups and vice versa during grazing make the sheep more tolerant towards new group members. As the group composition in European bison in the wild is very variable, as size and structure depend on season as groups mix and split up into other constellations, they therefore like the sheep should be more tolerant towards new group members. If the group composition has an effect on the activity budget of the European bison, my prediction was that animals in large groups had higher activity level than small groups. The results did not support my prediction as neither “walk” were higher nor “standing” were lower in the larger enclosures. However, it is hard to reach a clear conclusion about how the group size affects the behaviour of this species based on these results. Primarily this is due to the fact that the difference between the group sizes in this study was not large enough, only four individuals separates the largest group from the smallest one. Secondly, the average number of animals in a mixed group in the wild is 8-13 animals which are almost double the group sizes in this study. On the other hand, the size of the home ranges and the amount of food in the wild allows for larger groups than the captive conditions in this study. Optimal group size for captive wild animals is determined by different factors than those facing conspecifics in the wild (Price and Stoinski, 2007). The environmental pressures such as predation and availability of resources are eliminated or controlled in captivity; as a result group sizes are more flexible. But there might be social stress factors, a too large group in a small enclosure where avoidance of dominant conspecifics could be impossible in potential conflict situations (Morgan and Tromborg, 2007).
When comparing bulls to cows no significant differences were found in the overall comparison. A hypothesis has been proposed, the activity budget hypothesis (Mooring et al. 2005; Yearsley and Pérez-Barbería, 2005), which tries to explain the segregation outside the rutting season in ungulates. This hypothesis states that differences in activity, caused by the allometry of energy requirements, may be accentuated by sexual differences in the efficiency of fibre digestion (Yearsley and Pérez-Barbería, 2005). Further, the gender being less efficient at digesting fibre should select more easily digestible foods, which requires more searching time and consequently more time being active. The results of this study do not indicate that this hypothesis can be applied in this case. Additionally, the rejection of the activity budget hypothesis was also made in a study of a close relative to the European bison, the American bison (Bison bison) (Mooring et al. 2005) indicating that this hypothesis might not be applicable for the genus bison. One significant difference was found: cows in barren enclosures ran significantly more than bulls; actually the bulls did not run at all and cows on average 20s during a day. I also found a significant correlation between the amount of time running and age; younger bisons ran significantly more than the older. This could possibly explain the significantly more running of females in barren enclosures as all the bulls were older than eight years and a few of the cows were under the age of four. Relating to previous results, the amount of running was significantly higher in larger enclosures, but the running appeared not to be of the “exploring” kind but more playful behaviour. As the animals in the larger and more natural enclosures had lower mean age, therefore it is difficult to conclude that enclosure design really affects the running behaviour.
5.3 Effects of feeding enrichment
The result showed that the small feeding enrichment from the addition and scattering of browse had a significant effect on several behaviours in the barren enclosures. Although the browse was scattered to the degree possible, no significant increase in walking occurred, but a significant decrease of the total amount of standing was seen. The amount of eating from
natural sources, which the added browse was a part of, was doubled and the eating from supplemental sources was significantly reduced. This could be due to that when given a choice between receiving free food and performing species-appropriate behaviours to acquire the same food, the bison choose to the latter (Chang et al. 1999). In the natural enclosures, both the amount of time standing and the time spent walking decreased significantly. Most of the decrease in the amount of time walking originates from Kolmården Zoo. Due to an incident that occurred between the two observation periods, the bison were housed in their night enclosure (1400 m2) during the period with the browse addition, which seemed to affect their feeding in the reverse way compared to the other enclosures, i.e. increased eating from supplemental food sources and decreased time spent on eating from natural food sources. However, worth noting is that all of the supplemental feeding was presented in the night enclosure during the baseline observation as well as after the adding of browse. Browse on the other hand was presented in both the day and night enclosure By having access to the supplementary feeding at all time could explain the double amount of time spent on feeding from supplemental food and the decrease of time walking in Kolmården Zoo.
When considering the overall change before and after the enrichment, it was a larger effect in the barren enclosures. This could implicate that this kind of enrichments is suitable and effective in a barren enclosure as no larger effect was shown in naturalistic enclosures. Furthermore, scattering the food could give decreased inactivity to relatively low cost. Dishman et al. (2009) performed a feeding enrichment study on captive ring-tailed lemurs (Lemur catta) where the feeding boxes were scattered and extra browse was added to encourage exploratory behaviour. Their result showed that adding browse increased the activity level, which included moving and eating. This as well as my study show that not much is needed to stimulate the animals; sometimes scattering of the food can be enough to stimulate movement and feeding. Britt (1998) states that a reduction in the pooling of food resources and an increase in handling time are both likely to reduce feeding rates and increase the amount of time individuals spend on foraging.
