Foliar Concentration Of A Single Toxin Creates Habitat Patchiness For A Marsupial Folivore

Ecology, May, 2000 by Ivan R. Lawler, William J. Foley, Bart M. Eschler

IVAN R. LAWLER [1,3]

WILLIAM J. FOLEY [1]

BART M. ESCHLER [1,2]

Abstract. We examined intraspecific variation in susceptibility to herbivory by common ringtail possums (Pseudocheirus peregrinus) for two species of Eucalyptus, (E. polyanthemos and E. sideroxylon) and the chemical basis for that variation. Using a no-choice protocol, we observed dry matter intakes by common ringtail possums ranging from 3.28 to 44 g[cdot][(kg body mass).sup.-0.75][cdot][d.sup.-1] for E. polyanthemos and from 2.4 to 67 g[cdot][(kg body mass).sup.-0.75][cdot][d.sup.-1] for E. sideroxylon. We investigated, using correlative analyses, the relationships between dry matter intake and a range of foliage chemical characteristics, including measures of nutritional quality (total nitrogen, cellulose, hemi-cellulose, lignin, in vitro dry matter digestibility) and plant secondary chemistry (total phenolics, condensed tannins, cyanogenic glycosides, terpenes, and sideroxylonal, a recently identified Eucalyptus toxin). Significant relationships were identified only for terpenes (and 1,8-cineole in particular) and sideroxylonal. Bioassay experiments confirmed that the foliar concentration of sideroxylonal alone was sufficient to explain the variation observed. To undertake a field study of variation in foliar sideroxylonal and dry matter intake by possums of E. polyanthemos, we developed calibrations of both variables against the near infrared spectra of foliage samples. Acceptable calibration equations were developed, and we applied these to samples collected from a number of E. polyanthemos individuals within an area approximating the home range size of common ringtail possums. We found that foliar sideroxylonal varied from nil to 12.6 mg/g, while predicted dry matter intakes by possums ranged from nil to 42.8 g[cdot][(kg body mass).sup.-0.75][cdot][d.sup.-1]. We conclude that significant patchiness in nutritional quality of foliage, resulting from variation in foliar concentrations of a single compound, exists at a scale relevant to the feeding decisions of individual animals.

Key words: condensed tannin; Eucalyptus; habitat patchiness; herbivory; intraspecific variation; phenolic; Pseudocheirus peregrinus; ringtail possum; sideroxylonal; terpene.

INTRODUCTION

Patchiness in animal habitats is widely recognized as an important determinant of animal foraging. In studies of vertebrate foraging, patches are most often recognized in terms of the quantity of food available (Laca and Demment 1991). In contrast, the quality of the food resource is often harder to characterize, yet in some environments this may be the cause of the majority of habitat heterogeneity (Astrom et al. 1990). It is accepted that in many cases the concentration of plant secondary metabolites (PSMs) may limit the food intake of an herbivore, or deter it from feeding altogether (Cork and Foley 1991, Ganzhorn 1992), and this may contribute to habitat heterogeneity. However, the majority of studies of vertebrate foraging examine only interspecific differences in the amounts and types of PSMs present (Braithwaite et al. 1983, Oates et al. 1990, Ganzhorn 1992, Cork and Catling 1996) with some notable exceptions (Clausen et al. 1986, Reichardt et al. 1990, Snyder 1992). This is despite frequent observatio ns of animals showing preferences at the intraspecific level (i.e., between individuals within plant species). Interspecific comparisons may be obscured by qualitative differences in the foliar profiles of PSMs and relationships may be easier to discern within species, where the chemistry is qualitatively similar. Failure to recognize, and measure, such variation in field studies may lead to an inability to address the questions of interest. If some individual plants of an important food species are well defended against herbivores while neighboring conspecifics in the same area are not, measurement of leaf chemistry of only one or few individuals of each plant species may severely over- or under-estimate food availability and thus affect the interpretation of the resulting data.

The relationship between Eucalyptus species and their marsupial folivores is a case in point. Three species of arboreal marsupials are able to subsist on a diet solely of Eucalyptus foliage: the common ringtail possum (Pseudocheirus peregrinus), the greater glider (Petauroides volans) and the koala (Phascolarctos cinereus). As Eucalyptus is the dominant tree genus in Australia (Landsberg and Cork 1997), the interactions between Eucalyptus species and their marsupial herbivores have been the subject of numerous studies both in the field and with captive animals. While these studies have often identified preferences for individual trees within Eucalyptus species (Hindell and Lee 1987, Pahl 1987, Pahl and Hume 1990), the chemical basis for preferences at this level has been little studied. This has occurred for two reasons:

1) Captive animal studies have not specifically addressed intraspecific differences (Pahl and Hume 1990) and/or have relied on crude measures of PSMs such as "total" phenolics (e.g., Hume and Esson 1993) which group all compounds in the sample that bear the assayed functional group. These measures are insufficient to capture the wide variation in deterrent activity due to subtle differences in molecular structure between compounds sharing the same functional group (Clausen et al. 1986, Waterman and Kool 1994).