Ethology 110, 681—691 (2004) 2004 Blackwell Verlag, Berlin ISSN 0179–1613
Relationship between Serum Testosterone, Dominance and Mating Success in Pe`re David’s Deer Stags Li Chunwang, Jiang Zhigang, Zeng Yan & Yan Caie Institute of Zoology, Graduate School of the Chinese Academy of Sciences, Beijing, China
Abstract We conducted an experiment in the Beijing Milu Park to study the social behavior of male Pe`re David’s deer, and related social behavior to social position and serum testosterone level of the stags during rut. We classified the stags into three rank classes according to their rutting behavior: Ôharem masterÕ, ÔchallengerÕ and ÔbachelorÕ. We monitored the behaviors of four Ôharem mastersÕ, five ÔchallengersÕ and five ÔbachelorsÕ, and analyzed serum testosterone levels in blood samples of those 14 stags using radioimmunoassay. We defined the effectiveness value, E ¼ A/T, to assess the effectiveness of herding or mating attempts made by stags (ÔTÕ represents the frequency of herding or mating attempts made by a stag and ÔAÕ represents the frequency of herding or mating attempts accepted by hinds). We found that: (1) the Ôharem mastersÕ and the ÔchallengersÕ displayed more frequent rut and locomotive behaviors but fewer ingestion behaviors than the ÔbachelorsÕ; (2) serum testosterone levels in the Ôharem mastersÕ and the ÔchallengersÕ were higher than that in the ÔbachelorsÕ; (3) effectiveness value of herding attempts differed significantly between the three types of stags, being highest in the Ôharem mastersÕ and the lowest in the ÔbachelorsÕ; and (4) effectiveness value of mating attempts was significantly greater for the Ôharem mastersÕ than for the ÔchallengersÕ. We conclude that: (1) reproductive behavior of the Pe`re David’s deer stags is strongly associated with social rank; (2) social roles of Pe`re David’s deer stags during the rut are related to the testosterone secretion; and (3) rank class affects the mating opportunity of the stags. Correspondence: Jiang Zhigang PhD, Graduate of the Chinese Institute ofSchool Zoology, Academy of Sciences, Beijing 100080, China. E-mail:
[email protected] Introduction Most ungulates, such as red deer (Cervus elaphus), white-tailed deer (Odocoileus virginianus), pudu (Pudu puda), antelope and domestic horse, breed by establishing a hierarchical system (Clutton-Brock et al. 1982; Gosling 1986; U. S. Copyright Clearance Center Code Statement: 0179-1613/2004/11009–681/$15.00/0 www.blackwell-synergy.com
682
L. Chunwang, J. Zhigang, Z. Yan & Y. Caie
Miller et al. 1987; McDonnell & Murray 1995; Barto et al. 1998). Darwin (1871) pointed out that male competition for access to a limited number of females often results in high levels of male–male aggression. In cervids, many social behaviors that are associated with reproduction are regulated by gonadal steroid hormones (Miller et al. 1987; Nelson et al. 1990; Barto et al. 1998). However, data on relationships between social behaviors, social rank, and endocrine factors in Pe`re David’s deer (Elaphurus davidianus) are limited. Pe`re David’s deer display different social behavioral patterns seasonally (Li et al. 2001). As in red deer (Clutton-Brock et al. 1982), Pe`re David’s deer have a social hierarchy that becomes more marked during the rut (Schaller & Hamer 1978). Jiang (1999a) found that Pe`re David’s deer stags in different age groups adopt different reproductive strategies, and suggested that reproductive tactics of Pe`re David’s deer are age-dependent. Unanswered questions for the social behavior of Pe`re David’s deer include: (1) what is the relationship between social behavior and social rank? (2) what role does social rank play in breeding? In particular, do stags of different social rank have different behavioral strategies during the breeding season? Does the social rank of a stag affect its preference by hinds? (3) Is social rank related to gonadal steroid hormone secretion? In this research we focused on the relationships between social behavior, social status and serum testosterone levels in Pe`re David’s deer stags during rut. We hypothesized that testosterone is associated positively with the social rank of stags. We also hypothesized that the mating success of stags and the preference for stags by hinds in estrus is positively correlated with the stag’s social rank. To test the hypotheses, we compared the behaviors of the top-ranking stags with other stags and examined their serum testosterone levels. Intersexual selection of females exists in many ungulate species with a polygynous mating system (Clutton-Brock 1989). Hinds in heat exhibit estrus behaviors such as frequent urinating and receptivity (Li et al. 2001). Furthermore, the hinds in estrus also choose stags in response to their sexual advance, and preference for stags by hinds could reflect the status of stags (Clutton-Brock 1989). Thus, to examine the effectiveness of rutting behavior, we recorded the acceptance or rejection to the advances of stags by the hinds in estrus.
