@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Land and Food Systems, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Corti, Paulo"@en ; dcterms:issued "2009-07-27T19:16:26Z"@en, "2001"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Four hypotheses have been proposed to explain sexual segregation in sexually dimorphic ungulates. I tested two of these on a Dall's sheep (CV/'s dalli dalli) population in Kluane National Park in Yukon, Canada. In the first, the reproductive-strategy hypothesis, males are predicted to feed in the best foraging areas to enhance their condition for intrasexual competition. Females are predicted to use areas with lower predation risk to raise offspring. In the second, the sexual dimorphism-body size hypothesis, females should use the best forage areas to satisfy the nutritional demands of gestation and lactation, and males due to their greater absolute metabolic requirements and larger body size have to forage on more available forages, but lower quality. I found Dall's sheep had a high social segregation but limited habitat segregation between males and females. Males were further from security cover in more gentle terrain than were maternal groups that used cliff or talus slopes. Maternal groups were located at higher altitudes than were males because most security cover was close to mountain peaks. Lamb presence was a factor increasing predation risk and affecting maternal groups' behaviour and distribution towards security cover. Group distance from security cover was negatively correlated with the proportion of individuals lying down for maternal groups without lambs. When this group type left security cover they were constantly active, either feeding or moving. The forage density index (FDI) values varied significantly, with the areas used by males having higher FDIs than areas used by maternal groups. Nutritional components were similar, but fibre content was significantly higher in areas used by males. Signs of wolves (Canis lupus) and predation events on Dall's sheep were recorded only at male sites. These data support the idea that male areas have a greater risk of predation. My results primarily support the reproductive-strategy hypotheses, indicating that predation plays a key role in the development of sexual segregation in Dall's sheep. However, I also found evidence to support one prediction of the sexual dimorphism-body size hypothesis, where males use areas higher in forage availability but lower in quality than female areas."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/11266?expand=metadata"@en ; dcterms:extent "1945767 bytes"@en ; dc:format "application/pdf"@en ; skos:note "DALL'S S H E E P (Ovis dalli dalli Nelson, 1884) S E X U A L S E G R E G A T I O N INTERACTIONS B E T W E E N TWO H Y P O T H E S E S by PAULO CORTI L .V .Sc , University Austral of Chile, 1996 D.V.M., University Austral of Chile, 1996 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE D E G R E E OF MASTER OF S C I E N C E in THE FACULTY OF G R A D U A T E STUDIES (Department of Animal Science) We accept this thesis as conforming ^o^fer^flr/ed standard THE UNIVERSITY OF BRITISH COLUMBIA December 2000 © Paulo Corti, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Arytnr\\CtJ ^ C \\ a n c e The University of British Columbia Vancouver, Canada Date \"Decevvxloer 2.1 , Zooo II ABSTRACT Four hypotheses have been proposed to explain sexual segregation in sexually dimorphic ungulates. I tested two of these on a Dall's sheep (CV/'s dalli dalli) population in Kluane National Park in Yukon, Canada. In the first, the reproductive-strategy hypothesis, males are predicted to feed in the best foraging areas to enhance their condition for intrasexual competition. Females are predicted to use areas with lower predation risk to raise offspring. In the second, the sexual dimorphism-body size hypothesis, females should use the best forage areas to satisfy the nutritional demands of gestation and lactation, and males due to their greater absolute metabolic requirements and larger body size have to forage on more available forages, but lower quality. I found Dall's sheep had a high social segregation but limited habitat segregation between males and females. Males were further from security cover in more gentle terrain than were maternal groups that used cliff or talus slopes. Maternal groups were located at higher altitudes than were males because most security cover was close to mountain peaks. Lamb presence was a factor increasing predation risk and affecting maternal groups' behaviour and distribution towards security cover. Group distance from security cover was negatively correlated with the proportion of individuals lying down for maternal groups without lambs. When this group type left security cover they were constantly active, either feeding or moving. The forage density index (FDI) values varied significantly, with the areas used by males having higher FDIs than areas used by maternal groups. Nutritional components were similar, but fibre content was significantly higher in areas used by males. iii Signs of wolves (Canis lupus) and predation events on Dall's sheep were recorded only at male sites. These data support the idea that male areas have a greater risk of predation. My results primarily support the reproductive-strategy hypotheses, indicating that predation plays a key role in the development of sexual segregation in Dall's sheep. However, I also found evidence to support one prediction of the sexual dimorphism-body size hypothesis, where males use areas higher in forage availability but lower in quality than female areas. iv T A B L E O F C O N T E N T S A B S T R A C T ii T A B L E O F C O N T E N T S iv L IST O F T A B L E S v L IST O F F I G U R E S vi A C K N O W L E D G E M E N T S vii I N T R O D U C T I O N 1 M E T H O D S 7 STUDY A R E A AND SHEEP 7 TIME PERIOD 7 SAMPLING GROUP BEHAVIOUR, PREDATION, AND HABITAT 8 Group behaviour 8 Sheep behaviour 9 Definitions of behavioural variables 9 Predation risk 11 Measuring the degree of social and habitat segregation 11 Habitat characteristics by group type 13 VEGETATION SAMPLING 13 Percent of cover and forage density 13 Vegetation quality 15 STATISTICAL ANALYSES 16 Sheep data 16 Habitat and vegetation data 17 R E S U L T S 18 SEXUAL SEGREGATION AND SPATIAL DISTRIBUTION 18 GROUP CHARACTERISTICS AND SEXUAL SEGREGATION 21 HABITAT SELECTION BY GROUP TYPE 24 PREDATOR PRESENCE AT HOGE PASS 25 D I S C U S S I O N 28 R E F E R E N C E S 35 V LIST OF TABLES Table 1. Criteria used to assign a social group type to observations of Dall's sheep groups 8 Table 2. Summary of ranges for male and maternal groups at two locations in Hoge Pass during early summer 25 Table 3. Summary of vegetation and physiogeographic variables measured at feeding sites used by male and maternal groups at Area 1 at Hoge Pass in early summer 27 vi LIST O F FIGURES Figure 1. Sampling design used to measure the proportion of vegetation species in feeding sites of males and females at Area 1 14 Figure 2. Relative frequency distribution of all Dall's sheep groups as function of the percentage of adult males present in groups (class III - IV) 18 Figure 3. Distance from security cover for different group types of Dall's sheep.... 