During winter supplemental hay is added to some of the wild populations of European bison (Cabon-Raczynska et al. 1983; Pucek 2004). Supplementary winter feeding results in reduced locomotor activity since the animals in the mixed winter group stay in the immediate vicinity of the feeding places and do not move further away from them than 250-300 meters. Supplementary winter feeding with hay satisfies their food requirements and natural food such as bark and shoot is just supplementary. Therefore, the activity budget of European bison during winter in the wild is very similar to the activity budget in both the natural and barren enclosures in this study. Approximately 60% of daily activity is spent on resting (here lying, standing and ruminating), 30% is spent on feeding and the last 10% is spent on locomotor activity. In the snow-free period the opposite occurs, 60% is spent on feeding and 30% on resting, whereas the amount of movement is the same (Cabon-Raczynska et al. 1987). However, the amount of resting can be underestimated as this study observed the animals from dusk to dawn, excluding the night which is included in this study. These comparisons show that the patterns of activity may be specific, but that distribution of activity phases in time is modified by the season of the year and the feeding system. Therefore a study of the bison from this project during winter conditions might show more similarities between the enclosures types than during the summer.
5.4 Conservation implications
As the rearing environment has proven to influence the success of reintroduction of animals, it is of importance to relate the results in this study to possible implications for the conservation and to the potential for reintroduction of European bison into the wild. The results did show that the captive environment influences the daily activity; however, does this imply that a
bison in one or the other enclosure is more suited for reintroduction? Since the genetic characteristics of all Swedish bison are very similar, the differences between the two enclosure groups indicate that the animals still have the ability to respond appropriately to improved environmental quality. Therefore, it may indicate that the bison are able to respond to the captive environment and hence be susceptible to pre-release habitat training. Most reintroductions have been conducted using “soft” release procedures whereby animals undergo a period of acclimatisation directly pre-release, post-release or a combination of the two (Shepherdson 1994). There is also the “hard” release procedure where there is no pre-release training, and the animals are pre-released directly upon arrival to the pre-release site. Using soft release, the differences in activity between the bison from different enclosure types are likely to diminish over time and the longer the acclimatisation period is. However, few studies have compared the post-release survival of animals reared in different captive environments, and when acclimatisation periods are long this distinction becomes somewhat blurred (Shepherdson 1994). Furthermore, reintroduction success may often be higher with species whose behaviour is instinctive (Snyder et al. 1996) in contrast to animals whose behavioural repertoires are largely learned during their early years (Snyder et al. 1994). Ungulates may require less training in food acquisition and processing in comparison to primates and carnivores that may require more extensive training. Also, gregarious species living in herds may need less social training compared to animals that have a complex social structure (Kleiman 1989). The pairing of captive-born animals with experienced wild-caught individuals in reintroduction can be an effective supplement to training (Kleiman 1989).
In the case of European bison, due to the bottleneck that the species passed through in the early 20th century, there are very few systematic genetic differences. Today, as the populations have high inbreeding, high mean kinship coefficients and skewed founder contribution, it is difficult to find objective genetic criteria to select more or less suitable animals for release. However, the wild population have even higher inbreeding and mean kinship than the captive population, and in some wild populations not all founders are represented. Hence, when considering which of the animals in a captive population should be subject to release, the genetic suitability can be considered first and foremost, with minor consideration to the rearing environment, if a long acclimatisation period is conducted pre-release. Additionally, there are bison living in captivity which are not registered in the EEP programme. Therefore, the European bison EEP should devote more efforts to include these animals into the breeding programme. Furthermore, to develop the most effective pre-release procedures in suitable habitats and to review the management of the wild population by attempting to minimise the supplementary feeding during winter to speed up the adaptation process of European bison into contemporary woodlands (Pucek 2004).
