Functional Interactions

The picture is greatly complicated by the fact that interactions are really between communities of plants and communities of animals. Regionally and locally, wild ruminants form guilds, communities of organisms exploiting a similar resource base in a similar way. Within a guild, functional interactions may range from competitive to symbiotic, facultative to almost obligatory and reciprocal to unsymmetrical. Such interactions are not strictly characteristic of species pairs but also depend on their relative densities and the landscape.

Wild ruminants have radiated in marvelous variety and now occupy most of the world's biomes. At regional and local scales, there is ample evidence of complementarity with each species using foods, habitats and space in temporally distinctive ways. Much attention has been paid to summarizing and analyzing the multidimensionality of this ecological separation (Lamprey 1963, Ferrar & Walker 1974, Hudson 1976a,b, Greenacre & Vrba 1984).

The problem lies in interpreting these static descriptions since they may reflect either current interactive processes, historical events such as epidemics or the 'ghosts of competition past' (Connell 1980). These 'ghosts' may even involve habitats and competitors that have long since disappeared. On the one hand, such ecological specialization could be considered a mechanism which allows greater species packing, higher animal densities and higher overall grazing systems productivity. On the other, this apparent complementarity may reflect intense functional interactions which may make these systems fragile to human manipulation.

This ambiguity has only recently been appreciated and studies have finally progressed beyond static descriptions of patterns to the analysis of dynamic interactions. Perturbations are essential to reveal underling processes. Only by comparing pre-and post interactive niches and determining responses in population densities as well as resource use can we learn much about forces operating within these communities. Progress is frustrated by the logistic and sometimes ethical problems of conducting field experimentation. Consequently, most manipulative experiments have been conducted on livestock-wildlife interactions. Work on wildlife communities has depended on natural historical experiments such as disease or over-hunting (Caughley & Sinclair 1994). One caution is that given the inherently dynamic interaction between plants and animals, apparent populations responses of one species to the fortunes of another may be misleading unless unraveled from a very long time-series. In practice, this usually means that conclusions must draw from corroborative evidence based on both numerical and ecological/behavioural responses.

Studies on functional interactions within the herbivore guild should address four possible interrelationships: namely, behavioural interference, mediation by predators/parasites/pathogens, resource competition, and grazing facilitation (Fig. 5).

Fig. 5. Functional interactions within herbivore guilds.

Behavioural interference

Interference competition arises from direct behavioural interactions. Notes of aggressive inter-specific encounters are widespread in the literature. The problem is to interpret their ecological importance. The most practical approach is to evaluate spacing behaviour (de Boer & Prins 1990). Because spatial separation of animals can (and largely do) arise from landscape patterns, analysis of spatial relationships usually are based either on fine-scale temporal analysis or, less precisely, by examining residuals after landscape variation has been statistically removed. Beyond displacement, interference may arise from reductions in feeding rates as demonstrated in red deer hinds grazing near dominant individuals (Thouless 1990).

Mediation by predators, parasites and pathogens

Considerable attention has focused on the role of predators, parasites and pathogens as mediators of functional interactions (Holt 1977, Price et al. 1986).

An abundant herbivore may support relatively high predator densities which it plays against a rarer species less able to withstand predation pressure. The classic example is territorial predators of wildebeest or other abundant migratory species which switch to resident species as migrants continue their seasonal round (Fryxell et al. 1988). In western North America, abundance and distribution of moose seem to affect wolf predation on the much rarer woodland/mountain caribou (Seip 1992).

Predation may influence the structure of ungulate guilds in ways other than causing mortality. For example, the spatial relationship between zebra and wildebeest seems more related to predation than grazing facilitation (Sinclair 1985). Although zebra may appear to lead the succession of migratory herbivores in the Serengeti ecosystem, they may be at the leading edge of the wildebeest migration to avoid predation. Interspecific associations of territorial male African antelope also form in response to predation (Gosling 1979). In contrast, caribou seem to avoid areas occupied by moose because of greater encounter rates with wolves.

Most work on mediation deals with parasites, particularly the meningeal worm of eastern North America (Whitlaw & Lankester 1994). The meningeal worm (Parelaphostrongylus tenuis) is a benign parasite of white-tailed deer which causes heavy mortality in moose and caribou. It once was implicated in die-offs of moose and failure of introductions of caribou. The worm was considered to be a weapon of competition which allowed white-tailed deer to thrive in the face of competition from less well adapted hosts. However, the story seems much more complicated. Sympatric moose populations have increased in recent years and reanalysis of historical information now questions the assumption that meningeal worms did account for the dynamics of moose populations (Whitlaw & Lankester 1994). The most recent contribution to understanding the stability conditions of this system is Schmitz & Nudds (1994).

Resource competition

Resource partitioning (sensu stricto) implies scramble competition for resources, the intensity of which depends on common use of space, habitat and food. Where resources are limiting, niche overlap is taken as a measure of potential competition. The problem is that unless it is possible to compare pre- and post-interactive patterns of resource use, little can be concluded.

An alternative approach is to evaluate characteristics of animals which have bearing on their resource requirements. Body size and trophic adaptation (morphophysiological and behavioural) have attracted most attention (Illius & Gordon 1987, Hofmann 1989, Janis & Ehrhardt 1988, Murray & Brown 1993). Although the grazer-browser continuum accounts for much of the variation in trophic ecology of African wild ruminants (McNaughton & Georgiadis 1986, Owen-Smith 1991), the relative importance of size and morphophysiological specialization is debated (Gordon & Illius 1994).

Grazing facilitation

Grazing facilitation can occur at several levels. Over several growing seasons, grazers may encourage shrub succession to the benefit of browsers, a favour that may be reciprocated. Removal of plant litter in a previous grazing season by roughage grazers may hasten green-up next spring to the benefit of concentrate selectors (Gordon 1988). Grazing in one season may delay or at least stagger plant phenology and improve the nutritional quality of plants for herbivores using the range later in the season (Anderson & Scherzinger 1975).

The concept of grazing facilitation can be traced to Vesey-Fitzgerald (1960) who described the succession of elephants, buffalo, and smaller grazers such as topi in the Rukwa valley, Tanzania. Each member of the herbivore community opened vegetation improving conditions for the next in the sequence.

Bell (1971) elaborated this succession for large mammals of the Serengeti. He described successions locally along landscape catenas and regionally by migration around the Serengeti-Mara ecosystem. He generalized that zebra lead the succession breaking down stemmy swards and exposing leafy material for wildebeest which, in turn, created the grazing lawns favoured by gazelles. Although McNaughton (1976) confirmed the synergistic relationship between wildebeest and gazelles, the association between zebra and wildebeest is not as clear. First of all, most migratory wildebeest (about 1.3 million) do not benefit from the much smaller population of zebra (200,000) and often follow a different migratory path (Stelfox et al. 1986, Caughley & Sinclair 1994). The association appears more related to cover from predation (Sinclair 1985).

Both body size and trophic adaptations have been involked as explanations for the order of the succession (Bell 1971). Because of differential scaling of metabolic rate and digestive capacity, larger animals have an inherent edge on low quality forages (Demment & van Soest 1985). Hindgut fermentors are considered more tolerant of low quality forage than ruminants (Duncan et al. 1990). Refinement of the concept among ruminants is provided by Hofmann (1989).