19 Figure 4. Altitude used by different types of Dall's sheep groups 20 Figure 5. Group size by different types of Dall's sheep groups 21 Figure 6. Relative frequency distribution of the sheep groups and the total number of animals in relation to group size 22 Figure 7. Relative frequency distribution of the proportion of sheep lying down in groups as function of the distance to security cover 23 Figure 8. Proportion of lambs in maternal groups observed at different distances from security cover 24 VII ACKNOWLEDGEMENTS I thank: David Shackleton for his supervision and opportunity to work with him, and my committee, Daniel Weary and Greg Henry for help and corrections; Alton Harestad, Tina Buijs, and Scott Harrison for improving and commenting on my drafts; Alex Frid for advice and discussions in the field; Alan Norquay for helping with the mapping analysis; Kathreen Ruckstuhl for advice on mountain sheep behaviour; the Government of Yukon Department of Renewable Resources through Jean Carey for the opportunity to be part of its project in Kluane; Alex Frid and Carol Domes for assistance in the field; Parks Canada Yukon through Ray Breneman and Bruce Sundbo for park material facilitation and logistic information; Jim Shelford, Gilles Galzi, and Siva Kumarsingham for comments and help with nutritional analyses at Animal Science Laboratory, Faculty of Agricultural Sciences, UBC; Paul Alaback for comments on plant sampling; and Michael Pitt and Valerie Lemay for statistical advice. The Geographic Information Service UBC helped with aerial photograph material. I give special thanks to Tina for correcting my English and for her patience, and to my parents, Dante Corti and Mabel Gonzalez, to whom I dedicate this work. 1 INTRODUCTION In sexually dimorphic ungulate species, the adults associate extensively only during the rut or mating season, remaining segregated the rest of the year. This sexual segregation is particularly pronounced during the season of birth and early development of the young (Festa-Bianchet, 1988; Main, Weckerly & Bleich, 1996; Bleich, Bowyer & Wehausen, 1997). The main hypotheses that have been proposed to explain sexual segregation in ungulates have been reviewed in several works (i.e., Main & Coblentz, 1990; Miquelle, Peek & Van Ballenberghe, 1992; Main ef al., 1996; Bleich et al., 1997). These hypotheses generally suggest that this phenomenon results from differing energetic and reproductive strategies of the sexes (reviewed in Main et al., 1996), and four possible explanations have emerged. The first hypothesis, the \"reproductive-strategy hypothesis\", suggests that ecological factors determine sexual segregation (Festa-Bianchet, 1988; Main & Coblentz, 1990; 1996; Main ef al., 1996; Miquelle ef al., 1992; Bleich ef al., 1997). The explanation proposed is that females increase their Darwinian fitness through increased offspring survival, by selecting habitats with a lower predation risk for offspring, which also provide sufficient resources for lactation. Males on the other hand, increase their fitness by maximising body and weapon growth, which ultimately leads them to winning more fights and securing more mates. To achieve this end they use habitats with abundant food resources, even if there is a higher predation risk in such areas (Berger, 1991; Main et al., 1996; Bleich et al., 1997). The second, the \"sexual-dimorphism-body-size hypothesis\", proposes that factors related to physiological and morphological aspects of nutrition are responsible for sexual segregation (Clutton-Brock, Guinness & Albon, 1982; 2 Clutton-Brock, lason & Guinness, 1987; Beier & McCullough, 1990; Weckerly, 1993; reviewed in Main etal., 1996; Perez-Barberia & Gordon, 1998). The explanation is that because males are larger, they have greater absolute metabolic requirements than females. Consequently, males must seek areas with abundant forage. Further, to meet their nutritional requirements and forage efficiently, males must ingest larger amounts of forage per bite, even if such food has lower quality (higher fibre content). Because of their larger body size, males have lower relative metabolic requirements than do females, and can retain digesta longer to improve digestion efficiency (Barboza & Bowyer, 2000). By contrast, females can forage more efficiently than males because their narrower incisor arcade allows them to be more selective foragers. However, because of their smaller body size, females need higher quality foods to satisfy their relatively higher nutritional requirements than the males (Clutton-Brock etal., 1982, 1987; reviewed in Main etal., 1996). Also, unlike males, females are faced with the high demands of late gestation and especially of lactation. The \"social hypothesis\" postulates that social factors, especially male behaviours, lead to sexual segregation. Such factors can include interactions with same-sex peers, in particular the males' need to learn fighting skills, to evaluate potential rivals, and to establish dominance relationships. These same-sex interactions lead to different preferences of the two sexes (Villaret & Bon, 1995; Bon & Campan, 1996; reviewed in Main etal., 1996; Cransac etal., 1998). Although this hypothesis might explain sexual segregation, it does not explain how the segregation would lead to sex differences in habitat use that are commonly observed. 3 Recently, researchers suggested that sexual segregation may be driven purely by differences in the activity budgets of the adult sexes because they differ in size (Conradt, 1998a; Ruckstuhl, 1998, 1999; Ruckstuhl & Neuhaus, 2000), and not by sex differences in habitat requirements (Conradt, 1999). Body size differences which affect food requirements of the sexes (see \"sexual dimorphism-body size hypothesis\" above), should cause females to spend more time foraging and walking in search of higher quality food, than do males. These differences in foraging time make it difficult for the sexes to synchronise their movements and hence maintain cohesive groups (Conradt, 1998a; Ruckstuhl, 1998). This fourth hypothesis has been termed as the \"activity-budget hypothesis\" (Ruckstuhl, 1998; Ruckstuhl & Neuhaus, 2000). The first and the second hypotheses seem to be not mutually exclusive, because each is based on intersexual differences in reproductive strategy as they relate to energetics and security for offspring (reviewed in Main et al., 1996). But, both are distinct from the social interaction hypothesis, because the latter depends solely on social mechanisms and does not invoke differences in use of habitat as the primary cause. Also, while the \"social factors hypothesis\" provides a possible proximate mechanism for segregation and the \"activity-budget hypothesis\" for why the sexes may be unable to maintain cohesive groups, neither explains why the sexes should use different habitats. The objective of my study was to investigate the reproductive-strategy hypothesis (Main & Coblentz, 1990; Miquelle et al., 1992; Bleich ef al., 1997), and the sexual dimorphism-body size hypothesis (Clutton-Brock ef al., 1982, 1987; Beier & McCullough, 1990; Weckerly, 1993; Perez-Barberia & Gordon, 1998) on a Dall's sheep population during late spring and early summer. I decided to concentrate my 4 efforts on these two hypotheses, because among the several possible factors that might produce sexual segregation in ungulates, differences in energetic requirements, resources partitioning, and predation risk by sex have received the most support (i.e., Miquelle etal., 1992; Main etal., 1996; Bleich etal., 1997). This focus also allowed me to carry out my research within a reasonable period, unlike the \"social factors hypothesis\" which requires a study of several years fieldwork. My focus also fits with broader issues in ungulate biology because predation risk is considered a major factor affecting the evolution of ungulates in terms of social organisation, foraging behaviour, and morphology (Jarman, 1974; Geist, 1971; Molvar & Bowyer, 1994; Bleich et al., 1997). According to foraging theory, only the energetic benefit-time trade-off is important in an animal's decision to forage in specific areas (e.g., Stephens & Krebs, 1986). However, more recent studies have begun to investigate the importance of higher level trade-offs that cause some animals to forage in areas of high predation risk or to keep others away from areas with high food values (Houtman & Dill, 1998). Dall's sheep (Ovis dalli dalli) is a Caprinae species that segregates during spring and summer (Rachlow & Bowyer, 1998; Bowyer, Leslie & Rachlow, 2000). This species of mountain sheep is well-suited for studying sexual selection hypotheses because they have pronounced sexual dimorphism (females are approximately 60% of males body weight) and secondary sexual characteristics (Bowyer & Leslie, 1992). Also, they live in areas with high predator numbers and low human impact. The factors causing sexual segregation in this species have not been directly investigated. Researchers have assumed that the behaviour of Dall's sheep is similar to that of bighorn sheep (O. canadensis) (Bowyer et al., 2000), which has been studied more thoroughly (e.g., Geist & Petocz, 1977; Festa-5 Bianchet, 1988; Bleich etal., 1997; Ruckstuhl, 1998). Some aspects of the ecology of Dall's sheep, including the impacts of predation, have been documented in Alaska as well as in the Kluane region of Yukon (e.g., Murie, 1944; Hoefs & Cowan, 1979; Hoefs & Bayer, 1983; Nette, Buries & Hoefs, 1984; Sumanik, 1987; Frid, 1997; Nichols & Bunnell, 1999; Bowyer ef al., 2000). Dall's sheep find security from predators on cliff and talus slopes (Nichols & Bunnell, 1999; Bowyer et al., 2000) like many other Caprinae (Geist, 1971; Schaller, 1977). Based on both hypotheses, and assuming the possible ecological and physiological factors leading to sexual segregation on Dall's sheep, I made the following predictions about the behaviour of sheep in my study area: 1. Males have greater absolute metabolic requirements than do females and their habitat selection is primarily driven by food availability. Therefore, areas with high density of suitable forages are more likely to contain males. 