Quantifying vigilance and exploring the underlying mechanisms has been the subject of numerous studies. Less attention has focused on the complex interplay between contributing factors such as reproductive status, social rank, sex and group size. Reproductive status and social rank are of particular interest due to their association with mating behavior. Mating activities in rutting season may interfere with typical patterns of vigilance and possibly interact with social rank. In addition, balancing the tradeoff between vigilance and life maintenance may represent a challenge for gregarious ungulate species rutting under harsh winter conditions. We studied vigilance patterns in the endangered Przewalski's gazelle (Procapra przewalskii) during both the rutting and non-rutting seasons to examine these issues.
Field observations were carried out with focal sampling during rutting and non-rutting season in 2008–2009. Results indicated a complex interplay between reproductive status, social rank, sex and group size in determining vigilance in this species. Vigilance decreased with group size in female but not in male gazelles. Males scanned more frequently and thus spent more time vigilant than females. Compared to non-rutting season, gazelles increased time spent scanning at the expense of bedding in rutting season. During the rutting season, territorial males spent a large proportion of time on rutting activities and were less vigilant than non-territorial males. Although territorial males may share collective risk detection with harem females, we suggest that they are probably more vulnerable to predation because they seemed reluctant to leave rut stands under threats.
Vigilance behavior in Przewalski's gazelle was significantly affected by reproductive status, social rank, sex, group size and their complex interactions. These findings shed light on the mechanisms underlying vigilance patterns and the tradeoff between vigilance and other crucial activities.
Citation: Li C, Jiang Z, Li L, Li Z, Fang H, Li C, et al. (2012) Effects of Reproductive Status, Social Rank, Sex and Group Size on Vigilance Patterns in Przewalski's Gazelle. PLoS ONE 7(2): e32607. https://doi.org/10.1371/journal.pone.0032607
Editor: Jane M. Waterman, University of Manitoba, Canada
Received: September 5, 2011; Accepted: January 31, 2012; Published: February 28, 2012
Copyright: © 2012 Li et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was supported by Science and Technology Supporting Project, Ministry of Science and Technology, China (Monitoring of Important Biological Species and Demonstration of Key Conservation Technology Application in China, 2008BAC39B04, http://program.most.gov.cn/), Key Program of Knowledge Innovation Program of Chinese Academy of Sciences (KSCX2-EW-Z-4, http://www.bio.cas.cn/xmyj/yzscxzdxm/) and National Natural Sciences Foundation of China (Population patterns and structure of Przewalski's gazelle in anthropogenic landscape, No. 31070469, http://www.nsfc.gov.cn/nsfc/cen/xxgk/slzz.html). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Vigilance in animals may reduce the likelihood of being attacked and thus increase fitness by improving the ability to survive, obtain resources, reproduce and protect offspring , . However, vigilance behavior is often performed at the expense of other fitness-enhancing activities crucial for life maintenance and reproduction, such as foraging, resting and mating , . Quantifying vigilance patterns and exploring underlying influencing factors help us to understand how animals respond to potential threats and balance the tradeoff between vigilance and other crucial activities .
Factors influencing vigilance behavior have been studied in a wide variety of birds and mammals . In ungulates, vigilance is influenced by sex , , level of predation risk , , group size ,  and position in the herd . While not universal , many studies have detected a decrease in individual vigilance in larger groups, which is usually explained by the many-eyes effect, risk dilution or scramble competition , , . The first two mechanisms, related to anti-predatory vigilance, imply that the presence of more companions can lead to better threat detection and a simple dilution of risk upon attack. The third hypothesis argues that increased competition for limited resources in larger groups forces individuals to relax vigilance to increase their relative shares of limited resources .
Season is an important factor influencing activity patterns , , . Most studies examining the effect of season on vigilance have focused on the calving season, and have found that mothers with young spend more time vigilant , . Rutting season is also a critical period where animals spend considerable time and energy in mating activities . Vigilance patterns may differ with seasonal changes in reproductive status , . Thus far, however, studies concentrating on the effect of reproductive status and its interaction with other factors have been limited. Particularly, little attention has been paid to ungulates that rut under harsh winter conditions. For such species, balancing the tradeoff between vigilance and life maintenance while rutting could be a challenge. Given the time demands for rutting, especially for males, and the fact that food is often of poor quality in winter, maintaining adequate levels of vigilance may be quite difficult without compromising body condition , . Investigation of vigilance patterns under such conditions could help to understand strategies to allocate limited time to vigilance under predation risk.
During the rutting season, animals of different social ranks vary in mating tactics and in their allocation of time and energy to daily activities , . Vigilance patterns may thus differ in animals of different social ranks . A survey of the literature examining the effect of social rank on vigilance behavior in rutting ungulates reveals contradictory findings , , , . For instance, territorial males in impala (Aepyceros melampus) allocated more time to vigilance against intruding rivals . In elk (Cervus elaphus), however, rutting bulls engaged in courtship spent little time vigilant . Social rank is typically used to account for differences in vigilance between dominant and subordinate individuals , . However, discrepancy in the results so far may reflect interactions between social rank and species-specific mating systems, which suggests that more empirical studies are needed to investigate the effect of social rank on vigilance behavior during the rutting season.
