Heterogeneity in host compatibility is one of the main hypotheses proposed to explain uneven resistance to parasites and uneven parasite load between hosts. It suggests that differences between hosts modulate their predispositions as suitable environments for their potential parasites. Interesting studies of antiparasitic behavior have reported the existence of behavioral traits that are capable of removing foreign particles and of reducing the success of parasitic infections. These traits favor host neatness, although little is known about the heterogeneity of neatness. We used a standardized pseudoinfection with pseudoectoparasites (PEPs) to test the effects of sex, age, and season on the loss of PEPs by hosts as a means of exploring the factors determining neatness in the Iberian ibex Capra pyrenaica. Behavioral observations were also performed to analyze investment in antiparasitic behavior in terms of sex, age, and season. The life span of PEPs peaked in the period December–January, decreased with host age, and was longer in females than in males. Investment in antiparasitic behavior is also associated with both sex and age and season but in a different pattern with interactions between such factors. Our results disagree with the hypothesis that small-bodied animals should be less compatible to carry contact-transmitted particles, such as ectoparasites, in comparison with larger animals. This preexisting hypothesis is thus an inadequate way of predicting host neatness. Consequently, our experiment underlines the importance that nonimmunological traits play in determining heterogeneity in host compatibility to contact-transmitted foreign bodies and helps improve understanding of neatness and of host–parasite systems. Key words: antiparasitic behavior, body size principle, Capra pyrenaica, ectoparasites, grooming, host compatibility, neatness, ungulates. [Behav Ecol 22:1070–1078 (2011)]
to ectoparasites through a greater body surface to mass ratio (Hart et al. 1992). The body-size principle, resulting from the ‘‘programmed grooming model,’’ suggests that small- bodied animals should groom at a higher rate and conse- quently have fewer ectoparasites in comparison with larger animals (Hart et al. 1992). Thus, according to this principle, juveniles would groom more than adults (Mooring and Hart
1997; Mooring, Hart, et al. 2006) and, in dimorphic species, females would groom more than males (Mooring et al.
2002). To date, this principle has been mainly tested in un-
gulates (Mooring et al. 2000, 2002, 2004; Mooring, Hart, et al. 2006; Mooring, Patton, et al. 2006). In terms of non- immunological defenses against parasites and based on this
prevailing body-size principle, adult males are often assumed
to be more prone than other host classes to colonization by contact-transmitted foreign entities (such as ectoparasites).
Thus, adult males may have higher densities of ectoparasites
than females and juvenile (Schalk and Forbes 1997; Moore and Wilson 2002), in part due to their supposed lower in- vestment in antiparasitic behavior (Hart et al. 1992). This
expected nonimmunological predisposition to parasite colo-
nization and the ‘‘immunocompetence handicap’’ hypothe- sis (Folstad and Karter 1992) might even act additively to produce uneven distribution of parasites. Testosterone stim-
ulates the development of characteristics used in sexual se-
lection and at the same time reduces immunocompetence (Folstad and Karter 1992). Androgens would mediate the suppression of antiparasitic behavior as well (Mooring
et al. 1998; Mooring, Patton, et al. 2006).
However, recent studies have highlighted the fact that the body-size principle might not be as generalized and prepon-
derant as expected and place doubt on the ability of the
body-size principle to predict host neatness. Grooming, for example, has been observed to take place less often in juve- niles than in adults in the desert rodent Meriones crassus
(Hawlena, Bashary, et al. 2007; Hawlena et al. 2008), Cape
ground squirrels Xerus inauris showed no sex differences in terms of autogrooming (Hillegass et al. 2008), and the preliminary results of a test procedure using pseudoectopar-
asites (PEPs) did not match predictions based on the body-
size principle (Sarasa et al. 2009). Thus, these studies indicate that there is a need to study variability in neatness
further and to analyze its consistency with predictions
derived from the body-size principle.
In this study, we used PEPs (Sarasa et al. 2009) to test ex- perimentally the role of key factors of individual heterogene-
ity in patterns of uneven neatness in a sexually dimorphic
ungulate, the Iberian ibex Capra pyrenaica (Pe´rez et al.