2. Because security cover is usually highly localised and is used by all sheep at some time, there should be less forage in and near to security cover, because of greater foraging intensity than elsewhere. Consequently there should be less forage available near security and so sheep further from security cover will more likely be males because they need large quantities of food. 3. At least in spring and summer, females select habitats primarily driven by avoidance of predation risk to enhance offspring survival because young are the most vulnerable age class to predators. Therefore, areas with a high proportion of security cover (i.e., cliff and talus slopes) are more likely to contain females. Males, because of their larger body size, and resulting lower relative energetic requirements and greater gut capacity, are expected to spend more time ruminating and resting (lying down) than females. 7 METHODS Study Area and Sheep I collected data at Hoge Pass (61°19'N, 139°33'W), in eastern Kluane National Park Reserve, southwestern Yukon, Canada. The study site is part of the Donjek Mountain Range, which is isolated and rugged. Hoefs, Cowan & Krajina (1975) have described the phytogeography in the eastern area of Kluane, which has a semi-arid and continental climate. My site differed from those described by Hoefs et al. (1975) in that it was comprised mainly of alpine grassland without shrubs or tree cover. Cliff and talus slopes were the main components of the mountain summits. Large meadows occurred at the bases of steep and rocky terrain. For logistic and sampling reasons, I divided the total study site into two areas, Area 1 north of Hoge Creek and Area 2 south of Hoge Creek. I did this because Area 2 was difficult to access for vegetation sampling requiring at least a day hike to reach it. Giroux (1991) reported that the Dall's sheep population of the Donjek Range in 1991 totalled 682 animals, of which 181 were rams, 432 were ewes in maternal groups, and 69 were young of the year. However, this population may have diminished during 1999 because of a harsh winter season (A. Frid, pers. comm.). The estimation of the population size during the study period was approximately 90 males, 50 females, and 10 lambs. Time Period My fieldwork occurred from June 21 s t to July 15 t h of 1999, which encompasses the lactation period, a critical time in the annual cycle of female Dall's sheep (Bunnell, 1982; Rachlow & Bowyer, 1994; Rachlow & Bowyer, 1998; Bowyer et al., 2000). Also, sexual segregation in mountain sheep (Gv/'s spp.) is most 8 pronounced during lambing and lactation (Festa-Bianchet, 1988; Bleich ef al., 1997; Rachlow & Bowyer, 1998). The median of the birth season occurs in late spring, from around 18 to 27 of May; with lactation occurring throughout most of the summer following the birth period (Bunnell, 1982; Rachlow & Bowyer, 1994; Rachlow & Bowyer, 1998). Sampling Group Behaviour, Predation, and Habitat Group behaviour I arbitrarily considered a group to be a set of individuals present on the same aspect of the same slope, and no further than 100 m from each other {sensu Frid, 1997). Group type was classed as either \"male\" (adult and sub-adult males) (Geist, 1971) or \"maternal\" if 80% of the members were the same sex (excluding lambs and yearlings), and \"mixed\" if they were not (Table 1). Further, I recognised two types of maternal groups: those with and those without lambs. Table 1. Criteria used to assign a social group type to observations of Dall's sheep groups. Social group Group composition criteria Maternal with lambs Females > 80% of total individuals in a group. Lambs present. Yearlings may or may not be present. Maternal without lambs Females > 80% of total individuals in a group. No lambs present. Yearlings may or may not be present. Male Adult (class III - IV, 6 years old or older) and sub-adult (class I - II, 2.5-6 years old) males > 80% of a total group. Mixed Adult males + maternal groups Sheep behaviour I observed sheep during 21 morning surveys along a fixed, 6-km transect. Surveys consisted of walking the transect in search of sheep through an area of approximately 3,600 ha in Hoge Pass. I established observation points in areas where the scan width was widest and that enabled a good view of the surrounding mountains. I used a spotting scope (20-40 x 60) and binoculars (8 x 30). Minimum distance from sheep was 100 m, with not observed impact. I anticipated some degree of pseudoreplication in my observations because no animals were marked. To minimise its effects, I made one survey each day, to avoid recording the same group more than once a day (Machlis, Dodd & Fentress, 1985). Definitions of behavioural variables I used scan group sampling (Altmann, 1974; Martin & Bateson, 1993), to record the following variables for each group: distance to security cover, estimated visually in sheep torso-lengths from centre of group to the edge of nearest cover (cliff or talus slopes) then corrected using the topographic map; group size; group composition; and proportion of animals lying down. The location and altitude of sheep groups in the landscape were mapped on a 1:50,000 topographic map based on the centre of the group. Proximity to cliff and talus slopes provides security cover for mountain sheep because they are able to move in these areas more easily than most of their terrestrial predators (Geist, 1971; Bleich etal., 1997; Frid, 1997; Bleich, 1999; Nichols & Bunnell, 1999). Cliff slopes in general are very precipitous (slopes > 30°) and hazardous for predators of sheep, with lower quantities of forage because they 10 are primarily bedrock (Bleich etal., 1997; Frid, 1997; Rachlow & Bowyer, 1998; Bleich, 1999; Nichols & Bunnell, 1999). I used the number of sheep, other than lambs, to define group size. Young of the year were excluded from measures of group size because ungulates in this age class appear to recognise potential threats less readily than do older conspecifics (FitzGibbon & Lazarus, 1995; Bleich, 1999). Predation risk and distribution of forage are known to affect group size (Jarman, 1974) because the group reduces the risk of predation for individual sheep, such as through the dilution effect and higher predator detection (i.e., vigilance) (Jarman, 1974; Berger, 1978; Risenhoover& Bailey, 1985; FitzGibbon, 1990; Fox, Sinha & Chundawat, 1992; Molvar & Bowyer, 1994; Roberts, 1996; Frid, 1997). Composition of groups directly affects their predation risk. Male groups are less vulnerable to predator attacks than are maternal groups because of their larger body size (Festa-Bianchet, 1988; Bleich et al., 1997; Bleich, 1999). Maternal groups are especially vulnerable when females are accompanied by young, due to limited escape behaviour of offspring, particularly during the young's first days of life (Shackleton & Haywood, 1985; Berger, 1991; Frid, 1997; Rachlow & Bowyer, 1998). Ruminants that are not lying down (sleeping, resting, or ruminating) are generally either feeding or walking. This is related to differences in time budgets among group types, because the time that animals spend walking-feeding or lying down is related to the differences in body size and relative energetic requirements (Conradt, 1998a; Ruckstuhl, 1998, 1999). 