5.5 Conclusion
In conclusion, the results support the hypotheses stated except that there appeared to be no difference between bulls and cows in their activity budget. The animals in the naturalistic enclosures had a higher activity level, i.e. a higher amount of movement and foraging, than those in barren enclosures. It proved difficult to distinguish whether the enclosure type or the size had the larger impact on activity. However, as the movement and home-range in the wild are largely dependent on food availability, enclosure characteristics appear to be the main cause. Future studies could try to assess this by observing animals in small, naturalistic and large, barren enclosures. Further, to investigate if the European bison fully use the space that is provided or mainly use the space near food containers or sites with higher density of trees. Whether group size has an effect on activity budget cannot be determined from this study. The enrichment experiment showed that adding and scattering of browse have a large effect in barren enclosures where the amount of inactivity decreased. Scattering of browse can
therefore be regarded as a cheap and simple way to decrease inactivity and stimulate active foraging in more barren enclosures. The results obtained in this study imply that the European bison in both types of enclosures might be suited for reintroduction providing there is a soft release with a long acclimatisation period pre-release. Future studies in this area can be to investigate whether there is a difference in release success and survival of animals reared in different environments pre-release after a period of acclimatisation. If it would show that the captive rearing environment had a difference in post-release survival, the captive environments and its effect on activity would have larger implications than what I concluded from this study.
6 Acknowledgements
I would like to thank the management and staff at Avesta Visentpark, Borås Djurpark, Skånes Djurpark, Skansen, Lycksele Djurpark and Kolmården for enabling me to do this study and for all the help during the visits. Additionally, I want to thank my supervisor Prof. Mats Amundin for his help during this project and Annika Hofling for all the help and assistance during the data collection.
7 References
Allendorf, F. W. (1993) Delay of adaptation to captive breeding by equalizing family size.
Conservation Biology. 7, 416-419
Beissinger, S. R. (1997) Integrating behaviour into conservation biology: potentials and limitations. In: Behavioural approaches to conservation in the wild. Clemmons, J. R. and Buchholz, R. eds. pp. 23-47. Cambridge University Press, Cambridge.
van den Brink, W. J., (1980) The behaviour of Wisent and Bison in larger enclosures. Acta
Theriologica. 25, pp. 115-130
Britt, A. (1998) Encouraging natural feeding behaviour in captive-bred Back and white ruffed lemurs (Varecia varecia vareigata). Zoo Biology, 17, 379-392
Cabon-Raczynska, K., Krasinska, M., and Krasinski, Z. (1983) Behaviour and daily activity rhythm of European bison in winter. Acta Theriologica. 28, 273-299
Cabon-Raczynska, K., Krasinska, M., Krasinski, Z. A., and Wojcik, J. M. (1987) Rhythm of daily activity and behaviour of European bison in the Bialowieza forest in the period without snow cover. Acta Theriologica. 32, 335-372
Chang, T. R., Forthman, D. L. and Maple, T. L. (1999) Comparison of confined mandrill (Mandrillus sphinx) behaviour in traditional and “ecologically representative” exhibits. Zoo
Biology. 18, 163-176
Daleszczyk, K., Krasinska, M., Krasinski, Z. A. and Bunevich, A. N. (2007) Habitat structure, climatic factors, and habitat use by European bison (Bison bonasus) in Polish and Belarusian parts of the Bialowieza forest, Poland. Can. J. Zool. 85, 261-272
D’Eath, R. D., Tolkamp, B. J., Kyriazakis, I. and Lawrence, A. B., (2009) “Freedom of hunger” and preventing obesity: the animals welfare implications of reducing food quantity or quality. Animal Behaviour. 77, 275-288
Dishman, D. L., Thomson, D. M. and Karnovsky, N. J. (2009) Does simple feeding
enrichment raise activity levels of captive ring-tailed lemurs (Lemur catta)? Applied Animal
Behaviour Science. 116, 88-95
EAZA 2008. EAZA, Breeding programmes – overview EEPs/ESBs. http://www.eaza.net/ (Accessed 15 May 2009)
Håkansson, J. and Jensen, P. (2005) Behavioural and morphological variation between captive populations of red junglefowl (Gallus gallus) – possible implications for conservation.