We used Przewalski's gazelle (Procapra przewalskii) as a model species to investigate the effects of reproductive status, social rank, sex and group size on vigilance patterns in ungulates. Mating system of Przewalski's gazelle is characterized as the female traffic version of the hotspot hypothesis . After fights among male gazelles prior to winter rutting, each winner herds a harem of females that roam along a relatively fixed daily travel route. Losers forage apart with occasional challenges to dominant males resulting in sporadic mating chances. This type of mating system provides an ideal model to investigate behavioral strategies, especially the tradeoff between essential activities in socially foraging animals. Li et al.  and Li et al.  have found that vigilance in Przewalski's gazelle was influenced by group size, distance to human infrastructure and level of predation risk, and that these effects also varied between the sexes. Here, we intended to determine: 1) whether vigilance patterns differ between the rutting and non-rutting season; 2) if so, whether the effect of reproductive status differs between the sexes; and 3) whether and how social rank influences vigilance behavior during the rutting season. We predicted that gazelles would be more vigilant in rutting season because of increased demands for monitoring conspecifics with potentially different outcomes for males and females and for individuals of different social rank. We also attempted to understand how gazelles balance the tradeoff between vigilance, feeding, bedding and rutting activities based on time allocated to these essential behaviors.
We adhered to the “Guidelines for the use of animals in research” published in Animal Behaviour. Our research protocols have been approved by the Chinese Wildlife Management Authority. The study was observational involving no cruelty to animals and thus no review from the ethnic committee was required in China. All the work was carried out under the Wildlife Protection Law of the People's Republic of China.
Study Site and Species
Field observations were carried out using one population of Przewalski's gazelle in Shengge (37°27′48″ N, 98°33′45″ E) along the Upper Buha River Valley, located in the north-east of the Qinghai-Tibetan Plateau, China. Elevation in the distribution area of this population ranges from 3,500 m to 4,000 m above sea level. The region has an inland plateau semi-arid climate with dry, cold and long winters and a short frost-free period. Mean annual temperature is −1.5°C with the lowest record of −40°C. Annual precipitation is 330∼500 mm with mean evaporation of 1,300∼2,000 mm. Dominant vegetation type is alpine meadow with shrubs along the Buha River Valley. The main predator is wolf (Canis lupus) with an estimated population of ten individuals in this area. Tibetan fox (Vulpes ferrilata) is also common, contributing to some calf mortality . Human disturbance is rare.
Przewalski's gazelle is an endangered species endemic to the Qinghai-Tibetan Plateau , , . Since the 1950 s, the gazelle's population has significantly declined and is now restricted to several isolated areas around the Qinghai Lake in the north-east of the plateau , . The overall population size of the gazelle is estimated at 1,200∼1,600 with approximately 100 adult gazelles in Shengge . Sympatric Tibetan gazelle (Procapra picticaudata) occasionally forages with Przewalski's gazelle in the Shengge area , . Rutting of Przewalski's gazelle lasts from late December to mid-January . During the rutting season, in the winter 2008–2009 when this study was carried out, two types of groups were found: numerous non-territorial male groups and nine relatively stable reproductive groups composed of one territorial male and harem females. During the non-rutting season in the summer 2009, the two sexes mainly foraged apart (∼20 groups), which is referred to as sexual segregation , .
Field observations were carried out on sunny days from December 24, 2008 to January 15, 2009 (rutting season of Przewalski's gazelle) and from June 5 to June 30, 2009 (non-rutting season). During field work, we used focal sampling  to record behaviors from sunrise to sunset, i.e. 07:00 to 20:00 in summer and 08:30 to 18:00 in winter.
We selected groups of Przewalski's gazelle for behavioral observation as they were encountered along a fixed route. We defined a focal group as a collection of individuals all occurring within 50 m of one another . Group members were mentally numbered from left to right according to their location in the group. Using a random number generator, one focal individual was then selected among those present. In order to minimize the likelihood of pseudo-replication, no group was observed more than once on the same day, and we only selected one focal adult individual in a focal group , . With this random procedure, we estimated that all individuals within a category had the same chances of being sampled. Since it was not possible to mark individuals, the same individuals may have been watched on different days. However, this probably did not occur very often due to the large number of gazelles in the area. In addition, intervening changes in group size and spatial position within groups between observation days created very different contingencies for any successive observation on the same subjects. Therefore, it is reasonable to assume that pseudo-replication is not a major issue here.