11 Predation risk Several potential predators of adult Dall's sheep are found in Kluane: Coyotes {Canis latrans), wolverines {Gulo gulo), grizzly bears (Ursus arctos), lynx (Lynx canadensis), and wolves {Canis lupus) (Nichols & Bunnell, 1999). Wolves are considered the main predator of Dall's sheep in Kluane (Sumanik, 1987) and Denali (Murie, 1944) Parks, but Hoefs and Bayer (1983) considered coyote as the main predator at Sheep Mountain, in southeastern Kluane. New-borns may be taken by golden eagles (Aquila chrysaetos) (Nette et al., 1984) and red foxes {Vulpes vulpes) (Hoefs & Cowan, 1979). For my study, predator presence was recorded (type and location) if I observed predators or their signs (i.e., tracks and scats). Measuring the degree of social and habitat segregation The degree of social segregation was measured through the \"segregation coefficient\" (SC) developed by Conradt (1998b). The measure of habitat segregation is analogous to the SC measurement and uses the same interpretation (see Conradt, 1998b). I did not calculate a quantitative measure of spatial segregation because the size of the sample unit has a considerable effect on the degree of sexual segregation calculated; also when unit size is arbitrarily chosen it may lead to unreliable results (see Bowyer, Kie & Van Ballenberghe, 1996). The degree of social segregation was calculated using the following equation (after Conradt 1998b): SCsodai =1 - (NIX • Y ) . I k i = 1 [(x, • yd I (n, - 1)], where: x, is the number of males (adults and subadults) in the /th group; y is the number of females /th group; n, is the group size of the /th group (n, = x, + y); k is the 12 number of groups with at least two animals; X is the total number of males sampled (excluding solitary animals); V i s the total number of females sampled (excluding solitary animals); and N is the total number of males and females sampled during the study period (N = X+ Y). SCsodai takes a value of almost 0 in the case of no segregation and of 1 in the case of complete segregation. SCS0Ciai is independent of sex ratio, population density and group sizes (for more details see Conradt, 1998b). The degree of habitat segregation is obtained by the following equation (after Conradt, 1998b): SCnabitat =1 - (M / Z • W) * Z' / =f [(*/ ' W,) I (m, - 1)], where: z, is the number of males (adults and subadults) in the /th habitat type; w, is the number of females in the /th habitat type; m, is the number of males and females in the /th habitat type {m, = z,- + w,-); / is the number of habitat types which are used by at least two animals; Z is the total number of males sampled; W is the total number of females sampled; and M is the total number of males and females sampled (M = Z + W). Two habitat types were recognised that sheep groups used, security cover (cliff and talus slopes) and alpine grasslands. Like SC social, the expected value of SChabitat is stochastically independent of sex ratio, population density, and the density of use of various habitat types, including solitary animals if they overlap in habitat use with others (Conradt, 1998b). One important advantage of S C is that the degrees of social segregation and habitat segregation are directly comparable (Conradt, 1998b). 13 Habitat characteristics by group type I used all locations of sheep groups over the sampling period separated by group-type (male and maternal groups) to build habitat ranges for each group type. Using the software package \"California Home Range\" (CALHOME). C A L H O M E is normally used to determine utilisation distributions of single animal locations (Kie, Baldwin & Evans, 1994), but I applied the same procedure to group types using the centre of the group as the location. I calculated the probable 100 % minimum convex polygon (MCP) used by the different group types during the study. I also calculated the overlap proportion of M C P by the sexes in Areas 1 and 2, comparing the shared range with the total male and female ranges per area. I used a 1:62,000-scale air photo (September 1955, number A 14960 - 11) overlain with and a Universe Transverse Mercator (UTM) grid (cell size of 250 m2) to estimate the proportions of security cover and grasslands per hectare in the male and maternal areas, by counting the number of grid cells containing each of the two habitat types (Brower & Zar, 1979). I used the proportion of cliff and talus slopes as an index of security cover for mountain sheep (Bleich et al., 1997; Frid, 1997; Bleich, 1999; Nichols & Bunnell, 1999). Estimates of range areas were corrected for slope using a standard method (Geography 101, UBC). Vegetation Sampling Percent of cover and forage density Plots measuring 18 m x 30 m were located at the central point where a sheep group had been observed at least 3 days in a row at the Area 1 (n = 11 for maternal, and n = 13 for male groups), with four line transects and six Daubenmire quadrats per line (see Fig. 1). In each of the Daubenmire quadrants or subplots, I visually 14 estimated the proportion of plant species or types (i.e., graminoids and forbs), the proportion of the bare soil (ground or rock), and the proportion of dead vegetation (Daubenmire, 1959; in Alaback, 1986). Four of the Daubenmire quadrats were randomly selected, and the vegetation was clipped at ground level. I stored and air-dried the clippings in paper bags for vegetation quality analysis (see \"Vegetation quality\" below). Line transects (4.5 m apart) Daubenmire quadrats (20 cm x 50 cm) (5 m apart) 30 m ^ 18 m ^ Figure 1. Sampling design I used to measure the proportion of vegetation species in male and female feeding sites at Area 1. 15 I calculated a \"forage density index\" (FDI) using the biomass (air-dried weight) obtained from the clipped subplots and the proportion of grassland counted in the air photo, obtained from the range estimates for each group type at Area 1: FDI = (Biomass x Grassland proportion) / 100 I also measured landscape characteristics at each site and recorded physiogeographic variables such as geographic location, elevation, aspect, and slope using a Global Positioning System (GPS), altimeter, and a compass-clinometer. Vegetation quality I chose three standard laboratory methods for estimating forage quality: 1. Dry matter (DM) is considered a precise method for determining the standard amount of nutrients in a herbivore's diet (Robbins, 1983). This method standardises the nutrients content for comparisons and also shows the total content of solids in vegetation samples. 2. Crude protein (CP) is an indirect method for estimating the amount of protein based on the forage nitrogen content (Van Soest, 1982; Robbins, 1983). Because proteins are characteristically comprised of 16% nitrogen, the nitrogen content multiplied by 6.25 and termed the crude protein content (Robbins, 1983). The amount of crude protein in vegetation is positively correlated with forage quality (Van Soest, 1982). 3. Neutral detergent fibre (NDF) is a rapid method for estimating the total fibre in plant feed stuffs (Van Soest, Robertson & Lewis, 1991). The amount of fibre in 16 forage is negatively correlated with quality because the higher the NDF level the lower the digestibility (Robbins, 1983). Statistical Analyses All data were tested for normality through normal probability plots and Shapiro-Wilk's test (Zar, 1999), and homoscedasticity with Levene's test, using the S P S S 9.0 statistical package (SPSS, 1997). All hypotheses were tested with a level of significance of a = 0.05. Behavioural variables were not normally distributed and could not be successfully transformed; therefore, they were compared using non-parametric tests (Martin & Bateson, 1993). Sheep data The Kruskal-Wallis H-test (corrected for ties) was used for multiple comparisons to test significant differences in the means (Zar, 1999). If the Kruskal-Wallis /-/-test revealed a significant difference between group types for a given behavioural variable, each group was compared to the others using the non-parametric post-hoc Dunn's test (corrected for ties) (Zar, 1999). Correlations among behavioural variables for each group type were tested with the Spearman's rank correlation coefficient and only the significant ones were presented (Zar, 1999). The Kolmogorov-Smirnov Z-test was used to compare distributions (Zar, 1999). 