Behavioral Ecology Advance Access originally published online on January 25, 2006
Behavioral Ecology 2006 17(3):364-371; doi:10.1093/beheco/arj039
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Grooming in desert bighorn sheep (Ovis canadensis mexicana) and the ghost of parasites past
a Department of Biology, Point Loma Nazarene University, 3900 Lomaland Drive, San Diego, CA 92106, USA and b Department of Anatomy, Physiology and Cell Biology, School of Veterinary Medicine, University of California-Davis, Davis, CA 95616, USA
Address correspondence to M.S. Mooring. E-mail: mikemooring{at}ptloma.edu.
Received 15 July 2005; revised 13 December 2005; accepted 19 December 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Ectoparasites such as ticks have a negative effect on host fitness, whereas parasite-defense grooming is effective in removing ticks. The central control (programmed grooming) model proposes that animals engage in preventive tick-defense grooming in response to an internal timing mechanism, even in the absence of peripheral stimulation from parasites. This model predicts that smaller animals will groom more frequently than larger ones because of the higher cost of parasitism for a small animal (body size principle). The peripheral stimulation (stimulus driven) model predicts no size-related differences in grooming rate in the absence of tick bite irritation. We observed grooming behavior in a Chihuahuan desert population of bighorn sheep (Ovis canadensis mexicana), where ticks have been absent for perhaps thousands of years. Although not exposed to ticks, bighorns self groomed by means of oral and scratch grooming, albeit at very low rates compared to size-matched ungulates in both tick-infested and tick-free environments. Logistic regression and general linear models revealed both the probability that grooming was performed during a 10-min focal sample and the rate of grooming when it occurred was greater for younger, smaller age/sex categories of less body mass. Oral and scratch grooming were negatively associated with body mass during both years, with juveniles (X = 15 kg) grooming the most frequently and the oldest males (X = 70–85 kg) grooming the least. Assuming that programmed grooming evolved in a tick-infested environment, the current grooming behavior of this population is a relict of their ancestral environment, an adaptation to the "ghost of parasites past."
Key words: body size principle, desert bighorn, grooming, programmed grooming model, ticks.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Grooming is among the most commonly performed behavior patterns in mammals and other vertebrates. Although grooming can serve a diversity of functions, the most basic are the cleaning and maintenance of the pelage and the removal of ectoparasites (Hart et al., 1992
; Moore, 2002). Ticks are the most important ectoparasites of wild animals (Allan, 2001), and a substantial body of evidence points to the cost of ticks as a central force for the evolution of grooming in the natural environment (Allan, 2001; Hart, 1990; Mooring et al., 2004a). Tick-associated declines in growth are well documented for growing cattle (Kaiser et al., 1991; Little, 1963; Sutherst et al., 1983). For example, a moderate tick load on a calf can result in a 10- to 44-kg reduction in weight gain per year due to blood loss and tick-induced anorexia (Norval et al., 1988). A similar loss of reserves in a wild animal would be expected to have fitness-compromising consequences.
The efficacy of grooming in removing ectoparasites has been established through experimental studies in which grooming was restricted (e.g., Hart, 1990; Koch, 1981, 1988). In ungulates, this efficacy is illustrated by an experiment in which impala (Aepyceros melampus) were fitted with a neck harness that largely prevented oral self grooming. Harnessed impala harbored 20 times more adult female ticks than control impala wearing harnesses that permitted grooming (Mooring et al., 1996a). Given the costs of ectoparasite infestation, individuals with effective grooming behavior would be at a selective advantage. However, grooming also entails costs, such as compromised vigilance against predators (Maestripieri, 1993; Mooring and Hart, 1995a) and conspecifics (Mooring and Hart, 1995b; Mooring et al., 1996b), saliva loss from oral grooming (Ritter and Epstein, 1974), attrition of dental elements when teeth are used to groom (McKenzie, 1990), and hair loss from excessive scratching (Mooring and Samuel, 1999). An optimal grooming rate would, therefore, balance the cost of ticks against the costs of grooming.
Two physiological mechanisms are envisioned in the mediation of grooming: central control (Colbern and Gispen, 1988; Fentress, 1988; Nelson et al., 1975; Roth and Rosenblatt, 1967; Spruijt et al., 1992) and peripheral stimulation (Riek, 1962; Wikel, 1984; Willadsen, 1980). Although central and peripheral mechanisms operate concurrently, we assume that one predominates in the natural environment, overriding the other in the event of conflict. The mechanism involving peripheral stimulation, the "stimulus-driven model," postulates that grooming is a response to cutaneous stimulation, such as irritation from tick bites, probably triggered by the release of histamine from dermal mast cells at the site of the bite (Willadsen, 1980). The cutaneous irritation occurs after the ectoparasite has attached and started to feed. The mechanism of central control, the "programmed grooming model," postulates that a central regulatory or timing system periodically evokes bouts of grooming, which has the effect of removing ticks before they attach and begin to feed (Hart et al., 1992). Because engorging ticks are costly, preventive removal of ticks before they engorge would be adaptive. The recent finding that a mutant gene of Hoxb8 triggers excessive grooming in mice (Greer and Capecchi, 2002) supports the concept of a grooming control center in the central nervous system.