Methods Study Site and Animals
The Beijing Milu Park (397¢N, 11603¢E) has an area of 60 ha. Annual average temperature is 13.1C, with mean temperature of )3.4C in January and 26.4C in July. Average annual precipitation is about 600 mm. Dominant species in the vegetation are Eleusine indica, Eragrostis cilianensis, Digitaria sangunalis and Setaria viridis. A group of 20 Pe`re David’s deer was reintroduced to this Park from Woburn Abbey, England with the help of the Duke of Bedford in 1985. A second group of 17 Pe`re David’s deer joined the herd in 1987 (Jiang et al. 2000a).
Serum Testosterone, Dominance and Mating Success in Pe`re David’s Deer Stags
683
Pe`re David’s deer in the study population graze on natural vegetation in summer and autumn and they rely on supplementary feeds in winter and spring. There was a total of 144 deer in the Park in 2000, including 32 adult males, 58 adult females, 23 2-yr olds and yearlings and 31 newborns. Behavior Data Collection
We adopted the nomenclature used by Hogg & Forbes (1997) in describing male social status: (1) Ôharem mastersÕ, dominant stags who hold harems, (2) ÔchallengersÕ, without their own harems, challenge the Ôharem masterÕ and try to hold hinds. (3) ÔbachelorsÕ, who form the Ôbachelor groupÕ and stay away from female groups during the breeding season. A Pe`re David’s deer stag changes its social status from ÔbachelorÕ to ÔchallengerÕ, then to Ôharem masterÕ. Some of the stags may ultimately become Ôharem mastersÕ, but some of them may never reach the highest social rank during their lifetime (Jiang 1999a). During field observation, we observed four Ôharem mastersÕ, five ÔchallengersÕ and five ÔbachelorsÕ from 10 June to 31 August 2000, namely the whole rutting season. The ages of all stags were 4–6 yr. To avoid the problems of nonindependence in data, we ceased placing a stag in its previous social status group (for purposes of analysis) once it changed its rank, and chose another stag that corresponded with its previous social status to replace it. Thus, once the candidate stag joined one social status group, it was not included in the analysis of its previous groups. For example, the group of Ôharem masterÕ was composed of the first Ôharem masterÕ and three Ôharem mastersÕ that transformed from ÔchallengersÕ. Those three stags were not included in the ÔchallengerÕ group when we performed data analysis. To avoid the probable abnormality of behavioral patterns induced by anesthetics, we also stopped observing the stags immediately when they were captured. We resumed observation 4 d after blood sampling. We identified the individuals by dairy cattle ear tags that had been applied when they were calves. For those stags without ear tags or when ear tags were covered with mud, antler shape and facial characteristics were employed to identify individuals. At 8:00, 12:00 and 16:00 hours each day, we performed 10min observations on the whole group to find if the Ôharem masterÕ had changed its social status or not. We carried out behavioral observations every 5 d. On each observation day, we used the focal observation method to record the behaviors of the deer every 20 min (focal observation for each deer lasted 1.5 min) from 5:00 to 19:00 hours with SJ-1 Event Recorder (Jiang 1999b). All individuals could be seen at any time because of the relatively small paddock. The behavior recordings were made by focal observation with the following behavior categories described by Jiang (2000): Rut behavior. Anogenital sniffing, urine sniffing, flehmen, bellowing, wallowing, antler adorning, antler swags (grubs) mud, chasing hinds, herding hinds, chin resting, urine spraying, and pre-orbital gland marking. Mating behavior. Mounting and copulating. Agonistic behavior. Fighting and chasing other stags.