We defined five categories of individuals with respect to reproductive status, sex and social rank: females, non-territorial males and territorial males in rutting season and females and males in non-rutting season (Table 1). Social rank was specific to males (territorial and non-territorial males) during the rutting season. Territorial males were easily recognized by morphological characteristics (denser neck color, prominent larynx and exposed penis), rutting behaviors (herding and mounting females, mating roar, frequently marking rut stands with urine and chasing away intruding males) and relatively stable rut stands along female travel routes. Non-territorial males forage alone or in small single-sex groups away from territorial males and harem females , . Li et al.  found that mother gazelles with calves were generally more vigilant than females without calves. Therefore, to eliminate the effect of attending calves in females, we did not select mother gazelles as focal individuals in our field observations. This choice also makes the comparison of females between seasons simpler. Subadults were pooled with adults as they reach adult size in their first winter and were not practically distinguishable from adults in the field . Finally, we only considered single-species herds as focal groups to avoid the effect of “additional eyes” from different species , .
For each observation session, we recorded behavior, season, date, time, group size, herd composition, sex and social rank of the focal gazelle. We recorded six behavioral states: feeding, bedding, moving, vigilance, rutting activity and other behaviors. Feeding was defined as grazing or short bout of searching between grazing bouts with the head held below the shoulders. Bedding referred to rest and rumination while sitting on the ground. Moving was defined as walking or trotting with the head held above the horizontal plane without foraging. A gazelle was considered vigilant when it was standing with the head above shoulder level and scanning. Obvious alert scanning during other activities was also regarded as vigilance. Rutting activities in males included herding and guarding harem females, displaying, mounting, mating and marking rut stands with urine as well as chasing away intruders of the same sex. Females showed fewer and simpler rutting activities, including tail wagging and accepting mating . Other behaviors included behaviors that were not listed in the above categories, such as defecating, sneezing and scratching . For females and territorial males during the rutting season, we defined group size as the number of adults in the mixed-sex group including the dominant male and harem females. In the non-rutting season, group size was the total number of adult individuals in a group.
Behaviors were dictated as they occurred on an Mp3 digital recording pen inside an off-road vehicle using binoculars (8×42) or a telescope (20–60×63). Field observations have shown previously that Przewalski's gazelle habituate to parked vehicles, as found in other ungulates . Vehicles moving slowly could often approach gazelles within 100 m with no apparent reaction. To avoid possible disturbances from the vehicle, we slowly drove the vehicle toward the target group and stopped 150∼200 m away. Furthermore, we waited inside the vehicle several minutes prior to recording behaviors. We stopped a focal observation when the focal individual was out of sight, when the focal group composition changed or human disturbances occurred. Field observation and recording task was carried out by the same person.
Information from Mp3 records was transcribed into timed sequences of behavioral states in EXCEL spreadsheet with the ETHOM software version 1.0 (available at http://web.nchu.edu.tw/~htshih/ethom/intro_c.htm) . To reduce stochastic variation due to short sampling sessions and to increase data reliability and representativeness of the overall time budget in Przewalski's gazelle, observations shorter than five minutes were discarded, as was done in other studies , . To obtain percentages of time spent in different activities, the total amount of time spent in one activity during a focal observation was divided by the duration of the focal observation. In addition, we calculated the number of vigilance bouts, which represents the number of transitions to a vigilant state in a focal observation. Mean scan duration represented the amount of time spent vigilant divided by the number of vigilant bouts in one focal observation. Scan frequency represented the number of vigilant bouts divided by focal observation duration and expressed as frequency per min. Prior to statistical analysis, data were tested for normality with the Shapiro-Wilk Test. Percentages were arcsine-square-root transformed and mean scan durations were log10 transformed to meet assumptions of normality and homogeneity of variances. All statistical analyses were carried out with SAS v. 8.1 (SAS Institute Inc., Cary, NC, USA) with a level of statistical significance set at 0.05. Data below are shown as mean (± SE).
For the analysis of percentage time spent scanning and mean scan duration (the response variables), a linear model (GLM, PROC GLM in SAS) was used. We first considered the effects of reproductive status (two levels: rutting or non-rutting), sex (female or male), group size (continuous factor) and all interactions of these explanatory variables. Social rank (two levels: territorial or non-territorial) was specific to males during the rutting season. Therefore, we fitted an independent GLM to test the effect of social rank and group size and the interaction between the two, using data from territorial and non-territorial males during the rutting season. Non-significant effects (p>0.05) were excluded in the final models by backward elimination. We included time of day (three levels: before 11:00, 11:00–15:00, after 15:00) as a categorical factor in an earlier analysis but found no significant effect (F2,396 = 0.10, t = 0.32, p = 0.748 for percentage time spent scanning; F2,396 = 0.20, t = −0.45, p = 0.654 for scan duration) and thus did not consider this factor further.