17 Habitat and vegetation data Due to small sample sizes, the Mann-Whitney l/-test was use to compare male and maternal feeding sites in Area 1, in terms of vegetation species, physiogeographic data, and FDI (forage density index) (Zar, 1999). 18 R E S U L T S Sexual Segregation and Spatial Distribution A total of 209 group observations were sampled during the study period, and classed as male (n = 122), maternal with (n = 46) and without lambs {n = 40), and one mixed group. The single mixed group recorded was not considered for any analyses due to their small sample size, with the exception of the degree of social and habitat segregation where the totality of the population is considered (after Conradt, 1998b). I observed a high degree of social segregation between males and maternal groups (SC S 0 C /a / = 0.91), but there was a lower degree of habitat segregation (SChabitat = 0.31). The distribution of proportion of adult males class III -IV in groups shows that this male group type formed totally separate groups from maternal groups, and it was significantly different from a uniform distribution (Kolmogorov-Smirnov Z-test, Z = 3.146, n = 209, p < 0.001) (Fig. 2). 50 w 40 o. 3 O o> 30 TJ I 2 0 .a o 10 -i F=\\ F = l 1 P 0.05), (Fig. 3). 600 Males (n=122) Maternal+lamb (n=46) Maternal-lamb (n=40) Group category Figure 3. Distance from security cover for different group types of Dall's sheep observed at Hoge Pass. Box-plots represent the interquartile range, which contains the 50% of values. The whiskers represent the highest and lowest values containing the 80% of values, excluding outliers (o) and extreme values (+). The line across the box indicates the median. Different letters indicate significant differences among groups. 20 Male groups were observed at significantly lower altitudes (x = 1759.85 ± 133.4 m a.s.l.) than were maternal with lamb (x = 1880.2 ± 93.4 m a.s.l.), and maternal without lamb groups (x = 1866.7 ± 113.0 m a.s.l.) (%2 = 41.545, df= 2, p < 0.001). Both maternal with lamb and maternal without lamb groups were at similar altitudes (Q = 0.693, k = 3, p > 0.05), (Fig. 4). 2200 2000 ns 1800H CD 3 1600 140O 1200. 8 O Males Maternal+lamb (n=122) (n=46) Group category Maternal-lamb (n=40) Figure 4. Altitudes used by different types of Dall's sheep groups observed at Hoge Pass. Box-plots represent the interquartile range, which contains the 50% of values. The whiskers represent the highest and lowest values containing the 80% of values, excluding outliers (o) and extreme values (+). The line across the box indicates the median. Different letters indicate significant differences among groups. 21 Group Characteristics and Sexual Segregation Group classes were not significantly different in number of individuals (male groups: x = 8.78 ± 8.11, maternal with lamb groups: x = 7.57 ±8.12, maternal without lamb groups: x = 5.55 ± 7.45) (x 2 = 3.863, df= 2, p = 0.145) (Fig. 5). However, male group size was significantly positive correlated to distance from security cover (rs = 0.242, n = 122, p = 0.004). Males Maternal+lamb Maternal-lamb (n = 122) (n=46) (n=40) Group category Figure 5. Size by different types of Dall's sheep groups observed at Hoge Pass. Box-plots represent the interquartile range, which contains the 50% of values. The whiskers represent the highest and lowest values containing the 80% of values, excluding outliers (o) and extreme values (+). The line across the box indicates the median. Different letters indicate significant differences among groups. 22 The relative frequency distribution of group size of all the groups observed showed that single animal groups predominated, however, most of the animals of the populations were found in larger groups (Fig. 6). 1 10 19 28 37 46 Group size Figure 6. Relative frequency distribution of sheep groups and total number of animals in relation to group size. The proportion of animals lying down in groups did not significantly differ among group types (%2 = 2.342, df = 2, p = 0.31). However, the distribution of the proportion of the total animals lying down in relation to the distance to security cover varied significantly between males and both maternal groups (Kolmogorbv-Smirnov Z-test, p < 0.001). Also, the proportion of sheep lying down in maternal without lamb groups showed significantly negative correlations with distance to security cover (rs = -0.386, n = 40, p = 0.007) and group size (rs = -0.420, n = 40, p = 0.004). Lying 23 down animals in all categories tended to be closer to security cover, especially maternal groups (Fig. 7). 40 0 a Males b b Maternal+lamb Maternal-Iamb i — 15 5 10 20 Square-root of distance to security cover (m) 25 Figure 7. Relative frequency distribution of the proportion of sheep lying down in groups as function of the distance to security cover for different group types at Hoge Pass. Male distribution was significantly different than maternal with lamb and maternal without lamb groups (Z = 1.407, r?maies = 1 2 2 \"matemai+iamb = 46 /imaternai-iamb = .40, p = 0.038). Different letters indicate significant differences among groups. The distribution of proportion of lambs by distance to security cover decreased with increasing distance to security cover and it was significantly different from a uniform distribution (Z = 2.214, n = 86, p < 0.001) (Fig. 8). No maternal with lamb groups were observed > 15 m from security cover. 24 14 w 12 4 a 3 2 10 4 • - 8 4 (A (0 H 4 -o 2 4 0 0 - 5 6 - 10 11-15 15+ Distance from security cover (m) Figure 8. Relative frequency distribution of lambs in maternal groups observed at different distances from security cover at Hoge Pass. Habitat Selection by Group Type The 100% minimum convex polygon (MCP) ranges obtained from the locations of male and maternal groups in both areas showed that males occupied larger areas with a lower proportion of security cover (cliffs and talus) and a greater proportion of grasslands than did females (Table 2). They also showed a 26.8% overlap for the Area 1, and a total overlap in Area 2, where the females' range was included inside the males' range. The forage density index (FDI) for sites used by males (x = 0.18 ± 0.04) and maternal groups (x = 0.12 ± 0.03) in Area 1, differed significantly (Mann-Whitney U-test, U = 12, r?i = 13, n2 = 11, p < 0.001), indicating greater forage availability in male areas than maternal areas. Habitat selection, in terms of cover of main plant species, physiogeographic variables, and grassland nutritional quality, did not differ significantly between male 25 and maternal areas in Area 1 (Table 3). However, the proportion of lichen species was higher at maternal sites {U = 22, n-i = 13, n2 = 11, p = 0.04), while the Neutral Digestible Fibre (NDF %) was higher in male than maternal sites (U = 27, r?i = 13, n2 = 11, p = 0.01) (Table 3). Maternal sites were also located at significantly higher elevations than were male sites (U = 29.5, r?i = 13, n2 = 11, P = 0.015) (Table3). Table 2. Summary of ranges for male and maternal groups at two locations in Hoge Pass during early summer. Area 1 Area 2 Male Maternal Male Maternal Range size (ha) 829.5 663.3 331.8 166.0 Grassland % 49.1 35.2 20.7 7.4 Cliff / talus % 50.8 64.7 79.2 92.5 Predator Presence at Hoge Pass Potential predators observed in the study area were wolves, golden eagles, and red foxes. Although I did not see any grizzly bears, they are common at Hoge Pass (A. Frid, pers. comm., 1999), and the scats and forage holes of bears were seen frequently in areas used by male Dall's sheep. Wolverines also have been sighted here (A. Frid, pers. comm., 1999). Wolves were seen on two occasions. A single wolf unsuccessfully attacked a group of male sheep, and two wolves faced a group of adult males (class III - IV) but did not attack the sheep before leaving the area. These same two wolves were observed later that same morning watching a maternal with lamb group that was in 26 security cover. This group did not flee, remained in security cover, and the wolves left the area without attempting an attack. One fresh and three old wolf scats, as well as two sheep carcasses, one fresh (sex unknown) and one old carcass of female, were found in the males' range. No predator signs were found in maternal groups' range. 