The ecological consequences of variation in plants and prey for an omnivorous insect

Ecology, June, 1999 by Micky D. Eubanks, Robert F. Denno

INTRODUCTION

There is a growing awareness among ecologists that trophic interactions involving omnivores, animals that feed at more than one trophic level, are more prevalent in nature than previously supposed (Sprules and Bowerman 1988, Vadas 1989, Polis 1991, Whitman et al. 1994, Alomar and Wiedenmann 1996). As a result, authors have begun to incorporate interactions involving omnivores into current theories of population and community dynamics (Polis et al. 1989, Polis 1991, Morin and Lawler 1995, Polis and Strong 1996, Holt and Polis 1997). Because omnivores do not depend on a single resource, their population dynamics may differ markedly from those of herbivores or predators whose life histories and dynamics are strongly associated with resources at a single trophic level (Pimm and Lawton 1977, 1978, Morin and Lawler 1995, Polis and Strong 1996). Most importantly, feeding on alternative food sources may allow omnivores to persist when any single resource is scarce, potentially decoupling the dynamics of omnivores and their resources and promoting top-down control (Walde 1994, Morin and Lawler 1995, Moran et al. 1996, Polis and Strong 1996). However, models exploring the impact of omnivores or other complex consumers on community dynamics tend to assume that all resources are of equal or approximately equal value to the omnivore or consumer (Pimm and Lawton 1977, 1978, Morin and Lawler 1995, Polis and Strong 1996; but see Coli and Izraylevich 1997). This is an unrealistic assumption that contradicts the current movement to incorporate biological realism into models of food webs and community dynamics (Paine 1988, Hunter and Price 1992, Polis and Strong 1996).

Hence, a fundamental and essentially unanswered question concerning the ecological consequences of omnivory is how omnivores respond to variation in resources at more than one trophic level. For example, omnivores that feed on plants and prey encounter variation in the quality of both as food (Kiman and Yeargan 1985, Bjorndal 1991, Milne and Walter 1997). If resources at both trophic levels are roughly equivalent and essentially interchangeable, then feeding at more than one trophic level could provide omnivores with ecological flexibility by releasing these animals from a dependence on any single resource (Polis and Strong 1996). On the other hand, if resources at different trophic levels are not nutritionally equivalent, then omnivores may depend on a combination of resources and may be forced to simultaneously track resources at multiple trophic levels. Under these conditions, omnivores may be more sensitive to environmental variation than strict predators or herbivores and the hypothesized ecological flexibility associated with omnivory an illusion.

The goal of this study was to experimentally determine the effects of plant quality and prey species on the performance and population size of an omnivorous insect and to compare the response of omnivores and strict predators to variation in both plants and prey. By manipulating variation at two trophic levels, we quantified the degree of an omnivore's reliance on both prey and plant food. Furthermore, we explicitly tested the hypothesis that the ability to feed at two trophic levels promotes the population persistence of an omnivore when a resource at one trophic level is limited.

We studied a relatively simple, agricultural system consisting of an omnivorous "predator," the big-eyed bug (Geocoris punctipes), two of its common prey species, pea aphids (Acyrthosiphum pisum) and eggs of the corn earworm (Helicoverpa zea), and a shared host plant, lima beans (Phaseolus lunatus). Big-eyed bugs consume the pods and leaves of lima beans as well as pea aphids and corn earworm eggs. In turn, pea aphids and corn earworm larvae feed on the pods and leaves of lima beans. Preliminary observations and other studies of insect predation and herbivory suggest that lima bean plants and these prey species might vary in their nutritional quality as food for big-eyed bugs (Champlain and Scholdt 1966, Schumann and Todd 1982, Cohen and Debolt 1983). The specific objectives of this study were to: (1) determine the effects of variation in the presence of pods and prey species on the survival of big-eyed bugs, (2) determine if plant feeding allows big-eyed bugs to survive periods when prey are unavailable, (3) determine if the presence of pods and prey affects the dispersal of big-eyed bugs, (4) quantify the effect of pods on the population size of big-eyed bugs, and (5) compare the response of omnivores and strict predators to variation in lima beans and prey. These data represent one of the first and most comprehensive investigations of the combined effects of plants and prey on the dynamics of omnivores.

STUDY SYSTEM

Omnivores

Big-eyed bugs, Geocoris punctipes (Heteroptera: Lygaeidae), are abundant in lima beans, other agricultural crops, and old fields (Crocker and Whitcomb 1980). Big-eyed bugs are active in Maryland from mid-May until early October, producing three generations per year with peak densities in late July and early August. Big-eyed bugs are omnivorous, generalist "predators" and their diverse range of prey species includes aphids and lepidopteran eggs (Champlain and Scholdt 1966, Lawrence and Watson 1979, Crocker and Whitcomb 1980, Cohen and Debolt 1983). We have observed big-eyed bugs feeding on both pea aphids and corn earworm eggs in Maryland lima beans (Eubanks 1997). Big-eyed bugs also feed on the pods, seeds, and leaves of many plant species (Stoner 1970, Crocker and Whitcomb 1980, Naranjo and Stimac 1985, Thead et al. 1985). Despite their plant feeding habits, big-eyed bugs are not pests of lima beans or other agricultural plants, probably because their short proboscis and digestive enzymes do not cause extensive damage to plant tissues (Crocker and Whitcomb 1980, Nesci 1996).