A generalized linear model (PROC GENMOD in SAS), with a negative-binomial error structure and a logarithmic link function, was fitted to the heavily right-skewed scan frequency data. We used number of scanning bouts in focal observations as the response variable and included the natural logarithm of observation time as an offset because focal observation duration varied among individuals. The same fixed factors as used in the linear model were included in the model. Time of day was also removed due to its non-significant effect (χ2 = 1.02, p = 0.312, df = 2). The ratios of the deviance to degrees of freedom were approximately equal to one, indicating good fitness of the models to the data.
Pearson's correlation was used to test the relationship between percentage time spent scanning, feeding, bedding and rutting.
A total of 399 focal observations, amounting to 9,864 min, were collected over the two seasons (Table 1). Duration of focal observations ranged from 5.0 min to 32.9 min with an average of 24.7 min (±0.3).
Percentage time spent scanning
Percentage time spent scanning averaged 2.9% (±0.2%) and ranged from 0 to 33.4%. The final model without social rank revealed a significant effect of sex, reproductive status, group size and an interaction between sex and group size. The interaction between sex and group size indicated a negative effect of group size in females (β = −0.002±0.001, t = −2.72, p = 0.007) but not in males (β = −0.003±0.002, t = −2.04, p = 0.113) (Fig. 1). Males (3.6%±0.3%) spent more time scanning than females (2.0%±0.1%). Percentage time spent scanning increased in rutting (3.5%±0.3%) compared to non-rutting season (2.1%±0.1%). The independent model to test the effect of social rank indicated that non-territorial males (7.0%±0.9%) spent more time scanning than territorial males (1.6%±0.2%) during the rutting season (Table S1).
Scan duration averaged 10.3 s (±0.6) and ranged from 0 to 112.8 s. The final model without social rank revealed a significant effect of sex and group size. Scan duration decreased with group size (β = −0.008±0.003, t = −2.25, p = 0.025). Mean scan duration in females (11.3±1.0 s) was longer than in males (9.6±0.6 s). There was no significant difference between rutting (10.4±0.8 s) and non-rutting season (10.3±0.8). The independent model to test the effect of social rank indicated that there was no significant difference between territorial (9.0±0.8 s) and non-territorial males (9.6±0.6 s) during the rutting season (Table S2).
Scan frequency averaged 0.17 min−1 (±0.01) and ranged from 0 to 1.36 min−1. The final model without social rank revealed a significant effect of sex and an interaction between sex and group size. There was no overall significant effect of group size on scan frequency. The interaction between sex and group size indicated that scan frequency decreased with group size in females (β = −0.080±0.013, χ2 = 38.92, p<0.001) but not in males (β = −0.015±0.009, χ2 = 2.56, p = 0.110). Scan frequency was significantly higher in males (0.22±0.02 min−1) than in females (0.12±0.01 min−1). The independent model to test the effect of social rank indicated that scan frequency in non-territorial males (0.41±0.04 min−1) was significantly higher than in territorial males (0.11±0.01 min−1) during the rutting season (Table S3).
Relationship between time spent scanning, feeding, bedding and rutting
Time spent scanning was relatively small in Przewalski's gazelle. Gazelles devoted a large proportion of time to feeding and bedding, the other two main activities (Table 2). During the rutting season, percentage time spent scanning was negatively correlated with bedding (r = −0.34, p = 0.001, N = 87 in females; r = −0.23, p = 0.038, N = 81 in territorial males; r = −0.61, p<0.001, N = 68 in non-territorial males) but positively correlated with feeding (r = 0.24, p = 0.027, N = 87 in females; r = 0.26, p = 0.020, N = 81 in territorial males; r = 0.29, p = 0.018, N = 68 in non-territorial males). No significant relationship was found between percentage time spent scanning and rutting in territorial males (r = −0.09, p = 0.422, N = 81). Rutting activities did not occur frequently in non-territorial males and in females, preventing us for calculating the correlation between scanning and rutting in these animals.
During the non-rutting season, a negative correlation was found between percentage time spent scanning and feeding (r = −0.32, p = 0.002, N = 94 in females; r = −0.21, p = 0.080, N = 69 in males) but a positive correlation was documented between scanning and bedding (r = 0.21, p = 0.042, N = 94 in females; r = 0.29, p = 0.015, N = 69 in males).
Overall, the results illustrate that vigilance in Przewalski's gazelle reflects a complex interplay between reproductive status, social rank, sex and group size. Changes in time spent vigilant were caused by changes in both scan frequency and scan duration. A recent meta-analysis in birds found that the effect size related to scan frequency was usually larger than that for scan duration . A similar meta-analysis is lacking in mammals so that it is difficult to judge whether the pattern documented in gazelles here is common in other mammals. In the following, we focus on time spent vigilant as a proxy for vigilance since the effect of group size was consistent among all three measures of vigilance.
