Leadership Behavior in Relation to Dominance and
Reproductive Status in Gray Wolves, Canis lupus
Discussion
Several of the aspects of wolf behavior measured in this study seemed, a priori, to be linked to individual leadership potential. These metrics were variously successful. Nonfrontal leadership was rarely recorded and so contributed little to our understanding of individual roles. Initiation of pack behaviors was also infrequently observed, but our observations were consistent with the larger sample of individuals leading packs during travel.
Scent-marking, while having no direct bearing on leadership per se, provided the clearest indicator of dominant, breeding status, and our direct observations confirmed earlier work (Peters and Mech 1975; Haber 1977; Rothman and Mech 1979; Asa et al. 1990; Asa and Mech 1995; Mech 1995, 1999). All scent-marking was done by wolves of high status, and even a single observation of double scent-marking was sufficient to identify a wolf pair as dominant breeders. In this study, nonbreeders never marked, nor did most subordinate breeders; the only exception involved a female ascending to dominant status, who was observed scratching in the absence of the dominant pair. In another study, a female also ascending to dominant breeder status was seen flexed-leg urinating (Mech 1995).
The differences in scent-marking frequency between dominant males and females may be due to the apparent function of scent-marking in forming and maintaining a strong pair bond just prior to estrus. Dominant males marked at a high rate in both early winter and late winter, but dominant females seemed to mark less in late winter, after breeding ceased. Perhaps female scent marks function partly to indicate approaching receptivity. And the dominant male may maintain a higher level of territorial vigilance than the dominant breeding female, commensurate with his more extensive movements during the pup-rearing season (Mech 1999). We interpret differences in the scent-marking behavior of dominant breeding males and females as reflecting role differences relating to reproduction, not as differences in leadership.
The relative frequency that different individuals were recorded leading the pack during travel corresponded well to total time spent in the lead. Only the former data lent themselves to statistical analysis, which confirms the important role of breeding wolves, especially dominant breeders, in leading packs and prompting pack activities. We observed the highest frequency of leading by nonbreeding wolves in the Rose Creek pack, with 14-23 members. It is not clear whether this is a general pattern for large packs or simply an idiosyncrasy of the Rose Creek pack. In this study, variation in wolf leadership roles among packs was significant, consistent with the individual variation that is evident in virtually all studies of wolf behavior in captivity (e.g., Rabb et al. 1967; Zimen 1981; Fox 1971). Further work is necessary to determine whether large pack size reduces the almost exclusive tendency of dominant breeding wolves to lead smaller packs.
The strongest predictor of leadership in this study was high social rank, consistent with this general tendency previously identified for species with a strong hierarchical structure (Holekamp et al. 2000). Our findings should help resolve a continuing misperception, evident in the literature, that wolf packs are led by a single alpha wolf, the dominant male (Holekamp et al. 2000). Rather, a high-ranking male and female typically form the primary breeding pair and provide most leadership. In the case of gray wolves, contrary to the usual pattern for mammals, parental investment by both males and females is high, dominance within a pack is shared by the sexes, and pack-leadership responsibilities are shared by the sexes (Mech 2000). A similar pattern is seen in many species of social primates (Boinski 2000).
The results of this study suggest that individual variation in leadership patterns among wolf packs may be high and may vary with the season. We found division of leadership to be about equal between dominant males and females, at least in winter, in contrast to the pup-rearing season, when the male concentrates on travel and prey capture and the female is focused on rearing pups (Mech 1999). Individual variation may explain the occasional predominance of one wolf in leading the pack. Individual differences in age, experience, previous alliances, and temperament might influence ability to lead, but these influences have been poorly documented, even for social primates, which have been studied extensively (Boinski 2000).
Although dominant breeding wolves provided most leadership, we found that subordinate wolves, both breeders and nonbreeders, also provided leadership during travel. Dominant breeding wolves might share leadership in order to take advantage of pooled experience in a territory, although in this study a newly arrived dominant male showed the greatest tendency to lead. Shared leadership might reduce the energy expenditure of dominant individuals, such as in deep snow. Shallow snow prevailed in this study and we were unable to evaluate whether switching of leadership roles might increase with the cost of travel, which seems plausible. The ecological cost of travel for a 23-kg wolf was calculated to be 16% of daily energy expenditure (Steudel 2000), and this proportion would be higher for gray wolves, like those in Yellowstone National Park, that weigh twice this amount. Finally, shared leadership might spread the risks associated with travel, such as encountering fast water or thin ice. In two of the three packs in this study, the tendency of dominant females to lead declined in late winter, after the annual breeding season, when they are usually pregnant.
We caution against attaching undue significance to our observation that subordinate wolves sometimes led packs. Even though the subtle social interactions involved in travel coordination are difficult to observe, we found that dominant breeding wolves often made decisions that affected the pack's direction even when not in the lead. It is common in social species for subordinate individuals to closely monitor dominant leaders, so it is necessary to distinguish decision-makers, the true leaders, from initiators, who merely suggest a direction (Byrne 2000).
While two of the three packs in this study included multiple breeding females, relatives of the dominant breeding female, none of the packs had multiple male breeders that we knew of. This may be a general pattern that allows genetic relatives to participate in reproduction or ascend to dominant status. There is more uncertainty in determining genetic ties involving males than females, which perhaps explains the virtual absence of published records of multiple male breeders in a wolf pack. Persistence of subordinate female breeders in a pack appears to be at the discretion of the dominant breeding female. When subordinate females are allowed to remain within packs, their behavior (or, in this study, their role in pack leadership) appears to be tightly constrained by the dominant female breeder.
We have not used "alpha" because the value of this label has been questioned. Mech (1999) argued that the dominance hierarchy on which the structure of typical wolf packs containing a single breeding pair is based merely reflects parent-offspring dominance. However, in packs with multiple breeders, Mech (1999) allowed that there might be "intense rivalries such as those Haber (1977) reported" and that "the one use we may still want to reserve for 'alpha' is in the relatively few large packs comprised of multiple litters". Similarly, we found consistent differences in the roles played by dominant and subordinate female breeders. Dominant threats from the breeding pair usually prevent breeding by auxiliary wolves (Mech 1970; Peterson 1977), and indeed, the evolution of dominance behavior within wolf packs with multiple breeding females may be a result of selection pressure to adjust pack reproduction to food availability (Zimen 1976). Because wolves are usually food-limited, dominance relationships channel reproductive efforts into a single pair. As short-hand for "dominant breeder", "alpha" does seem to be appropriate for packs of multiple breeders. However, it falsely implies a hierarchical system in which each wolf assumes a place in a linear "pecking order" (Mech 1999).
We observed incipient leadership behavior in a subordinate male (nonfrontal leadership) just prior to dispersal and, with better documentation, among subordinate breeding females in the form of low-level scent-marking (when dominant wolves were absent), leading the pack, and initiating pack behaviors. A subordinate breeding female, related to the dominant female and her potential successor, may exhibit limited leadership and thus provide in the future a more successful transition in pack leadership that enhances matrilineal genetic survival.
