Consequence for plantain chemistry and growth when herbivores are attacked by predators

Ecology, March, 1996 by Nancy E. Stamp, M. Deane Bowers

INTRODUCTION

Whether, and if so how, the presence of natural enemies alters the impact of insect herbivores on their host plants remains a central issue in understanding tritrophic-level interactions in natural and managed ecosystems (Price et al. 1980). More specifically, the degree to which invertebrate predators alter the foraging patterns of insect herbivores (Bernays and Graham 1988, Montllor and Bernays 1993) and its consequences for host plants are unclear (Price et al. 1980). In several studies, the presence of invertebrate predators caused herbivorous prey to disperse (Ito and Miyashita 1968, Yamanaka et al. 1972, Nakasuji et al. 1973, Mansour et al. 1981, Stamp and Bowers 1988, 1992b, 1993), but the effect on the host plant was not reported. Invertebrate predators can reduce herbivore damage and this may indirectly lead to increased plant yield (Gould and Jeanne 1984, Whittaker and Warrington 1985, Riechert and Bishop 1990, Ito and Higashi 1991, Clark et al. 1994), but the mechanisms involved were not determined in these studies. Reduction in herbivory may reflect predators killing herbivores and/or herbivores altering their behavior in the presence of predators. Predators that hunt in different ways, such as stink bugs, which search leaves and stalk prey, vs. wasps, which search the ground as well and straightforwardly attack prey, are likely to have different indirect effects on herbivores.

The concept that "enemies should be considered as mutualists with plants and part of plant defense" (Price et al. 1980:58; Price 1986) is a generally accepted tenet of the theory of plant-herbivore interactions. Nevertheless, the idea has proven very difficult to document (reviewed by Hare 1992). If this relationship is indeed mutualistic, a net gain, relative to costs, is required for both the plant and the natural enemies of the herbivore. The fact that an herbivore that eats a plant is killed by a predator is not sufficient evidence of a net gain for both plant and predator. For example, the growth of an insect predator can be reduced by prey that contain plant toxins (Stamp et al. 1991, Malcolm 1992, Paradise and Stamp 1993, Traugott 1994). Negative effects, such as this, would jeopardize the development and maintenance of a mutualism.

Here we report the results of an experiment on plant-herbivore interactions in the presence and absence of two levels of two different types of invertebrate predators. Our system consisted of two lepidopteran species (Junonia coenia Hubner: Nymphalidae and Spilosoma congrua Wlk.: Arctiidae) that differ in feeding specialization on narrow-leaved plantain (Plantago lanceolata L.: Plantaginaceae) and which were subjected to predation by wasps (Polistes fuscatus Fabricius: Vespidae) and stink bugs [Podisus maculiventris (Say): Pentatomidae]. This system is particularly appropriate for study of the effects of insect predators because of the degree to which we can manipulate this system and document the consequences. Specifically, our objectives were to examine defensive chemistry (iridoid glycosides), nutritional quality (leaf nitrogen), and growth response (aboveground plant mass) of plantain subjected to herbivory by caterpillars of either the specialist Junonia, or the generalist Spilosoma, in the presence and absence of either wasps or stink bugs. Some evidence of insect predators having an effect on host plants by way of the predators' interactions with insect herbivores would be alterations: (1) in plant mass, and/or (2) in the concentrations of plant defensive chemicals.

METHODS

The system

We contrasted a specialist feeder with a more generalized feeder because these two types tend to behave differently, in amount of movement among plants, exposure to predators, and defense against predators (Stamp 1992, Stamp and Bowers 1992a). We hypothesized that predators would have different direct and indirect effects on these two types of herbivores, which in turn might have different consequences for the host plants. Junonia coenia caterpillars are specialists on plants containing iridoid glycosides (Bowers 1984), and narrow-leaved plantain is one of their major host plants (Shapiro 1974, Scott 1975, 1986, Bowers 1984). This plant species contains the iridoid glycosides catalpol and aucubin (Duff et al. 1965, Bobbitt and Segebarth 1969). Aucubin is the biosynthetic precursor of catalpol (Damtoft et al. 1983). Catalpol is the more toxic of these two iridoid glycosides to generalist herbivores (Bowers and Puttick 1988, Bowers 1991, 1992). Consequently, the amount of catalpol relative to total iridoid glycosides is biologically significant. Iridoid glycosides serve as feeding stimulants for Junonia larvae (Bowers 1984). Several species of generalist caterpillars also eat plantain (Tietz 1972). For some arctiid caterpillars, plantain ranks relatively high in acceptability (Villiard 1975, Dethier 1988a, b, 1989). For this study we chose a tiger moth caterpillar, Spilosoma congrua Wlk. (Arctiidae), which occurs in eastern North America and feeds on a wide variety of herbaceous plants, including plantain (Tietz 1972). Feeding by these species can induce higher iridoid glycoside concentrations in P. lanceolata (Bowers and Stamp 1993).