First, we consider the effect of reproductive status. Both females and non-territorial males in Przewalski's gazelle spent more time vigilant during the rutting than the non-rutting season. The seasonal increase in vigilance probably reflects increased threats associated with rutting as well as environmental changes between seasons , , . First, seasonal change associated with rutting is probably a driving force for vigilance patterns, as noted in other taxa , , . During the rutting season, increased aggressive behaviors of dominant males, as a result of changes in physiological status , , may force subordinate males and females to allocate more time to monitoring conspecifics , , , . Territorial males in Przewalski's gazelle frequently herded and guarded females to keep them inside their rut stands. Territorial males roamed among females and approached them to identify and forcibly mount sexually recipient ones , . Aggressive territorial males harass female gazelles and may even cause physical injuries , , . Therefore, extra scanning by females in rutting season could be directed at territorial males. Non-territorial males are generally subadults or weaker individuals that lost fights to establish dominance at rut stands located along the daily routes taken by females for grazing and drinking , . Such failure probably keeps non-territorial males vigilant against aggressive territorial males throughout the rutting season . Non-territorial males typically foraged and traveled far away from territorial males and frequently scanned against potential threats.
In addition to the demands caused by rutting, vigilance may also be influenced by harsh winter conditions, which reduce food quality. Poor foraging during the rutting season forces Przewalski's gazelle to spend more time feeding, as has been noted in other northern hemisphere ungulates , . Interestingly, gazelles which faced greater demands on vigilance and foraging during rutting reduced time spent bedding without compromising vigilance. Similar results are found in other studies which investigated tradeoffs between vigilance and other activities within the overall time budget , . Whether the reduction in time spent bedding has any consequences on fitness remains to be determined in these gazelles.
We found that territorial males spent the least amount of time on vigilance, which is consistent with results from some ungulate studies , ,  but not with others , . For males, vigilance patterns may reflect the interplay between species-specific mating strategy and social rank , . Unlike impala and captive Père David's deer (Elaphurus davidianus), in which dominant males (harem masters) guard fixed rutting territories and are frequently challenged by bachelors , , territorial males in Przewalski's gazelle spend a large proportion of time herding and guarding roaming harem females. Non-territorial males in Przewalski's gazelle are rarely seen near or inside rut stands to challenge territorial males . Head-on confrontations between territorial males and non-territorial males are thus rarely seen after the establishment of dominance status. In addition, we did not observe any intrusions from other territorial males probably because rut stands are quite scattered (>500 m) in the study area and each harem master only herds his own harem females. Presumably, territorial males benefit from this relatively stable social rank by relaxing vigilance against intruding rivals. Other hypotheses to explain lower vigilance appear less likely. For instance, territorial males during the short rutting season allocate a large proportion of time and energy to rutting activities , , , which may cause a reduction in vigilance against potential threats , . However, we found that rutting did not interfere with scanning in territorial males. It is conceivable that territorial males benefit from the vigilance of their harem females while non-territorial males are forced to be more vigilant because they forage in smaller groups. In contrast, we found that group size did not influence vigilance in males.
Low vigilance in territorial males may not unduly influence the risk of being preyed upon. First, any risk is diluted with harem females , . Furthermore, both guarding and herding harem females keep territorial males active. During these activities, territorial males can conceivably perceive disturbances detected by more alert females. Nevertheless, we observed that territorial males did not flee away immediately upon detecting an approaching threat, as females did. Territorial males seemed reluctant to leave rut stands. Similar delayed fleeing has also been documented in other mammals and birds , , . This, together with poor body condition resulting from extensive rutting but less feeding , , may make territorial males more vulnerable to predation.
As documented by Li et al. , we found a negative group size effect on vigilance but in female gazelles only. The sexual difference probably reflects different targets of vigilance in males and females. Any benefits related to collective detection and risk dilution in males is probably compensated by increased monitoring of females in larger groups. The mitigating effect of conspecific monitoring on the group size effect has been predicted and documented in other species , . Females probably monitor both males and predators and therefore benefit from the presence of more companions.
What causes the decrease in individual vigilance with group size in females? We argue that the scramble competition hypothesis , ,  is unlikely to address this pattern. In spite of low food quality during the rutting season, the alpine meadow where the gazelles forage is homogenous and large, limiting the role of food competition on vigilance . Instead, the observed decline in individual vigilance in large groups probably reflects increased safety caused by the presence of more eyes to detect threats and more bodies to dilute risk , , . We suggest that risk dilution in larger groups may apply to predation threats as well as threats from territorial males. During the rutting season, territorial males herd and often harass females , , acting as conspecific threats. The presence of several female companions in larger groups can dilute sexual harassment from the dominant male and allow females to relax vigilance in larger groups. This could be examined in future work by documenting time spent interacting with males for an individual female in groups of different sizes.
Overall, our results indicate that patterns of vigilance in Przewalski's gazelle reflect a complex interplay between reproductive status, social rank, sex, and group size. Similar studies across a broad range of ecological factors will shed more light on factors underlying vigilance in animals and the tradeoff between vigilance and other crucial activities.