27 Table 3. Summary of vegetation and physiogeographic variables measured at feeding sites used by male and maternal groups at Area 1 at Hoge Pass. Group type Male {n = 13) Maternal (n =11) Variables X SD X SD % vegetation cover Dryas spp. 22.2 12.6 20.5 15.2 Other graminoids 6.7 4.3 5.0 2.9 Kobresia spp. 12.8 12.9 11.9 1.0 Mosses 3.4 5.6 2.6 1.6 Lichens * 1.0 1.1 3.6 2.7 Salix spp. 4.5 6.8 6.4 5.8 Forbs 9.7 6.5 7.0 5.0 Dead matter (litter) 35.7 14.3 38.2 8.8 Rock / bare ground 12.9 9.3 11.3 9.0 Nutritional variables % Vegetation dry matter 94.5 0.3 94.4 0.3 % Neutral detergent fibre * 47.4 3.5 43.5 2.7 % Crude protein 14.4 2.4 15.0 2.0 Site variables Altitude (m a.s.l.)* 1782.7 84.1 1868.6 48.4 Slope (degrees) 23.2 9.6 28.2 7.8 * means significantly different at p < 0.05 28 DISCUSSION During the period of my study, Dall's sheep showed distinct sexual segregation and used the available habitat components in different ways according to sex. I found that males were distributed in areas with a higher densities of forages, which had a higher proportion of non-digestible fibre than areas used by females. The areas used by males were larger than those used by maternal groups, and males ranged significantly further from security cover than did any maternal group type. Female areas contained a higher proportion of security cover than did male areas, and females rarely ranged from this security cover even when unaccompanied by young. The presence of offspring appeared to further restrict females to areas in or near security cover. The higher use of security cover by maternal groups supports a major prediction of the reproductive-strategy hypothesis, while the use by males of areas with abundant but lower quality food, supports a key prediction of the sexual dimorphism-body size hypothesis. Thus elements from both hypotheses would explain differences in response to predation risk and use of available food resources among Dall's sheep groups, at least during late spring and early summer. Although I calculated a low degree of habitat segregation, the sexes used the two main habitat types in Hoge Pass in different ways, producing a high degree of social segregation (after Conradt, 1998b). Habitat segregation was low because both sexes depend to some degree on rugged terrain for security cover, as do other species of Caprinae (Geist, 1971; Schaller, 1977; Bleich, 1999), and both sexes must leave cliff and talus slopes to forage and satisfy nutritional demands. Thus, the vegetation samples for maternal areas were taken in the alpine grassland and not in the cliff and talus slopes, which determined not significant differences between male 29 and maternal areas together with a small sample size. Also, recognising only two habitat types, could limit apparent \"choices\" for sheep. However, the distance animals were from security cover varied by sex according to the different levels of trade-off between predation risk and forage abundance, both of which are lower in security cover (Bleich et al., 1997; Rachlow & Bowyer, 1998; Bleich, 1999). Both types of maternal groups made heavy use of security cover that was typically at high elevation, free of snow during late spring and summer, and had patchily distributed forage. Despite some overlap was observed between ranges used by the sexes, male groups used significantly lower altitudes and were further from security cover than were females. It appeared that males did not depend on cliff and talus slopes as much as did maternal groups as has been found in other species (Shank, 1985; Bleich, 1999). I found that maternal groups with lambs were restricted almost entirely to areas in or near to security cover. This has also been reported by others, suggesting that offspring presence implies a greater susceptibility to predation risk (Shackleton & Haywood, 1985; Festa-Bianchet, 1988; Berger, 1991; Rachlow & Bowyer, 1988, 1994, 1998; Bon, Joachim & Maublanc, 1995; Bleich, 1999). Although no significant differences were observed between maternal groups with and without lambs in terms of distance to security cover, the absence of young allowed maternal groups to exploit resources located slightly further from security cover (i.e., the range of distances was greater in groups without lambs), which is also observed by others (e.g., Clutton-Brock etal., 1982; Festa-Bianchet, 1988; Bon etal., 1995). I observed an increase in group size with increased distance from security cover only for male group, which also ranged further from security cover than 30 maternal groups. Such an increase is an expected anti-predator behaviour in ungulates (Jarman, 1974; Berger, 1978; Risenhoover& Bailey, 1985; FitzGibbon, 1990; Fox, Sinha & Chundawat, 1992; Molvar & Bowyer, 1994; Roberts, 1996; Frid, 1997). Although maternal with lamb groups were always in, or very close to, security cover, they still formed relatively large groups. This might be because security cover was limited in area and so tended to concentrate females. Three maternal groups without lambs were observed at distances between 90 and 100 m from security cover; one was extremely large, which could be associated with the distance from security, but the other two were very small and seemed to be travelling between areas. These observations are also consistent with the idea that females experience a greater predation risk than the males. I did not observe differences in the proportion of active vs. inactive animals in different group types. But, when groups were seen further from security cover, their members tended to be more active, either walking or feeding, than were those in smaller groups that were closer to or in security cover, especially maternal groups without lambs. Except for maternal groups with lambs, sheep near or in security cover were less likely to be feeding and more likely to be either resting or ruminating. This use of security cover for all activities by the maternal groups with lambs, was similar to what has been observed in lactating female mouflon (O. orientalis) (Bon etal., 1995). The higher quality of forage in more elevated maternal areas may be attributed to phenological differences related to altitude during spring and summer (Klein, 1965; Hoefs, 1979; Albon & Langvatn, 1992) that happened to coincide with the distribution of the security cover in Hoge Pass. Perhaps by avoiding areas used heavily by females that had low forage density indices (FDI), males were able to find 31 more abundant food and higher FDIs, thus optimising their foraging, despite lower food quality. The lower quality of food should not affect adult males, because they have higher absolute requirements and larger digestive systems, allowing them to survive on lower quality forage than females. However, they are not as tolerant as females, to low quantities of food, even if this food is of high quality (Clutton-Brock etal., 1982, 1987; Shank, 1985; lllius & Gordon, 1990). My findings that males use areas with higher abundance, but lower quality food, suggest that males may be less constrained in habitat selection than are females. First, males are not involved in rearing young, which in spring and early summer are particularly vulnerable to predators because they are small, slow, and inexperienced (Shackleton & Haywood, 1985; Berger, 1991; Frid, 1997; Rachlow & Bowyer, 1998). Also, males have a large body size and massive horns, and are fast runners, and so should be less susceptible to predation (Bleich, etal., 1997; Bleich, 1999). Consequently, males should be able to bear a higher risk of encountering predators than can females, especially those accompanied by lambs. Thus males should be free to exploit those areas with more suitable and available forage (Shank, 1985; Berger, 1991). Also, it has been suggested that females are able to feed more efficiently than males in depleted food resources, because of their smaller body size and narrower muzzle (Clutton-Brock ef al., 1982, 1987). The narrower muzzle means a smaller bite size that allows them to selectively obtain sufficient food quality in these areas, while a smaller body size requires