Interaction between food availability and predation mortality mediated by adaptive behavior.

Key words: antipredator behavior; behavioral indirect effects; density dependence; growth-rate-mortality-rate trade-offs; interaction modification; regulation; Rana catesbeiana; Tramea lacerata.

INTRODUCTION

Many animals must actively search for their food. The faster an animal searches, or the more time spent searching, the more likely it is to encounter food items. However, both of these components of activity will also increase an animal's encounter rate with predators (Gerritsen and Strickler 1977). Prey movement can also enhance prey detection by some predators (Curio 1976, Taylor 1984). Thus acquiring resources to produce additional offspring decreases the probability of surviving long enough to realize the production of those offspring.

This trade-off is particularly clear in larval anurans (tadpoles). Tadpoles feed by scraping attached algae and detritus or filtering suspended particles while swimming (Seale and Wassersug 1979, Wassersug and Hoff 1979). More active tadpoles grow faster and are often better competitors both interspecifically (Woodward 1982. Morin 1983, Lawler 1989, Werner 1991, 1992a) and intraspecifically (Skelly and Werner 1990). However, more active tadpoles also appear to be more vulnerable to their invertebrate predators (interspecific; Lawler 1989, Azevedo-Ramos et al. 1992; intraspecific; Skelly 1994).

Because reproduction and survival, the two components of Darwinian fitness, are both functions of activity, an adaptive resolution of the trade-off can be proposed in terms of activity. Models that incorporate risk of mortality as a trade-off with resource acquisition (McNamara and Houston 1987, 1994, Abrams 1993, Werner and Anholt 1993) predict reduced activity when mortality risk increases. Considerable empirical evidence exists to support this prediction (reviewed in Sih 1987, Lima and Dill 1990). Rarely tested, though, is the prediction that foraging activity should also be lower at high food levels (McNamara and Houston 1987, 1994, Abrams 1993, Werner and Anholt 1993) when high food levels are not transient (McNamara and Houston 1994). There is considerable empirical evidence for this prediction as well (reviewed in Werner and Anholt 1993). This prediction stands in contrast to the case where mortality risk is independent of foraging activity. When this is the case, foraging effort should increase at higher resource levels (Norberg 1981, Dunbrack and Giguere 1987, McNamara and Houston 1994). Therefore, changes in behavior with food level provide a critical test between situations where mortality risk is a function of foraging effort and those where it is not.

If the trade-off between food acquisition and mortality is indeed mediated by foraging activity, we should also be able to demonstrate that increased food level leads to a decreased predation mortality rate because activity has been reduced at the higher resource level.

In this study we measured the activity of larval anurans (bullfrog tadpoles, Rana catesbeiana) at different resource levels when caged larval dragonflies (Tramea…