Biological Conservation. 122, 431-439
Jaman, M. F. and Huffman, M. A. (2008) Enclosure environment affects the activity budgets of captive Japanese macaques (Macaca fuscata). American Journal of Primatology. 70, 1133-1144
Jørgensen Meisfjord, G. H., Andersen, I. L., Berg, S. and Boe, K. E. (2009) Feeding, resting and social behaviour in ewes housed in two different group sizes. Applied Animal Behaviour
Science. 116, 198-203
Kleiman, D. G. (1989) Reintroduction of captive mammals for conservation. BioScience. 39, 152-161
Krasinska, M., Krasinski, Z. A. and Bunevich, A. N. (2000) Factors affecting the variability in home range size and distribution in European bison in the polish and Belarussioan parts of the Bialowieza forest. Acta Theriologica. 45, 321-334
Lee, J. S. F. and Berejikian, B. A. (2008) Effects of the rearing environment on average behaviour and behavioural variation in steelhead. Journal of Fish Biology. 72, 1736-1749 Leone, E. H., and Estevez, I., (2008) Use of space in the domestic fowl: separating the effects on enclosure size, group size and density. Animal Behaviour. 76, 1673-1682
Mallapur, A. and Chellam, R. (2002) Environmental influences on stereotypy and the activity budget of Indian leopards (Panthera pardus) in four zoos in southern India. Zoo Biology. 21, 585-595
Maple, T. L. and Perkins, L. A. (1996) Enclosure furnishings and structural environmental enrichment. In Wild mammals in captivity – principles and techniques, eds D. G. Kleiman, M. E. Allen, K. V. Thompson and S. Lumpkin, pp. 212-222. University of Chicago Press, Chicago
Mignon-Grasteau, S., Biossy, A., Bouix, J. Faure, J-M., Fisher, A. D., Hinch, G. N., Jensen, P., Le Neindre, P., Mormède, P., Prunet, P., Vandeputte, M. and Beaumont, C. (2005) Genetics of adaptation and domestication in livestock. Livestock Production Science. 93, 3-14 Miller, K. A. and Mech, J. A. (2005) The differential effects of four types of environmental enrichment on the activity budgets, fearfulness and social proximity preference of Japanese quail. Applied Animal Behaviour Science. 95, 169-187
Mooring, M. S., Reisig, D. D., Osbourne, E. R., Kanallakan, A. L., Hall, B. M., Schaad, E. W., Wiseman, D. S. and Huber, R. R. (2005) Sexual segregation in bison: a test of multiple hypotheses. Behaviour. 142, 897-927
Morgan, K. N., and Tromborg, C. T. (2007) Sources of stress in captivity. Applied Animal
Behaviour Science. 102, 262-302
Newberry, R. C. (1995) Environmental enrichment: Increasing the biological relevance of captive environments. Applied Animal Behaviour Science. 44, 229-243
Olech, W. 2008. Bison bonasus. In: IUCN 2008. 2008 IUCN Red List of Threatened Species. http://www.iucnredlist.org/details/2814 Downloaded on 10 February 2009
Pollard, K. A., and Blumstein, D., T. (2008) Time allocation and the evolution of group size.
Animal Behaviour. 76, 1683-1699
Price, E. O. (1984) Behavioural aspects of animal domestication. The quarterly review of
biology. 59, 1-32
Price, E. O. (1999) Behavioral development in animals undergoing domestication. Applied
Animal Behaviour Science. 65, 245-271
Price, E. E., and Stoinski, T. S. (2007) Group size: Determinants in the wild and implications for the captive housing of wild mammals in zoos. Applied Animal Behaviour Science. 103, 255-264
Pucek, Z. (ed.); Pucek, Z., Belousova, I.P., Krasiñska, M., Krasiñski, Z.A. and Olech, W. (comps.). (2004). European Bison. Status Survey and Conservation Action Plan. IUCN/SSC Bison Specialist Group. IUCN, Gland, Switzerland and Cambridge, UK. ix + 54 pp.
Shepherdson, D. (1994) The role of environmental enrichment in the captive breeding and reintroduction of endangered species. In: Creative Conservation; Interactive management of
wild and captive animals. Olney, P. J., Mace, G. and Feistner, A. (eds.) Chapman & Hall,
London
Snyder, N. F. R., Koenig, S. E., Koschmann, J., Snyder, H. A. and Johnson, T. B. (1994) Thick billed Parrot releases in Arizona. Condor. 96, 845-862
Snyder, F. R., Derrickson, S. R., Beissinger, S. R., Wiley, J. W., Smith, T. B., Toone, W. D. and Miller, B. (1996) Limitations of captive breeding in endangered species recovery.
Conservation Biology. 10, 338-349
Yearsly, J. M. and Pérez-Barbería, F. J. (2005) Does the activity budget explain sexual segregation in ungulates. Animal Behaviour. 69, 257-267