less forage quantity than the males (Gordon & lllius, 1988; lllius & Gordon, 1990; Perez-Barberia & Gordon, 1998). Clutton-Brock et al.'s (1982, 1987) explanation for sexual segregation in ungulates based on body size was founded on their observations of red deer 32 (Cervus elaphus) on the Island of Rhum (Scotland). I think that the unnatural conditions on this island (lack of large predators) may have resulted in an explanation unique to these conditions, and one that does not necessarily apply elsewhere such as in North America. Predation pressure on young was not considered by Clutton-Brock et al. (1982, 1987), while males and females were assumed to be equally free to feed anywhere on the island. The density of deer on Rhum was such that forage supplies were heavily utilised especially in winter and early spring. Females were suggested to out-compete males in areas with high quality but low biomass of forage. They attributed this to the females' narrower muzzle and smaller body size. The quantity of forage in areas used by females was so low that males could not obtain sufficient food to meet their requirements. Instead they had to feed in areas with higher quantities but poorer quality, even feeding on kelp and other marine algae. A final factor that may have influenced the deer's behaviour (and hence led to Clutton-Brock et al.'s explanation) is that this species is primarily adapted to forest habitat, but on Rhum, as in most other places in Scotland, forests no longer exist and the red deer are secondarily adapted to open habitats and to a higher proportion of grasses and forbs. This lower quality habitat is probably the reason Scottish red deer are some of the smallest in the world (Geist, 1998). In addition, Kie and Bowyer (1999) reported that following predator control, sexual segregation in a white-tailed deer (Odocoileus virginianus) population decreased, and females began to use areas used by males. Predation risk, and secondarily resource availability, probably played a key role in the evolution of sexual segregation in ungulates (Miquelle ef al., 1992; Main et al., 1996; Bleich etal., 1997; Bleich, 1999). In northwestern North America, where the richness of large predators is still present, sexually dimorphic ungulates show 33 marked sexual segregation (e.g., Miquelle et al., 1992; Bleich et al., 1997; the present work). On the other hand, in environments which lack large predators, sexually dimorphic ungulates do not always exhibit as pronounced sexual segregation, and mixed groups are common year-round. In the absence of large predators, I expect that habitats would be exploited by males and females solely according to differences in their nutritional requirements. And further that only females with very young offspring should segregate until the offspring have developed sufficient locomotor skills to keep up with the group (Shackleton & Haywood, 1985; Bon etal., 1995). An example of a sexually dimorphic species with limited sexual segregation is found in the genus Hippocamelus of the Cervidae family, endemic of the South American Andes (Merkt, 1987; Frid, 1999). The huemul deer (H. bisulcus) inhabiting Chilean coastal Patagonia shows aggregations of males and females despite distinct sexual dimorphism (Frid, 1999), only females with offspring were segregated from males and other females, dwelling mainly in cliff terrain either singly or in small groups (Frid, 1994). The same has been observed for the taruca deer (H. antisensis) in the southern Peruvian Andes, where only females with offspring maintain separate groups (Merkt, 1985; 1987). The cougar (Puma concolor) is the only potential large predator of adult ungulates remaining in the Andes mountain range, but it lives at low densities (Iriarte ef al., 1990). Both deer species naturally inhabit zones currently lacking more than one species of large carnivore, this situation would allow them to not segregate (see Kie & Bowyer, 1999). Culpeo foxes (Dusicyon culpaeus) however, are relatively common and suspected to prey on young Hippocamelus, which may lead females with young of the year to stay in security cover away from conspecifics (Frid, 1999). 34 In conclusion, I attribute the sexual segregation of Dall's sheep in Hoge Pass during late spring and early summer, mainly to predation risk factors experienced by different group types, thus supporting the reproductive-strategy hypothesis. However, the fact that the males used areas with lower quality forage, but higher availability, may lend support to the sexual dimorphism body-size hypothesis, however, differences in food requirements could simply be a consequence of body size differences between the sexes, which are the result of other selection pressures (e.g., large body size in males selected through intrasexual selection). 35 REFERENCES Alaback, P.B. (1986). Biomass regression equations for understory plants in coastal Alaska: effects of species and sampling design on estimates. Northwest Sci., 60: 90-103. Albon, S.D. & Langvatn, R. (1992). Plant phenology and benefits of migration in a temperate ungulate. Oikos, 65: 502-513. Altmann, J . (1974). Observational study of behaviour: sampling methods. Behaviour, 49: 227-267. Barboza, P.S. & Bowyer, R.T. (2000). Sexual segregation in dimorphic deer: a new gastrocentric hypothesis. J. Mammal., 81: 473-489. Beier, P. & McCullough, D.R. (1990). Factors influencing white tail deer activity patterns and habitat use. Wildl. Monogr., 109: 1-51. Berger, J . (1978). Group size, foraging, and antipredator ploys: an analysis of bighorn sheep decisions. Behav. Ecol. & Sociobiol., 4: 91-99. Berger, J . (1991). Pregnancy incentives, predation constrains and habitat shifts: experimental and field evidence for wild bighorn sheep. Anim. Behav., 41: 61-77. Bleich, V .C . (1999). Mountain sheep and coyotes: patterns of predator evasion in a mountain ungulate. J. Mammal., 80: 238-289. Bleich, V . C , Bowyer, R.T. & Wehausen. J.D. (1997). Sexual segregation in mountain sheep: resources or predation? Wildl. Monogr., 134: 1-50. Bon, R. & Campan, R. (1996). Unexplained sexual segregation in polygamous ungulates: a defence of an ontogenetic approach. Behav. Processes, 38: 131-154. Bon, R., Joachim, J . & Maublanc, M.L. (1995). Do lambs affect habitat use by lactating female mouflons in spring in areas free of predators? J. Zool., Lond., 235: 43-51. Bowyer, T.R., Kie, J .G . & Van Ballenberghe, V. (1996). Sexual segregation in black-tailed deer: effects of scale. J. Wildl. Manage., 60: 10-17. Bowyer, T.R. & Leslie, D.M. (1992). Ovis dalli. Mamm. Species, 393: 1-7. Bowyer, T.R., Leslie, D.M. & Rachlow, J.L. (2000). Dall's and Stone's sheep. In Ecology and management of large mammals of North America: 491-516. Demaries, S. & Krausman, P.R. (Eds.). Upper Saddle River: Prentice Hall Inc. Brower, J .E . & Zar, J .H. (1979). Field and laboratory methods for general ecology. Dubuque: Brown Company Publishers. Bunnell, F.L. (1982). The lambing period of mountain sheep: synthesis, hypotheses, and test. Can. J. Zool., 60: 1-14 Clutton-Brock, T.H., Guinness, F.E. & Albon, S.D. (1982). Red deer: behaviour and ecology of two sexes. Chicago: University of Chicago Press. Clutton-Brock, T.H., lason, G.R. & Guinness, F.E. (1987). Sexual segregation and density-related changes in habitat use in male and female red deer (Cervus elaphus). J. Zool., Lond., 211: 275-289. Conradt, L. (1998a). Could asynchrony in activity between the sexes cause intersexual social segregation in ruminants? Proc. Roy. Soc. Lond. (B), 265: 1359-1363. Conradt, L. (1998b). Measuring the degree of sexual segregation in group-living animals. J. Anim. Ecol., 67: 217-226. 