The programmed grooming model makes predictions that contrast with those of the stimulus-driven model (Hart et al., 1992). One prediction of programmed grooming is that animals grooming the most will have the fewest ticks (due to preventive removal), whereas the stimulus-driven model predicts that animals with the most ticks will groom the most (due to cutaneous irritation). The programmed grooming model also predicts that animals of smaller body size will groom more frequently than those of larger size as a reflection of their greater surface area-to-mass ratio (the body size principle). Because smaller animals have less blood volume per unit of surface area, they pay a higher cost in blood loss per given density of ticks. Assuming an equal rate of tick infestation, small-bodied animals should groom at a higher rate than larger animals to maintain a lower density of ticks. This should occur even in a tick-free environment. The stimulus-driven model, on the other hand, would predict that larger animals (with a greater number of ticks on their larger body surface) will groom more. The results of several studies support the predictions of the programmed grooming model, showing an inverse relationship between grooming rate and tick density (Mooring and Hart, 1995b; Mooring et al., 1996a,b; Olubayo et al., 1993) and an inverse relationship between body size and frequency of grooming for both interspecific comparisons (Hart et al., 1992; Mooring et al., 2000, 2004a) and developmental stages within the same species (Hart and Pryor, 2004; Mooring and Hart, 1997a,b; Mooring and Samuel, 1998a,b; Mooring et al., 2002).
The desert bighorn sheep (Ovis canadensis mexicana) of the Red Rock Wildlife Area (RRWA) in the Chihuahuan desert of New Mexico provide a unique opportunity to test the programmed grooming model and to examine the interaction of the costs and benefits of grooming. Throughout most of the American southwest, bighorn sheep are parasitized primarily by Dermacentor hunteri, a desert-adapted tick that in the adult stage is found almost exclusively in and around the ears (Crosbie et al., 1997). However, these ticks are absent in the Chihuahuan desert, apparently because the desert woodrat (Neotoma lepida), which hosts the immature tick stages, is absent (Crosbie et al., 1997). There is reason to believe that bighorns of RRWA have lived for multiple generations, perhaps thousands of years, without exposure to ticks (see Discussion). The absence of ticks not only eliminates one of the primary benefits of grooming, but the modulation of grooming by tick saliva or histamine (Hart, 1997) would also be absent. On the other hand, predators are abundant, maintaining an important and persistent cost of grooming in terms of distraction from antipredator vigilance (Mooring et al., 2004b).
The unique population of bighorns at Red Rock allowed us to make two testable predictions arising from the programmed grooming model. As mentioned, previous studies have established that grooming distracts from vigilance (Cords, 1995; Maestripieri, 1993; Mooring and Hart, 1995a). One prediction is that, with risk of tick infection absent as a primary benefit of grooming, but with risk of predation present as a continuing cost, the central regulatory mechanism would evolve to produce a much lower rate of grooming than comparable animals in tick-infested environments. The second prediction is that some minimal level of grooming would persist despite the current absence of ticks, reflecting the importance of grooming in ancestral bighorns that lived in a tick-infested environment. Because of the ghost of parasites past (cf. Byers, 1997), we predicted that the rate of grooming would follow the body size principle seen in other ungulates (including ancestral bighorn sheep), in which younger and smaller animals groom significantly less than their larger adult conspecifics.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study site and animals
The RRWA in the Chihuahuan desert of New Mexico (32° 44' N, 108° 41' E) consists of 5.4 km2 of rolling hills and steep cliffs along the western banks of the Gila River at 1280- to 1495-m elevation. The desert scrub habitat is dominated by scattered juniper (Juniperus monosperma) and mesquite (Prosopis glandulosa, Prosopis velutina) trees and shrubs of whitethorn (Acacia constricta), catclaw (Acacia gregii), wait-a-minute bush (Mimosa biuncifera), sotol (Dasylirion wheeleri), ocotillo (Fouquieria splendens), yucca (Yucca baccata), and cholla and prickly pear cacti (Opuntia spp.). The population of bighorn sheep inhabiting the RRWA was established in 1972 by the New Mexico Department of Game and Fish (NMDGF) as a breeding population to augment free-ranging populations and reintroduce sheep to traditional habitat. Bighorn at RRWA are derived from native stock from the San Andres Mountains, New Mexico (three rams, 13 ewes), and Sonora, Mexico (five ewes) (Monson, 1980
). Over 200 capture-related examinations by NMDGF confirmed the absence of ticks and mites on bighorns (NMDGF, unpublished data).
The RRWA facility is surrounded by a 2.6 m woven game fence, with interior fences that divide the reserve into five pastures; bighorn pass freely though gates in the interior fencing, and thus have access to all five pastures. Although the presence of the game fence prevents bighorn from leaving the reserve, it does not prevent other animals from entering. Peccaries (Pecari tajacu) regularly dig under the perimeter fence to the reserve, providing entrance for predators such as coyote (Canis latrans), bobcat (Lynx rufus), and mountain lion (Felis concolor); mountain lions can also enter by climbing the wooden fence posts. Mountain lions have been the most important predators of adult bighorn in recent years (Mooring et al., 2004b; NMDGF, unpublished data).