684
L. Chunwang, J. Zhigang, Z. Yan & Y. Caie
Ingestive behavior. Feeding and drinking. Rest behavior. Rest kneeling, sitting and lying down. Locomotive behavior. Stepping, walking and running. Standing. Standing. Others. All other behaviors not listed above. We recorded all herding and mating attempts made by each stag we observed, and the responses of hinds. The herding attempts include chasing and herding hinds. The mating attempts include mounting and copulating. The responses of hinds include accepting and rejecting. Herding. A stag chases hinds and tries to make a harem. Accepting. A hind shows positive response without escaping while a stag is herding or mating. Rejecting. A hind shows negative response and runs away while a stag is herding or mating. Blood Sample Collection and Hormone Assay
Eight ÔchallengersÕ appeared during the whole rut season in 2000. However, to ensure that the sampling procedure was independent, we captured each stag only once. The Ôharem mastersÕ were sampled while they were masters; ÔchallengersÕ and ÔbachelorsÕ were sampled in the same way. We sampled all the stags we observed on 24 July (one harem master, one challenger and one bachelor), 30 July (one harem master, one challenger and two bachelors), 2 August (one harem master, two challengers and one bachelor) and 12 August (one harem master, one challenger and one bachelor). All animals in this study were cared for in accordance with the principles and guidelines of the Chinese Wildlife Management Authority. Each stag was immobilized between 6:00 and 8:00 hours by Miannaining (the major component of which was xylazine hydrochloride, Veterinary Institute, Changchun University of Agriculture and Animal Science, China) (1.5–2.0 ml/100 kg) with a G.U.T.50 Universal Gas Applicator (Teleject GMBH, Ro¨merberg, Germany). As quickly as possible (usually within 15 min), we collected 10 ml blood samples from the jugular vein into tubes. We kept the blood samples in a cooler until they were centrifuged (within 1 h). After centrifugating each sample at 2000 g for 15 min at room temperature, we transferred 2 ml serum to a tube and kept it at )20C for further assay. After blood sampling, the deer was injected intramuscularly with Suxingling No. 4 (the major component of which was yohimbine hydrochloride; Veterinary Institute, Changchun University of Agriculture and Animal Science, China) (2.5–4.0 ml/ 100 kg) to speed up recovery from anesthesia. To ensure the animals recovered well, we kept watch on the stags that we sampled for the following 3 d, so that we could find any abnormality and help them in a timely manner. No further interference was allowed to disturb these stags. Three days later, we found that the stags behaved normally, with an increase in ingestive behavior.
Serum Testosterone, Dominance and Mating Success in Pe`re David’s Deer Stags
685
Steroid Radioimmunoassay
We used iodinated steroid radioimmunoassay kits (Beijing North Institute of Biological Technology, Beijing, China) to determine the hormone concentration in the processed serum samples. We added duplicate 50 ll serum samples to sample tubes, and duplicate 50 ll aliquots of testosterone standards to other tubes. Then we added 100 ll aliquots of antiserum to the standards and the samples. In addition, we set up the tubes containing 100 ll buffer and 50 ll Ô0Õ testosterone standards but no antiserum in duplicate for estimation of non-specific binding and total radioactivity. For the steroids assay, we added 100 ll 125I-testosterone (approx. 0.25 lCi/ml) to all tubes and incubated the whole overnight at 4C. After the addition of 500 ll charcoal-dextran [0.125% Norit-A Charcoal (Beijing North Institute of Biological Technology, Beijing, China), 0.0125% dextran in 0.1 m phosphate-buffered saline (PBS), pH 7.0] and a 30-min incubation at 25C, we centrifuged tubes at 3500 g for 15 min, and used a SN-682 radioimmunoassay gamma counter (Shanghai Hefu Photo-electricity Instrument Co. Ltd, Shanghai, China) to count the radioactivity of the supernatant fraction for 2 min. We defined sensitivities by 10% displacement from the B0 binding. Sensitivity of testosterone was 0.1 ng/ml. Intra- and interassay coefficient of variation of testosterone was