37 Conradt, L. (1999). Social segregation is not a consequence of habitat segregation in red deer and feral soay sheep. Anim. Behav., 57: 1151-1157. Cransac, N., Gerard, J.F. , Maublanc, M.L. & Pepin, D. (1998). An example of segregation between age and sex classes only weakly related to habitat use in mouflon sheep (Ovis gmelini). J. Zool., Lond., 244: 371-378. Festa-Bianchet, M. (1988). Seasonal range selection in bighorn sheep: conflicts between forage quality, forage quantity, and predator avoidance. Oecologia, 75: 580-586. FitzGibbon, C D . (1990). Mixed-species grouping in Thomson's and Grant's gazelles: the antipredator benefits. Anim. Behav., 39: 1116-1126. FitzGibbon, C D . & Lazarus, J . (1995). Anti-predator behaviour of Serengeti ungulates: individual differences and population consequences. In Serengeti II: research, management and conservation of an ecosystem: 274-296. Sinclair, A .R .E . & Arcese, P. (Eds.). Chicago: University of Chicago press. Fox, J.L., Sinha, S .P . & Chundawat, R.S. (1992). Activity patterns and habitat use of ibex in the Himalaya mountains of India. J. Mammal., 73: 527-534. Frid, A. (1994). Observations on habitat use and social organization of a huemul Hippocamelus bisulcus coastal population in Chile. Biol. Conserv. 67: 13-19. Frid, A. (1997). Vigilance by female Dall's sheep: interactions between predation risk factors. Anim. Behav., 53: 799-808. Frid, A. (1999). Huemul [Hippocamelus bisulcus) sociality at a periglacial site: sexual aggregation and habitat effects on group size. Can. J. Zool., 77: 1083-1091. Geist, V. (1971). Mountain sheep: a study in behavior and evolution. Chicago/London: University of Chicago Press. Geist, V. (1998). Deer of the World: their evolution, behavior, and ecology. Mechanicsburg: Stackpole Books. Geist, V. & Petocz, R.G. (1977). Bighorn sheep in winter: do rams maximize reproductive fitness by spatial and habitat segregation from ewes? Can. J. Zool., 55: 1802-1810. Giroux, S. (1991). Dall sheep survey, Donjek range, Kluane National Park Reserve. TR-91-05/KLU. Gordon, I.J. & lllius, A.W. (1988). Incisor arcade structure and diet selection in ruminants. Func. Ecol., 2: 15-22. Hoefs, M. (1979). Flowering plant phenology at Sheep Mountain, southwest Yukon Territory. Can. Field-Nat, 93: 183-187. Hoefs, M. & Bayer, M. (1983). Demographic characteristics of an unhunted Dall sheep (Ow's dalli) population in southwest Yukon, Canada. Can. J. Zool., 61: 1346-1357. Hoefs, M. & Cowan, I.M. (1979). Ecological investigation of a population of Dall sheep (Ov/'s dalli dalli Nelson). Syesis, 12: 1-83. Hoefs, M., Cowan, I.M. & Krajina, V . J . (1975). Phytosociological analysis and synthesis of Sheep mountain, southwest Yukon Territory, Canada. Syesis, 8: 125-228. Houtman, R. & Dill, L.M. 1998. The influence of predation risk on diet selectivity: a theoretical analysis. Evol. Ecol., 12: 215-262. lllius, A.W. & Gordon, I.J. (1990). Variation in foraging behaviour on red deer and i the consequences for population demography. J. Anim. Ecol., 59: 89-101. Iriarte, J.A., Redford, K.H., Franklin, W.L. & Johnson, W.E. (1990). Biogeographic variation of food habits and body size of the American puma (Felis concolor). Oecologia, 82: 185-190. Jarman, P.J . (1974). The social organization of antelope in relation to their ecology. Behaviour, 48: 216-267 Kie, J .G . , Baldwin, J.A. & Evans, C .J . (1994). CALHOME, home range analysis program electronic. User's Manual. Kie, J .G . & Bowyer, R.T. (1999). Sexual segregation in white-tailed deer: density-dependent changes in use of space, habitat selection, and dietary niche. J. Mammal., 80: 1004-1020. Klein, D.R. (1965). Ecology of deer range in Alaska. Ecol. Monogr., 35: 259-284. Machlis, L.P., Dodd, W.D. & Fentress, J .C . (1985). The pooling fallacy: problems arising when individuals contribute more than one observation to the data set. Z. Tierpsychol., 68: 201-214. Main, M.B. & Coblentz, B.E. (1990). Sexual segregation among ungulates: a critique. Wildl. Soc. Bull., 18: 204-210. Main, M.B. & Coblentz, B.E. (1996). Sexual segregation in Rocky mountain mule deer. J. Wildl. Manage., 60: 497-507. Main, M.B., Weckerly, F.W. & Bleich, V .C. (1996). Sexual segregation in ungulates: new directions for research. J. Mammal., 77: 621-630. Martin, P. & Bateson, P. (1993). Measuring behaviour: an introductory guide. (2nd edn). Cambridge: Cambridge University Press. Merkt, J .R. (1985). Social structure of Andean deer (Hippocamelus antisensis) in southern Peru. M.Sc. thesis, University of British Columbia. Merkt, J.R. (1987). Reproductive seasonality and grouping patterns of the north Andean deer or taruca (Hippocamelus antisensis) in southern Peru. In Biology and management ofCervidae: 388-401. Wemmer, C C . (Ed.). Washington: Smithsonian Institution Press. Miquelle, D.G., Peek, J .M. & Van Ballenberghe, V. (1992). Sexual segregation in Alaskan moose. Wildl. Monogr., 122: 1-57. Molvar, E.M. & Bowyer, T.R. (1994). Cost and benefits of group living in a recently social ungulate: the Alaskan moose. J. Mammal., 75: 621-630. Murie, A. (1944). The wolves of Mount McKinley. U.S. Nat. Park Serv.: Fauna Ser. 5: 1-238 Nette, T., Buries, D. & Hoefs, M. (1984). Observation of golden eagle (Aquila chrysaetos) predation on Dall sheep (Ow's dalli dalli) lambs. Can. Field-Nat., 98: 252-254. Nichols, L. & Bunnell, F.L. (1999). Natural history of thinhorn sheep. In Mountain sheep of North America: 23-77. Valdez, R. & Krausman. P.R. (Eds.). Tucson: University of Arizona Press. Perez-Barberia, F.J. & Gordon, I.J. (1998). The influence of sexual dimorphism in body size and mouth morphology on diet selection and sexual segregation in cervids. Acta Vet. Hung., 46: 357-367. Rachlow, J.L. & Bowyer, R.T. (1988). Interannual variation in timing and synchrony of parturition in Dall's sheep. J. Mammal., 72: 487-492. Rachlow, J.L. & Bowyer, R.T. (1994). Variability in maternal behaviour by Dall's sheep: environmental tracking or adaptive strategy? J. Mammal., 75: 328-337. 41 Rachlow, J.L. & Bowyer, R.T. (1998). Habitat selection by Dall's sheep (Ov/'s dalli): maternal trade-offs. J. Zool., Lond., 245: 457-465. Risenhoover, K.L. & Bailey, J.A. (1985). Foraging ecology of mountain sheep: implications for habitat management. J. Wildl. Manage., 49: 797-804. Roberts, G. (1996). Why vigilance declines as group size increase. Anim. Behav., 51: 1077-1086. Robbins, C.T. (1983). Wildlife feeding and nutrition. New York: Academic Press. Ruckstuhl, K.E. (1998). Foraging behaviour and sexual segregation in bighorn sheep. Anim. Behav., 56: 99-106. Ruckstuhl, K.E. (1999). To synchronise or not to synchronise: a dilemma for young bighorn males? Behaviour, 136: 805-818. Ruckstuhl, K.E. & Neuhaus, P. (2000). Sexual segregation in ungulates: a new approach. Behaviour, 137: 361-377. Schaller, G.B. (1977). Mountain monarchs: wild sheep and goats of the Himalaya. Chicago: University of Chicago Press. Shackleton, D.M. & Haywood, J . (1985). Early mother-young interactions in California bighorn sheep, Ovis canadensis californiana. Can. J. Zool., 63: 868-875. Shank, C C . (1982). Age-sex differences in diets of wintering Rocky mountain bighorn sheep. Ecology, 63: 627-633. Shank, C C (1985). Inter- and intra-sexual segregation of chamois (Rupricapra rupicapra) by altitude and habitat during summer. Z. Saugetierkd., 50: 117-125. S P S S Inc. (1997). S P S S 9.0 for Windows. Chicago: S P S S Inc. Stephens, D.W. and J.R. Krebs. (1986). Foraging Theory. Princeton: Princeton University Press. Sumanik, R.S. (1987). Wolf ecology in the Kluane region, Yukon Territory. M.Sc. thesis, Michigan Technological University. Van Soest, P.J . (1982). Nutritional ecology of the ruminant. Portland: Durham and Downey Inc. Van Soest, P.J . , Robertson, J.B. & Lewis, B.A. (1991). Symposium: carbohydrate methodology, metabolism, and nutritional implications in dairy cattle. J. Dairy Sci., 74: 3583-3597. Villaret, J .C . & Bon, R. (1995). Social and spatial segregation in Alpine ibex {Capra ibex) in Bargy, French Alps. Ethology, 101: 291-300. Villaret, J . C , Bon, R. & Rivet, A. (1997). Sexual segregation of habitat by the Alpine ibex in French Alps. J. Mammal., 78: 1273-1281. Weckerly, F.W. (1993). Intersexual resource partitioning in black-tailed deer: a test of the body size hypothesis. J. Wildl. Manage., 57: 475-494. Zar, J . (1999). Biostatistical analysis. (4th edn). Upper Saddle River: Prentice Hall Inc. "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2001-05"@en ; edm:isShownAt "10.14288/1.0089797"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Animal Science"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Dall’s sheep (Ovis dalli dalli Nelson, 1884) sexual segregation : interactions between two hypotheses"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/11266"@en .