Most populations of desert bighorn range over vast, rugged areas in small groups that are intolerant of human presence and extremely difficult to find and observe (Krausman and Shackleton, 2000). Because the sheep at Red Rock are confined to the reserve by a perimeter game fence, population density tends to be high for desert bighorn (14–20 sheep per km2 in recent years). This allowed us to locate and observe a significant proportion (>50%) of the population on a daily basis without radio telemetry and provided a unique opportunity to collect behavioral data. Bighorn populations are typically sexually segregated, with males and females occupying separate seasonal ranges, although with temporal and spatial overlap (Krausman et al., 1999; Shackleton et al., 1999). Previous work on this population has revealed that this is the case for RRWA bighorn sheep (Mooring et al., 2003). Because male and female groups at Red Rock overlapped in space use, and no ticks have ever been recovered from either sex, we believe sexual segregation had no influence on ectoparasite exposure.
This study was conducted over 2 years (year 1: June–July 1999; year 2: June–August 2000). The bighorn population size varies from year to year due to predation and other natural mortality and the periodic capture and removal of animals to augment other populations. Annual ground surveys conducted in late April of each year of our study by NMDGF indicated a population of 108 bighorns in year 1 and 76 in year 2. Because we could not capture and weigh the animals under observation, we used the mean mass for each age/sex category from the review by Hansen and Deming (1980) in the analysis.
Behavioral observations
Work was done primarily on foot, using the 30 km of primitive road system or hiking cross-country; we sometimes used a four-wheel drive all terrain vehicle (ATV). Whenever we sighted a herd of bighorn we recorded group size, composition, and the map reference location. The following age/sex categories were recorded, based on body size and horn configuration (Geist, 1971; Hansen and Deming, 1980): juveniles (lambs < 1 year), yearling females (1-year-old ewes), yearling males (1-year-old rams), females (adult ewes 
2 years), and males (adult rams 2 years). Adult rams were further categorized by horn curl as either Class 2 (half curl), Class 3 (three-fourth curl), or Class 4 (three-fourth to full curl) males, with horn thickness and brooming (wear) of tips assisting in identification of ram class (Geist, 1971; Hansen and Deming, 1980). Class 3 and Class 4 males were later collapsed into one category (Class 3/4) because many of the rams were intermediate in horn configuration, and assigning a particular class would have been arbitrary.
Behavioral observations were opportunistic. When we sighted a group of bighorns, we conducted 10-min focal samples on group members as long as they stayed in sight. The selection of what individuals to observe and when was random, and the likelihood of repeat sampling of the group at the time of observation was slight because group members would be spread out and we could "identify" individuals in the short term by their location. Observations were conducted primarily during the morning hours when sheep were most active; bighorn use an energy-minimizing strategy to avoid heat stress during the hot summer months and spend most of the afternoons bedded down in shade (Krausman and Shackleton, 2000; Krausman et al., 1999; Simmons, 1969, 1980). Observations were made with 10x binoculars and 15–60x spotting telescopes; sheep were unmarked and were usually 500–1000 m away.
We recorded all grooming that was observed during 10-min focal animal samples (Altmann, 1974), with focal animals chosen by random sampling from the herd, as mentioned. Ten-minute samples were chosen because sheep often did not stay in view for very long. We only conducted observations when the focal animal was standing because animals are less likely to groom when recumbent. An exception was made for lambs because they frequently alternate between lying down and standing up and often groom while reclining; observations on lambs were initiated only when the animal was standing but were continued if the lamb reclined (e.g., Mooring and Hart, 1997a).
Grooming took the form of oral grooming (using the lower incisors to scrape through the pelage) and scratching with the hoof of the hind leg. Oral grooming was directed to body regions posterior to the head and neck, and scratching was directed to the anterior portion of the body, primarily the head, neck, chest, and upper forelegs, as reported for Rocky Mountain bighorn (Geist, 1971). The number of grooming movements (i.e., oral grooming teeth scrapes or hind leg scratches) and the body area groomed were recorded. An "episode" consisted of each individual grooming motion, and "bouts" were defined as an uninterrupted sequence of episodes separated from any subsequent bout by an interval of at least 5 s and/or a switch to another body region. Grooming bouts and episodes were later extrapolated to grooming per hour and per 12-h day.
Data were written directly into notebooks and entered into laptop computers back at camp. Behavioral observations totaled 338 h over the two summers (year 1: 166 h; year 2: 172 h). Interobserver reliability tests were conducted in years 1 and 2 to assess the level of agreement among observers. For focal observation reliability tests in year 1, an experienced observer (MSM) observed the same focal animal with one other observer at a time (TAF, JEB, ICF) for 15 10-min samples. In year 2, all four observers (MSM, TAF, TTN, DDR) collected data on the same focal animal at the same time for 15 10-min samples. Pearson correlations showed that interobserver reliability exceeded 0.90 for both summers. For year 1, the mean correlation coefficient for focal samples was r = .91 (range: r = .83–1.00), while for year 2 the mean correlation coefficient was r = .93 (range: r = .82–1.0).
Statistical analysis
For analyses, we compared grooming rate measures (oral and scratch grooming bouts and episodes) with independent variables associated with differences in body mass. "Category" was coded as follows: 1 = juveniles, 2 = yearling females, 3 = adult females, 4 = yearling males, 5 = Class 2 adult males, 6 = Class 3/4 adult males. Body mass means from Hansen and Deming (1980) related to each of the categories as follows: 1 = 15 kg, 2 = 41 kg, 3 = 48 kg, 4 = 52 kg, 5 = 70 kg, 6 = 85 kg. Thus, category differences also reflect mass differences. In addition to category/mass, we included age (1 = juveniles, 2 = yearlings and adults) and sex (1 = female, 2 = male) in the model as additional independent variables to control for the confounding effect of these factors (age and sex both covary with category and mass). "Year" (1 = 1999, 2 = 2000) was also included as an independent variable in the models.