2002). Observations of behavior were also performed to ana- lyze investment in antiparasitic behavior. According to the
body-size principle, we expected to find greater investment
in antiparasitic behavior in small ibexes than in large ones
(Mooring and Hart 1997; Mooring et al. 2002; Mooring Hart, et al. 2006). Thus, we expected neatness (expressed as the
investment in antiparasitic behavior and the ability to remove PEPs) to decrease with age (prediction 1) and males to be less neat than females (prediction 2).
In addition, host–pathogen interactions are usually highly
status dependent. More specifically, the mating season might signal a period of change in the levels of physiological modulators of grooming behavior (Mooring, Patton, et al.
2006) and in investment in antiparasitic behavior (Mooring
et al. 1996a). These changes relate to the link between host behavior and opportunism in parasite life cycles (Tinsley
1990), and in the case of the ecology of the Iberian ibex, autumn rutting is generally considered to be the starting
point and hence one of the main determining factors of the seasonal outbreaks of sarcoptic mange caused by the
contact-transmitted mite Sarcoptes scabiei (Pe´ rez et al. 1997; Leo´ n-Vizca´ıno et al. 1999). This burrowing mite consumes living cells and tissue fluid from the skin of its host (Pence and Ueckermann 2002). Sarcoptic mange is a major deter- mining factor of Iberian ibex ecology, population dynamics, and management practices (Pe´ rez et al. 2002, 2011; Sarasa et al. 2010, 2011). Thus, during the mating season, we ex- pected to find a season-dependent reduction in antiparasitic behavior coinciding with a peak in the life span of colonizing ectoparasites and of PEPs, as well as greater neatness during the rest of the year (prediction 3).
MATERIALS AND METHODS Study site
The experiment was performed in a large enclosure (35 ha) containing a stock reservoir population of Iberian ibex (Espa- cio Natural de Sierra Nevada [ENSN], lat 3709#N long 3031#E, Granada, southern Spain). This enclosure was built in the
1990s to protect a pool of ibexes trapped in the surrounding
Sierra Nevada Natural Space from mange. The enclosure that contains the stock reservoir population only prevents expo-
sure of the mange-free ibexes inside to potentially mangy
ibexes on the outside. As a result, the stock reservoir popula- tion is free of mange but not free of other parasites. During the study period, 41 males and 46 females that had been
marked as kids with numbered ear tags, ranged freely within
this enclosure. Ibexes had previously developed tolerance to the observer (Sarasa et al. 2009) to avoid the behavioral alter- ations associated with the presence of a human observer
(Crofoot et al. 2010). The natural food supply available in
the enclosure was complemented on a daily basis with addi- tional forage provided in several mangers and their surround-
ing areas to prevent the monopolization of feeding sites and
food by dominant individuals (Appleby 1980).
To avoid interobserver variability, all the fieldwork was car- ried out by the first author. The experiment was based on a pseudoinfection protocol using PEPs whose characteristics have been previously described (Sarasa et al. 2009). PEPs are waxed wooden triangular markers that mimic innocuously contact-transmitted parasitic infections in host species (Sarasa et al. 2009). PEPs mimic several mechanical features of ectoparasites and can be transmitted by contact to hosts. PEPs are under the effects of the microhabitat conditions of the body surface of their host, just as real contact-transmitted parasites do (Sarasa et al. 2009). Nevertheless, the host–PEP interaction is only an imitation of real host–parasite interac- tions because PEPs are unaffected by host immunity, are innocuous to the host, and do not exhibit microhabitat preferences (e.g., parasite mobility and preferences [Murray
1990; Crompton 1997; Christe et al. 2007; Khokhlova et al.
2011]) (Sarasa et al. 2009). Consequently, PEPs enable the importance of the nonimmunological features that
influence the compatibility between hosts and contact-
transmitted parasites to be investigated, while controlling for parasite biology. Whatever the preestablished parasitic community of the host, real contact-transmitted parasites
attempting to colonize a new host have to overcome to the
microhabitat conditions of the host body surface prior to successfully establishing themselves. Likewise, PEPs are un- der the effects of the microhabitat conditions of the host
body surface that lead to PEPs loose (Sarasa et al. 2009),
and we characterized the outcome of such confrontations as an indicator of host compatibility and neatness. PEPs were
prepared just prior to animal manipulation (before dawn) to minimize handling time and to optimize the features of the PEPs during pseudoinfection. Food was used to attract all the ibexes into a small bottlenecked space that is habitually used for managing animals in the large enclosure. In the selection of the pseudoinfected ibexes, we ensured that all sex- and age-classes were represented (Supplementary Table S1). When handled, animals were blindfolded, physically re- strained, and each pseudoinfected with 44 PEPs distributed over the whole skin surface of the host according to a stan- dardized protocol (each host had different color PEPs). An- imals were then released into the large enclosure (Day
0 post-pseudoinfection ¼ D0ppi). The dynamics of the PEPs on each individual were monitored daily for 3 weeks using
10 3 40 Bushnell binoculars and a Canon EOS 400D camera with a 70–300 mm lens. The short-range observation distance
and extensive photographic data were key factors in the fine
monitoring of the dynamics of the PEPs.