Overall effects of reproductive status, social rank, sex, group size and interactions between factors on percentage time spent scanning in Przewalski's gazelle were tested using a linear model (PROC GLM in SAS).
Overall effects of reproductive status, social rank, sex, group size and interactions between factors on mean scan duration in Przewalski's gazelle were tested using a linear model (PROC GLM in SAS).
Overall effects of reproductive status, social rank, sex, group size and interactions between factors on scan frequency in Przewalskis' gazelle were tested using a generalized linear model with a negative-binomial error structure and a logarithmic link function (PROC GENMOD in SAS).
We thank the Administration of Qinghai Lake National Nature Reserve and the Wildlife & Nature Reserve Management Bureau of Qinghai Province for support. We are grateful to Wu Yonglin and Shi Yong for their assistance in the field work and Li Xinhai for help with data analysis. We thank two anonymous reviewers for useful comments on the paper.
Conceived and designed the experiments: Chunlin Li ZJ ZL. Performed the experiments: Chunlin Li. Analyzed the data: Chunlin Li ZJ LL. Contributed reagents/materials/analysis tools: Chunlin Li ZJ HF Chunwang Li. Wrote the paper: Chunlin Li ZJ. Revised the manuscript and reviewed the final draft of the manuscript: Chunlin Li ZJ LL ZL HF Chunwang Li GB.
- 1. Treves A (2000) Theory and method in studies of vigilance and aggregation. Anim Behav 60: 711–722.
- 2. Caro T (2005) Antipredator defenses in birds and mammals. Chicago, US: University of Chicago Press.
- 3. Illius AW, Fitzgibbon C (1994) Costs of vigilance in foraging ungulates. Anim Behav 47: 481–484.
- 4. Fortin D, Boyce MS, Merrill EH, Fryxell JM (2004) Foraging costs of vigilance in large mammalian herbivores. Oikos 107: 172–180.
- 5. Inger R, Bearhop S, Robinson JA, Ruxton G (2006) Prey choice affects the trade-off balance between predation and starvation in an avian herbivore. Anim Behav 71: 1335–1341.
- 6. Childress MJ, Lung MA (2003) Predation risk, gender and the group size effect: does elk vigilance depend upon the behaviour of conspecifics? Anim Behav 66: 389–398.
- 7. Winnie J, Creel S (2007) Sex-specific behavioural responses of elk to spatial and temporal variation in the threat of wolf predation. Anim Behav 73: 215–225.
- 8. Lung MA, Childress MJ (2007) The influence of conspecifics and predation risk on the vigilance of elk (Cervus elaphus) in Yellowstone National Park. Behav Ecol 18: 12–20.
- 9. Frid A (1997) Vigilance by female Dall's sheep: interactions between predation risk factors. Anim Behav 53: 799–808.
- 10. Shorrocks B, Cokayne A (2005) Vigilance and group size in impala (Aepyceros melampus Lichtenstein): a study in Nairobi National Park, Kenya. Afr J Ecol 43: 91–96.
- 11. Underwood R (1982) Vigilance behaviour in grazing African antelopes. Behaviour 79: 81–107.
- 12. Beauchamp G (2008) What is the magnitude of the group-size effect on vigilance? Behav Ecol 19: 1361–1368.
- 13. Fairbanks B, Dobson FS (2007) Mechanisms of the group-size effect on vigilance in Columbian ground squirrels: dilution versus detection. Anim Behav 73: 115–123.
- 14. Lima SL, Bednekoff PA (1999) Back to the basics of antipredatory vigilance: can nonvigilant animals detest attack? Anim Behav 58: 537–543.
- 15. Rieucau G, Giraldeau LA (2009) Group size effect caused by food competition in nutmeg mannikins (Lonchura punctulata). Behav Ecol 20: 421–425.
- 16. Beauchamp G, Ruxton G (2003) Changes in vigilance with group size under scramble competition. Am Nat 161: 672–675.
- 17. Lei R, Jiang Z, Liu B (2001) Group pattern and social segregation in Przewalski's gazelle (Procapra przewalskii) around Qinghai Lake, China. J Zool (Lond) 255: 175–180.
- 18. Pepin D, Cargnelutti B, Gonzalez G, Joachim J, Reby D (2001) Diurnal and seasonal variations of roaring activity of farmed red deer stags. Appl Anim Behav Sci 74: 233–239.
- 19. Richard E, Morellet N, Cargnelutti B, Angibault JM, Vanpé C, et al. (2008) Ranging behaviour and excursions of female roe deer during the rut. Behav Process 79: 28–35.
- 20. Rieucau G, Martin JGA (2008) Many eyes or many ewes: vigilance tactics in female bighorn sheep Ovis canadensis vary according to reproductive status. Oikos 117: 501–506.