Because most subjects were not individually recognizable at the distances that observations were conducted, grooming means for individual sheep could not be calculated. Another complication was that grooming did not occur frequently, with a high percentage of 10-min focal observations including no grooming. Thus, a large number of 10-min focal observation samples were required to accurately characterize grooming rates (year 1: n = 993; year 2: n = 1030). The analytical approaches specified below were used to address problems caused by possible repeated sampling of the same individuals. Correspondingly, the statistical conclusions apply only to the herd studied.
Following the procedure of Mooring (1995), we used mean daily grooming rate for each age/sex category as our sampling unit for exploratory analysis. Because the size, composition, and location of bighorn groups varied from day to day, each day was considered a random sample from the pool of each age/sex category, which was replaced daily. Each new day of observation started out with a fresh sampling, with care taken not to duplicate any animals on that day. Because the sampling unit was a composite measure representing the characteristic grooming rate of each age/sex category of bighorn, repeat sampling of individuals was not an issue provided that some individuals were not overrepresented in the focal samples (which we took care to avoid by random sampling).
Analysis of daily means by ranked ANOVA revealed that oral grooming among age/sex categories was significantly different during both 1999 (bouts: F5,215 = 4.84, p < .0001; episodes: F5,215 = 4.87, p < .0001) and 2000 (bouts: F5,171 = 14.78, p < .0001; episodes: F5,171 = 16.25, p < .0001). We then used a generalized linear mixed model (GLMM) to investigate the influence of mass on grooming rates derived from daily means. Linear mixed models are an extension of the general linear model (GLM) that permits data to exhibit correlation and nonconstant variability (Milner et al., 1999; SPSS, 2001). Using the "Mixed Models" option in SPSS 11.5, we fitted a mixed model using mass as the fixed effect, with date, category, sex, and year as random variables. Results were similar whether mass was log transformed or untransformed. GLMM revealed that mass influenced grooming rate when sex and year were controlled for (oral bouts: F1,358 = 19.4, p = .0001; oral episodes: F1,358 = 12.7, p = .0001; scratch bouts: F1,358 = 6.7, p = .01; scratch episodes: F1,358 = 17.6, p = .0001). These exploratory results concurred with subsequent results based on focal samples (see below) and suggested that the influence of mass on grooming rate was not an artifact of repeated sampling.
Using the original data set, we ran logistic regression (Kuter et al., 2004) on all focal samples to ask whether the probability that any grooming at all was performed during 10-min focal samples varied according to age/sex category and mass of bighorn. Differences between specific pairs of categories were examined using contrasts. Subsequently, we ran a GLM ANOVA (Kuter et al., 2004) on only those 10-min focal samples in which grooming occurred, asking whether the rate of grooming varied according to bighorn category and mass. Natural log transformations were used to accommodate the assumption of normality. Multiple comparisons were performed using Fisher's least significant differences test. All analyses were two tailed, with the level of significance set as 
= .05.
To explore how grooming by desert bighorn ewes compared with females of other species, we examined previous studies and extracted female mass data from reference sources (Burt and Grossenheider, 1980; Haltenorth and Diller, 1980; MacDonald, 1984; Nowak, 1999).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bighorn sheep of all age/sex categories engaged in sheep-typical grooming, but at a very low level. In fact, in 94% of the 10-min observations (n = 2023), no grooming was observed. Those grooming most frequently were juveniles, while older males groomed the least. Combining data from both years, juveniles oral groomed during 15% of 354 10-min samples, compared with the oldest males that oral groomed during only 1% of 378 samples (Table 1). Females and yearlings oral groomed during 5–8% of their focal samples. Extrapolating to hourly rates, juveniles oral groomed at the mean rate of 1.63 bouts (15 episodes) per hour, while older males delivered a mean of 0.08 bouts (0.3 episodes), and adult females and yearlings delivered oral grooming at intermediate rates (Table 2). To examine the magnitude of grooming differences throughout daylight hours, we next converted our data to grooming per l2-h day. Juveniles delivered up to 25 bouts (235 episodes) of oral grooming per l2-h day, compared with older males that delivered a maximum of only one bout (five episodes) during the same time period.
|
|
Of the independent variables, only category/mass was significant, and analyses were significant for bouts only. For analysis of the likelihood of any oral grooming occurring during a 10-min observation, we found that the probability of oral grooming was significantly greater for younger animals of lower body mass (logistic regression: n = 2023, df = 5, Wald
2 = 57.3, p = .0001). Further analyses revealed that juveniles were significantly more likely to oral groom during a 10-min sample than were adult females and males (p < .0001), and both female yearlings (p = .049) and female adults were more likely to oral groom than were Class 2 males (p = .01) and Class 3/4 males (p = .04). Rates of grooming by male yearlings were not significantly different from those of other categories of sheep. These results, revealing that categories of bighorn sheep of lower body mass were more likely to groom in a 10-min focal observation period than those of larger size, corresponded to the body size prediction of the programmed grooming model. The analysis using just those observations in which oral grooming occurred (GLM) gave the same results (Figure 1), with category/mass the only significant factor (GLM: n = 125, df = 5, F5,119 = 3.62, p = .004). Further analysis by Fisher's least significant difference test revealed that juveniles oral groomed at a higher rate than adult females (p = .0006), and both juveniles (p = .006) and yearling females (p = .03) oral groomed at a higher rate than Class 2 adult males. All other comparisons were nonsignificant.