Rutting in the Iberian ibex occurs in autumn between mid-November and mid-December (Fandos 1991). Contact-
transmitted parasites are supposed to take advantage of the
mating season to propagate within populations and the peak in occurrence of major parasites such as S. scabiei takes place between January and March (Pe´rez et al. 1997;
Leo´ n-Vizca´ıno et al. 1999; Granados et al. 2007). To examine
inter-seasonal variability of host neatness before, during, and after these key periods in the ecology of the Iberian ibex, the whole experimental procedure was repeated 5 times (I: end of
August–early September 2007; II: mid October–early
November 2007; III: end of November–mid December 2007; IV: mid-January–early February 2008, and V: March 2008).
PEPs were found to be a suitable tool for this purpose because
their characteristics (in particular their adherence) were sta- ble between the considered periods (Sarasa et al. 2009). Dur- ing the experiment, we also recorded the peak of the rutting
season, that is, the period in which tending and coursing be-
havior in males is especially intense and frequent and in which females do not systematically avoid and are more receptive to courting males. Different individual hosts were considered for
each season to avoid inter-seasonal autocorrelation. A total of
57 individuals were pseudoinfected and monitored through- out the experiment (Supplementary Table S1).
To examine investment in antiparasitic behavior related to age, sex, and season, we also scan-sampled behavior
(Altmann 1974) in the year after the pseudoinfections using the same calendar of 5 sampling periods. This protocol was based on the relative phenological stability of the biology of the Iberian ibex, well illustrated by the fact that the peak in rutting occurred as normal in mid-November during the years of pseudoinfection and behavioral observations, as de- scribed by previous studies of the life cycle of the species (Fandos 1991). For each period, 9–13 samplings were performed (70% in morning and 30% in afternoon), during which the behavior of all the observed animals (individually marked with ear tags) was scan sampled to estimate the rel- ative importance (RI) of antiparasitic behavior in their activity budgets and its seasonality. We distinguished ‘‘intra- specific cleaning behavior’’ (ISCB) from ‘‘exo-scratching be- havior (ESB).’’ ISCB is the maintenance of the coat using teeth, horns, or any other part of the body and includes self-grooming, allogrooming, and hindleg scratching de- scribed in other studies (Mooring et al. 2002, 2004; Mooring, Patton, et al. 2006). ESB is the voluntary rubbing of parts of the body on environmental components such as trees, fen- ces, or rocks. Note that ESB is not normally considered in major studies of antiparasitic behavior (Mooring et al. 2000,
2002, 2004), maybe due to the difficulties in sampling such behavior in free-ranging individuals and in deficient environ- ments such as zoological parks. Nevertheless, alternative strategies to ISCB such as ESB (including wallowing) or cleaning symbiosis could be consistently associated with par- asite life cycles (McMillan et al. 2000) and might even be highly effective in removing parasites (Bezuidenhout and Stutterheim 1980). During our observations, we also re- corded rutting behavior (tending and coursing behavior in males). Samplings were performed at least 6 h apart and almost always 24 h apart in order to reduce/avoid temporal autocorrelation.
In our behavioral data set, we analyzed the factors associated with the mean investment in antiparasitic behavior per in- dividual in order to focus the study on the importance of explanatory variables rather than on the rarity of occur- rence of antiparasitic behavior. Thus, the mean investment in total and in categories of antiparasitic behavior was esti- mated for each individual in each replication period. As suggested by Verbeke and Molenberghs (2000), we per- formed an exploratory data analysis of our behavioral data set. We tested for temporal autocorrelation (a relationship