- 21. Liley S, Creel S (2008) What best explains vigilance in elk: characteristics of prey, predators, or the environment? Behav Ecol 19: 245–254.
- 22. Balmford A, Turyaho M (1992) Predation risk and lek-breeding in Uganda kob. Anim Behav 44: 117–127.
- 23. Reboreda JC, Fernandez GJ (1997) Sexual, seasonal and group size differences in the allocation of time between vigilance and feeding in the greater rhea, Rhea americana. Ethology 103: 198–207.
- 24. Guillemain M, Caldow RWG, Hodder KH, Goss-Custard JD (2003) Increased vigilance of paired males in sexually dimorphic species: distinguishing between alternative explanations in wintering Eurasian wigeon. Behav Ecol 14: 724–729.
- 25. Festabianchet M (1988) Seasonal range selection in bighorn sheep: Conflicts between forage quality, forage quantity, and predator avoidance. Oecologia 75: 580–586.
- 26. Michelena P, Deneubourg JL (2011) How Group Size Affects Vigilance Dynamics and Time Allocation Patterns: The Key Role of Imitation and Tempo. PLoS ONE 6: e18631.
- 27. Jiang Z, Li C, Zeng Y, Fredrik W (2004) “Harem defending” or “challenging”: alternative individual mating tactics in Père David's deer under different time constraint. Acta Zool Sin 50: 706–713 (in Chinese).
- 28. Willisch C, Ingold P (2007) Feeding or resting? The strategy of rutting male Alpine chamois. Ethology 113: 97–104.
- 29. Wolff JO (1998) Breeding strategies, mate choice, and reproductive success in American bison. Oikos 83: 529–544.
- 30. Wolff JO, Van Horn T (2003) Vigilance and foraging patterns of American elk during the rut in habitats with and without predators. Can J Zool 81: 266–271.
- 31. Marino A, Baldi R (2008) Vigilance patterns of territorial guanacos (Lama guanicoe): The role of reproductive interests and predation risk. Ethology 114: 413–423.
- 32. You Z, Jiang Z, Li C (2011) Location of rut stands vs. mating opportunities in Przewalski's gazelle: A field test of the “Resource-based Hypothesis” and “Female Traffic Version of the Hotspot Hypothesis”. Curr Zool 57: 701–708.
- 33. Li C, Jiang Z, Feng Z, Yang X, Yang J, et al. (2009) Effects of highway traffic on diurnal activity of the critically endangered Przewalski's gazelle. Wildl Res 36: 379–385.
- 34. Li Z, Jiang Z, Beauchamp G (2009) Vigilance in Przewalski's gazelle: effects of sex, predation risk and group size. J Zool (Lond) 277.
- 35. IUCN SSC Antelope Specialist Group (2008) Procapra przewalskii. In IUCN Red List of Threatened Species: Version 2009.1. Available from http://www.iucnredlist.org (accessed September 20, 2011).
- 36. Jiang Z (2004) Przewalski's gazelle. Beijing, China (in Chinese): China Forestry Publishing House.
- 37. Leslie DM Jr, Groves CP, Abramov AV (2010) Procapra przewalskii (Artiodactyla: Bovidae). Mamm Species 42: 124–137.
- 38. Jiang Z, Lei R, Han X, Li C (2003) A review on the researches of Przewalski's gazelle. Chin J Zool 38: 129–132 (in Chinese with English abstract).
- 39. Yang J, Jiang Z, Zeng Y, Turghan M, Fang H, et al. (2011) Effect of anthropogenic landscape features on population genetic differentiation of Przewalski's gazelle: Main role of human settlement. PLoS ONE 6: e20144.
- 40. Li C, Jiang Z, Ping X, Cai J, You Z, et al. (2010) Current status and conservation of Przewalski's gazelle Procapra przewalskii. Oryx: (in press).
- 41. Li Z (2008) Competition and coexistence mechanisms of sympatric Przewalski's gazelle and Tibetan gazelle in upper Buha River, Qinghai-Tibet Plateau. PhD thesis, Institute of Zoology, Chinese Academy of Sciences, Beijing, China.
- 42. Ruckstuhl KE, Neuhaus P (2005) Sexual segregation in vertebrates: ecology of the two sexes. Cambridge, UK: Cambridge University Press.
- 43. Martin P, Bateson P (1993) Measuring behaviour: an introductory guide. Cambridge, UK: Cambridge University Press.
- 44. Li Z, Jiang Z (2008) Group size effect on vigilance: evidence from Tibetan gazelle in Upper Buha River, Qinghai-Tibet Plateau. Behav Process 78: 25–28.
- 45. You Z, Jiang Z (2005) Courtship and mating behaviors in Przewalski's gazelle Procapra przewalskii. Acta Zool Sin 51: 187–194 (in Chinese with English abstract).
- 46. Jacobsen OW, Ugelvik M (1994) Effects of presence of waders on grazing and vigilance behavior in breeding wigeon, Anas penelope. Anim Behav 47: 488–490.