|
With regard to scratch grooming, logistic regression showed that there were differences among the age/sex categories (logistic regression: n = 2023, df = 5, Wald
2 = 13.1, p = .02). Further analyses showed that female adults were significantly more likely to scratch groom during a 10-min sample than were male yearlings (p = .002), and male yearlings were more likely to scratch groom than Class 2 males (p = .05) or Class 3/4 males (p = .007). There were no other significant differences among categories of sheep. Sex had no significant effect on probability of scratching. Among all age/sex categories, bighorns were more likely to scratch during year 1 compared to year 2 (logistic regression: n = 2023, df = 1, Wald 2 = 3.85, p = .05). GLM analysis of samples in which scratch grooming occurred showed that, again, there were differences among the age/sex categories in the number of scratch bouts performed (GLM: n = 134, df = 5, F5,128 = 3.91, p = .002), and Fisher's least significant differences revealed that juveniles scratched at a higher rate than yearling males (p = .02), adult females (p = .004), and adult males (Class 2: p = .0002; Class 3/4: p = .003). Both yearling females and yearling males scratched more than did Class 2 adult males (p = .04). All other comparisons were nonsignificant. While the predictions of some level of grooming, and an inverse relationship between body mass and grooming rate, were borne out, the prediction of a very low level of grooming could not be approached with a comparison to grooming by bighorn in tick-infested habitats because such grooming data are not available. As an alternative approach, we compared grooming in Red Rock bighorn females with adult females of other ungulate species that are exposed to tick infestation in their natural habitats and for which grooming rates have been published. Although no statistical comparisons could be made, data from impala and Grant's gazelle (Gazella granti), with the same approximate body mass, showed that members of these species all had grooming rates exceeding that of bighorns at Red Rock (Table 3). Elk (Cervus elaphus), moose (Alces alces), and bison (Bison bison) exhibited grooming rates that overlapped with those of bighorn at Red Rock, although these species are 5–11 times larger in mass.
|
Because these comparisons involved animals exposed to ticks that would have modulated the rate of programmed grooming above baseline, another comparison was made with ungulates of eight species of similar body size, which were housed in a tick-free zoological park or research facility (Table 4). These comparisons revealed that females of all eight species had mean grooming rates exceeding those of the Red Rock bighorn females.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Patterns of oral grooming were the same in bighorn sheep at Red Rock as seen in other ungulates, including goats (Capra hircus) and sheep (Ovis aries), in which the lower incisors are combed through the pelage in upward movements. The pattern of scratch grooming with the hooves of the hind feet were also the same (e.g., Hart and Pryor, 2004
; Mooring et al., 2000, 2002, 2004a). Because grooming in the bighorn was performed in the absence of ticks or other obvious stimuli, a central control mechanism provides the most logical explanation. The stimulus-driven model would not predict the differential rates of grooming found among age/sex categories in the present study.
Although they groomed, the bighorn did so very infrequently compared with wild populations of other species exposed to ticks of similar or greater body mass (Table 3) or ungulates of similar mass maintained in tick-free environments (Table 4). These results may relate to different selection pressures operating on the central grooming center that controls the baseline rate of grooming. Oral grooming, in which the head is turned toward the pelage, would compromise scanning for predators, and thereby increase the risk of predation. Because the Red Rock population is exposed to predators (Mooring et al., 2004b), grooming in this population carries a cost that would not be compensated for by the benefit of tick removal. If tick saliva, or some other agent of tick bite stimulation, increases grooming rate (Hart, 1997), this factor was also missing. We would predict that Sonoran or Mojave desert bighorn, which are exposed to D. hunteri ticks (Crosbie et al., 1997) as well as predators, would groom at significantly higher rates than bighorn sheep of this study, even if they were maintained in an ectoparasite-free environment.
When grooming did occur in bighorn at Red Rock, the rate of oral grooming was inversely associated with body mass. Juveniles (X = 15 kg) groomed the most and adult males, with the greatest body mass (X = 70–85 kg), groomed the least. Adult females and yearlings (X = 41–52 kg) groomed at intermediate rates. These results are consistent with the intraspecific body size prediction that grooming rate decreases as animals grow older and larger. Intraspecific comparisons between young animals and adult counterparts have been made for impala, bison, elk, domestic goats, and domestic sheep (Hart and Pryor, 2004; Mooring and Hart, 1992, 1997a,b; Mooring and Samuel, 1998a,b). In all cases, small neonates and young groomed at a much higher rate than adults. For example, impala lambs oral groomed at two to three times the rate of their mothers (Mooring and Hart, 1992, 1997a,b), while the prevalence of moderate to severe tick infestation was two to five times greater on adult impala compared with lambs (Gallivan et al., 1995). In addition, a comparative study of 25 bovid species observed in a tick-free zoological park (Mooring et al., 2002) supported the prediction that juveniles would groom more frequently than adults.
In prior studies of the body size principle in impala and goats (Hart and Pryor, 2004; Mooring and Hart, 1997a), matched-pair comparisons indicated that juveniles ceased to groom at significantly higher rates than adults when their mass increased to approximately 30% of adults. In this study, the difference in body mass between bighorn juveniles (15 kg) and adult females (60 kg from capture operations; Mooring et al., 2003) when grooming rate was no longer significant was about 20%.
Although body mass (category) was the only significant independent variable in our analyses, it does not fully explain the magnitude of the grooming rate differences observed. Because the body size prediction is based on the allometry of surface-to-mass ratio, grooming rate is predicted to scale to mass2/3 (Clutton-Brock and Harvey, 1983; Schmidt-Nielsen, 1984). For the range of masses observed in this study (15–85 kg), the allometric relationship alone would predict that the grooming rate of the largest adult males be approximately half (56%) that of juveniles or about 14 bouts per l2-h day. In fact, the biggest males in this study groomed no more than one bout per 12 h, or only 4% the rate of juveniles. Thus, other factors may be needed to explain the dramatic difference in grooming rate between juveniles and adult males. For example, another prediction of the programmed grooming model is that breeding males will groom less than females due to the higher cost of relaxed vigilance for breeding males of polygynous mating systems (the vigilance principle). Sexual dimorphism in grooming rate has been shown in previous studies of ungulates (Hart et al., 1992; Mooring and Hart, 1995b; Mooring and Samuel, 1998a; Mooring et al., 1996b, 1998, 2002).
To our knowledge, this is the first reported case of ectoparasite-defense behavior persisting in a wild population in the absence of the ectoparasites. That bighorns at Red Rock exhibited programmed, tick-defense grooming in the absence of tick exposure raises the question of current utility versus evolutionary inertia in their grooming system. Byers (1997) has elegantly shown that much of the behavior of pronghorn antelope (Antilocapra americana) can be understood as adaptations to ancestral predation risk (the ghost of predators past), even though those predators have long been extinct. The San Andres population of bighorn, which contributed 16 of the 21 founder animals to the RRWA population, is not presently exposed to D. hunteri or other ticks (Crosbie et al., 1997) and probably was not exposed to ticks in the recent past. (Sonora, Mexico, from which the other five founders originated, is within the distribution of D. hunteri.) Thus, the majority of the founding stock have not been exposed to ticks for an unknown period of time, perhaps hundreds or thousands of years. We conclude that the relationship between body size and grooming patterns exhibited by bighorn sheep at Red Rock do not reflect current defense against ticks but were adaptations to the threat of ticks in their ancestral environment. According to this perspective, the programmed grooming system of Red Rock desert bighorn sheep is a relict of the ghost of parasites past.
|
ACKNOWLEDGEMENTS |
|---|
We thank the NMDGF for permission to study the bighorn sheep of RRWA as well as the use of the ATV during summer 2000. Eric Rominger and Heather Whitlaw provided logistical and administrative support, caretaker Don Graves provided valued assistance in the field, and contractors Alton Ford and Steve Harvill freely shared their knowledge with us. We are especially grateful to Betsy and Louie Cabral, who opened their ranch to us, allowing us to set up tents and live in their caboose "guest house" during two summers. Mitch Watnik and Neil Willets gave valuable statistical advice. Suggestions by Thelma Rowell, Kay Holekamp, and Philipp Heeb improved a previous version of the manuscript. Funding for this study was provided by the Point Loma Nazarene University (PLNU) Department of Biology, the Research Associates, a PLNU Alumni Association award, a PLNU Wesleyan Center Scholarship and Fellowship, a PLNU Research and Special Projects grant, and by the NMDGF. The Welseyan Center for 21st Century Studies provided office space away from distractions for the preparation of the manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Allan SA, 2001. Ticks (Class Arachnida: Order Acarina). In: Parasitic diseases of wild mammals, 2nd ed (Samuel WM, Pybus MJ, Kocan AA, eds). Ames: Iowa State University Press; 72–106.
Altmann J, 1974. Observational study of behavior: sampling methods. Behaviour 49:227–267.[Medline]
Burt WH, Grossenheider RP, 1980. A field guide to the mammals of North America north of Mexico, 3rd ed. Boston: Houghton Mifflin.
Byers JA, 1997. American pronghorn: social adaptations and the ghost of predators past. Chicago: University of Chicago Press.
Clutton-Brock TH, Harvey PH, 1983. The functional significance of variation in body size among mammals. Spec Publ Am Soc Mammal 7:632–663.
Colbern DL, Gispen WH (eds), 1988. Neural mechanisms and biological significance of grooming behavior. Ann N Y Acad Sci 525:1–438.[Web of Science][Medline]
Cords M, 1995. Predator vigilance cost of allogrooming in wild blue monkeys. Behaviour 132:559–569.