- 47. Li Z, Jiang Z, Beauchamp G (2010) Nonrandom mixing between groups of Przewalski's gazelle and Tibetan gazelle. J Mammal 91: 674–680.
- 48. Stankowich T (2008) Ungulate flight responses to human disturbance: a review and meta-analysis. Biol Conserv 141: 2159–2173.
- 49. Shih HT, Mok HK (2000) ETHOM: event-recording computer software for the study of animal behavior. Acta Zool Taiwanica 11: 47–61.
- 50. Pelletier F, Mainguy J, Cote SD (2009) Rut-induced hypophagia in male bighorn sheep and mountain goats: foraging under time budget constraints. Ethology 115: 141–151.
- 51. Mooring MS, Patton ML, Reisig DD, Osborne ER, Kanallakan AL, et al. (2006) Sexually dimorphic grooming in bison: the influence of body size, activity budget and androgens. Anim Behav 72: 737–745.
- 52. Mooring MS, Patton ML, Lance VA, Hall BM, Schaad EW, et al. (2004) Fecal androgens of bison bulls during the rut. Horm Behav 46: 392–398.
- 53. Li C, Jiang Z, Jiang G, Fang J (2001) Seasonal changes of reproductive behavior and fecal steroid concentrations in Père David's deer. Horm Behav 40: 518–525.
- 54. Beauchamp G (2003) Group-size effects on vigilance: a search for mechanisms. Behav Process 63: 111–121.
- 55. Beauchamp G, Livoreil B (1997) The effect of group size on vigilance and feeding rate in spice finches (Lonchura punctulata). Can J Zool 75: 1526–1531.
- 56. Reale D, Bousses P, Chapuis JL (1996) Female-biased mortality induced by male sexual harassment in a feral sheep population. Can J Zool 74: 1812–1818.
- 57. Sundaresan SR, Fischhoff IR, Rubenstein DI (2007) Male harassment influences female movements and associations in Grevy's zebra (Equus grevyi). Behav Ecol 18: 860–865.
- 58. Tchabovsky AV, Krasnov BR, Khokhlova IS, Shenbrot GI (2001) The effect of vegetation cover on vigilance and foraging tactics in the fat sand rat Psammomys obesus. J Ethol 19: 105–113.
- 59. Armitage KB, Salsbury CM, Barthelmess EL, Gray RC, Kovach A (1996) Population time budget for the yellow-bellied marmot. Ethol Ecol Evol 8: 67–95.
- 60. Clutton-Brock TH, Guinness FE, Albon SD (1982) Red deer: behaviour and ecology of two sexes. Chicago: University of Chicago Press.
- 61. Bateson P (1983) Mate choice. Cambridge, UK: Cambridge University Press.
- 62. Clutton-Brock T, Deutsch J, Nefdt R (1993) The evolution of ungulate leks. Anim Behav 46: 1121–1138.
- 63. Lima SL, Dill LM (1990) Behavioral decisions made under the risk of predation - a review and prospectus. Can J Zool 68: 619–640.
- 64. Liu B, Qian Z, Zhang B, Liu S, Li M (2009) Diurnal activity budgets and energy balance strategy of Mongolian gazelle (Procapra gutturosa) before, during and after the rut around Dalai Lake, Inner Mongolia. Zool Res 30: 694–698.
- 65. Laundré JW, Hernández L, Altendorf KB (2001) Wolves, elk, and bison: reestablishing the “landscape of fear” in Yellowstone National Park, USA. Can J Zool 79: 1401–1409.
- 66. Gutzwiller KJ, Marcum HA, Harvey HB, Roth JD, Anderson SH (1998) Bird tolerance to human intrusion in Wyoming montane forests. Condor 100: 519–527.
- 67. Li C, Jiang Z, Tang S, Zeng Y (2007) Evidence of effects of human disturbance on alert response in pere david's deer (Elaphurus davidianus). Zoo Biol 26: 461–470.
- 68. Taylor AR, Knight RL (2003) Wildlife responses to recreation and associated visitor perceptions. Ecol Appl 13: 951–963.
- 69. Yoccoz NG, Mysterud A, Langvatn R, Stenseth NC (2002) Age- and density-dependent reproductive effort in male red deer. Proc R Soc Lond B 269: 1523–1528.
- 70. McElligott AG, Naulty F, Clarke WV, Hayden TJ (2003) The somatic cost of reproduction: what determines reproductive effort in prime-aged fallow bucks? Evol Ecol Res 5: 1239–1250.
- 71. Beauchamp G (2001) Should vigilance always decrease with group size? Behav Ecol Sociobiol 51: 47–52.
- 72. Favreau FR, Goldizen AW, Pays O (2010) Interactions among social monitoring, anti-predator vigilance and group size in eastern grey kangaroos. Proc R Soc B 277: 2089–2095.