Crosbie PR, Goff WL, Stiller D, Jessup DA, Boyce WM, 1997. The distribution of Dermacentor hunteri and Anaplasma sp. in desert bighorn sheep (Ovis canadensis). J Parasitol 83:31–37.[CrossRef][Medline]
Fentress JC, 1988. Expressive contexts, fine structure, and central mediation of rodent grooming. Ann N Y Acad Sci 525:18–26.[Web of Science][Medline]
Gallivan GJ, Culverwell J, Girdwood R, Surgeoner GA, 1995. Ixodid ticks of impala (Aepyceros melampus) in Swaziland: effect of age class, sex, body condition and management. S Afr J Zool 130:178–l86.
Geist V, 1971. Mountain sheep: a study in behavior and evolution. Chicago: University of Chicago Press.
Greer JM, Capecchi MR, 2002. Hoxb8 is required for normal grooming behavior in mice. Neuron 33:23–34.[CrossRef][Web of Science][Medline]
Haltenorth T, Diller H, 1980. A field guide to the mammals of Africa including Madagascar. London: Collins.
Hansen CG, Deming OV, 1980. Growth and development. In: The desert bighorn: its life history, ecology and management (Monson G, Sumner L, eds). Tucson: University of Arizona Press; 152–171.
Hart BL, 1990. Behavioral adaptations to pathogens and parasites: five strategies. Neurosci Biobehav Rev 14:273–294.[CrossRef][Web of Science][Medline]
Hart BL, 1997. Effect of hormones on behavioral defenses against parasites. In: Parasites and pathogens: effects on host hormones and behavior (Beckage N, ed) New York: Chapman & Hall; 210–230.
Hart BL, Hart LA, Mooring MS, Olubayo R, 1992. Biological basis of grooming behaviour in antelope: the body size, vigilance and habitat principles. Anim Behav 44:615–631.[CrossRef]
Hart BL, Pryor PA, 2004. Developmental and hair-coat determinants of grooming behaviour in goats and sheep. Anim Behav 67:11–19.[CrossRef]
Kaiser MN, Sutherst RW, Bourne AS, 1991. Tick (Acarina:Ixodidae) infestations on zebu cattle in northern Uganda. Bull Entomol Res 81:257–262.
Koch HG, 1981. Suitability of birds and mammals as hosts for immature stages of the lone star tick, Amblyomma americanum (Acari: Ixodidae). J Med Entomol 18:93–98.[Web of Science][Medline]
Koch HG, 1988. Suitability of white-tailed deer, cattle and goats as hosts for the lone star tick, Amblyomma americanum (Acari: Ixodidae). J Kans Entomol Soc 61:251–257.
Krausman PR, Sandoval AV, Etchberger RC, 1999. Natural history of desert bighorn sheep. In: Mountain sheep of North America (Valdez R, Krausman PR, eds). Tucson: University of Arizona Press; 139–191.
Krausman PR, Shackleton DM, 2000. Bighorn sheep. In: Ecology and management of large mammals of North America (Demarais S, Krausman PR, eds). Upper Saddle River: Prentice-Hall; 517–544.
Kuter M, Nachtsheim C, Neter J, Kutner MH, Nachtsheim CJ, Li W, 2004. Applied linear statistical models, 5th ed. New York: McGraw-Hill.
Little DA, 1963. The effect of cattle tick infestation on the growth rate of cattle. Aust Vet J 39:6–10.
MacDonald D (ed), 1984. All the world's animals: hoofed mammals. New York: Torstar Books.
Maestripieri D, 1993. Vigilance costs of allogrooming in macaque mothers. Am Nat 141:744–753.[CrossRef]
McKenzie AA, 1990. The ruminant dental grooming apparatus. Zool J Linn Soc 99:117–128.
Milner JM, Elston DA, Albon SD, 1999. Estimating the contributions of population density and climatic fluctuations to interannual variation in survival of Soay sheep. J Anim Ecol 68:1235–1247.[CrossRef]
Monson G, 1980. Distribution and abundance. In: The desert bighorn: its life history, ecology and management (Monson G, Sumner L, eds). Tucson: University of Arizona Press; 40–50.
Moore J, 2002. Parasites and the behavior of animals. New York: Oxford University Press.
Mooring MS, 1995. Effect of tick challenge on grooming rate by impala. Anim Behav 50:377–392.[CrossRef]
Mooring MS, Benjamin JE, Harte CR, Herzog NB, 2000. Testing the interspecific body size principle in ungulates: the smaller they come, the harder they groom. Anim Behav 60:35–45.[CrossRef][Web of Science][Medline]
Mooring MS, Blumstein DT, Stoner CJ, 2004a. The evolution of parasite-defence grooming in ungulates. Biol J Linn Soc 81:17–37.[CrossRef]
Mooring MS, Fitzpatrick TA, Benjamin, JE, Fraser IC, Nishihira TT, Reisig DD, Rominger EM, 2003. Sexual segregation in desert bighorn sheep (Ovis canadensis mexicana). Behaviour 140:183–207.[CrossRef]
Mooring MS, Fitzpatrick TA, Nishihira TT, Reisig DD, 2004b. Vigilance, predation risk, and the allee effect in desert bighorn sheep. J Wildl Manag 68:519–532.[CrossRef]
Mooring MS, Gavazzi AJ, Hart BL, 1998. Effects of castration on grooming in goats. Physiol Behav 64:707–713.[CrossRef][Medline]
Mooring MS, Hart BL, 1992. Reciprocal allogrooming in dam-reared and hand-reared impala fawns. Ethology 90:37–51.
Mooring MS, Hart BL, 1995a. Costs of allogrooming in impala: distraction from vigilance. Anim Behav 49:1414–1416.[CrossRef]
Mooring MS, Hart BL, 1995b. Differential grooming rate and tick load of territorial male and female impala, Aepyceros melampus. Behav Ecol 6:94–101.
Mooring MS, Hart BL, 1997a. Self grooming in impala mothers and lambs: testing the body size and tick challenge principles. Anim Behav 53:925–934.[CrossRef]
Mooring MS, Hart BL, 1997b. Reciprocal allogrooming in wild impala lambs. Ethology 103:665–680.[Web of Science]
Mooring MS, McKenzie AA, Hart BL, 1996a. Grooming in impala: role of oral grooming in removal of ticks and effects of ticks in increasing grooming rate. Physiol Behav 59:965–971.[CrossRef][Medline]
Mooring MS, McKenzie AA, Hart BL, 1996b. Role of sex and breeding status in grooming and total tick load of impala. Behav Ecol Sociobiol 39:259–266.[CrossRef][Web of Science]
Mooring MS, Reisig DD, Niemeyer JM, Osborne ER, 2002. Sexually and developmentally dimorphic grooming: a comparative survey of the Ungulata. Ethology 108:1–24.[CrossRef][Web of Science]
Mooring MS, Samuel WM, 1998a. Tick defense strategies in bison: the role of grooming and hair coat. Behaviour 135:693–718.
Mooring MS, Samuel WM, 1998b. Tick-removal grooming by elk (Cervus elaphus): testing principles from the programmed-grooming hypothesis. Can J Zool 76:740–750.[CrossRef]
Mooring MS, Samuel WM, 1998c. The biological basis of grooming in moose (Alces alces): programmed versus stimulus-driven grooming. Anim Behav 56:1561–1570.[CrossRef][Web of Science][Medline]
Mooring MS, Samuel WM, 1999. Premature loss of winter hair in free-ranging moose (Alces alces) infested with winter ticks (Dermacentor albipictus) is correlated with grooming rate. Can J Zool 77:148–156.[CrossRef]
Nelson WA, Keirans JE, Bell JF, Clifford CM, 1975. Host-ectoparasite relationships. J Med Entomol 12:143–166.[Web of Science][Medline]
Norval RAI, Sutherst RW, Kurki J, Gibson JD, Kerr JD, 1988. The effect of the brown ear-tick Rhipicephalus appendiculatus on the growth of Sanga and European breed cattle. Vet Parasitol 30:149–164.[CrossRef][Web of Science][Medline]
Nowak RM, 1999. Walker's mammals of the world, 6th ed. Baltimore: Johns Hopkins University Press.
Olubayo RO, Jono J, Orinda G, Groothenhuis JG, Hart BL, 1993. Comparative differences in densities of adult ticks as a function of body size on some East African antelopes. Afr J Ecol 31:26–34.
Riek RF, 1962. Studies on the reactions of animals to infestation with ticks. Aust J Agric Res 13:532–550.[CrossRef]
Ritter RC, Epstein AN, 1974. Saliva lost by grooming: a major item in the rat's water economy. Behav Biol 11:581–585.[CrossRef][Web of Science][Medline]
Roth L, Rosenblatt JS, 1967. Changes in self-licking during pregnancy in the rat. J Comp Physiol Psychol 63:397–400.[CrossRef][Web of Science][Medline]
Schmidt-Nielsen K, 1984. Scaling: why is animal size so important? New York: Cambridge University Press.
Shackleton DM, Shank CC, Wikeem BM, 1999. Natural history of Rocky Mountain and California bighorn sheep. In: Mountain sheep of North America (Valdez R, Krausman PR, eds). Tucson: University of Arizona Press; 78–138.
Simmons NM, 1969. Heat stress and bighorn behavior in the Cabeza Prieta Game Range, Arizona. Desert Bighorn Counc Trans 13:55–63.
Simmons NM, 1980. Behavior. In: The desert bighorn: its life history, ecology and management (Monson G, Sumner L, eds). Tucson: University of Arizona Press; 124–144.
Spruijt BM, Van Hooff JARAM, Gispen WH, 1992. Ethology and neurobiology of grooming behavior. Physiol Rev 72:825–852.
SPSS, 2001. Advanced models 11.0. Chicago: SPSS Inc.
Sutherst RW, Maywald GF, Kerr JD, Stegeman DA, 1983. The effect of cattle tick (Boophilus microplus) on the growth of Bos indicus x B. taurus steers. Aust J Agric Res 34:317–327.[CrossRef]
Wikel SK, 1984. Immunomodulation of host responses to ectoparasite infestation. An overview. Vet Parasitol 14:321–339.[CrossRef][Web of Science][Medline]
Willadsen P, 1980. Immunity to ticks. Adv Parasitol 18:293–313.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Hawlena, D. Bashary, Z. Abramsky, I. S. Khokhlova, and B. R. Krasnov Programmed versus stimulus-driven antiparasitic grooming in a desert rodent Behav. Ecol., September 1, 2008; 19(5): 929 - 935. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Hillegass, J. M. Waterman, and J. D. Roth The influence of sex and sociality on parasite loads in an African ground squirrel Behav. Ecol., September 1, 2008; 19(5): 1006 - 1011. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||