Floral characters link herbivores, pollinators, and plant fitness

Ecology, Sept, 1997 by Sharon Y. Strauss

INTRODUCTION

Both herbivores and pollinators have been shown to have large effects on plant fitness (e.g., references in Crawley 1983, Jones and Little 1983, Marquis 1992). Although much is known about the individual effects of herbivores and pollinators, relatively few studies have examined the interactive effects of these species on plant ecology and evolution. The focus of this feature is to illustrate how simultaneous consideration of several forces acting on organisms in concert may provide us with a broader understanding of evolutionary and ecological pressures faced by individuals. In this context, I will present some ideas on how our approach to measuring plant fitness in the past may have biased us against detecting combined effects of herbivores and pollinators on plant fitness. I will discuss ideas, as well as describe some of my own experiments in progress that pertain to these areas. Because we lack information to address many of the possibilities raised here, this discussion will serve mostly to provide approaches that may help us understand how pollination and herbivory together affect plant fitness.

OVEREMPHASIS ON SEED PRODUCTION AS A MEASURE OF FITNESS

Typically, the effects of pollinators and herbivores have been assessed by measuring plant seed production, both in terms of seed number and seed mass. Male plant fitness, or the number of seeds sired by pollen, is responsible for half the genes contributed to the next generation in sexually reproducing plants, but has been ignored until relatively recently. A new focus on male plant fitness (e.g., Young and Stanton 1990, Devlin et al 1992, Snow 1994, Quesada et al. 1995, Mutikainen and Delph 1996, Delph et al. 1997) has opened the door to examining effects of herbivores and pollinators on total plant fitness. A few important studies have examined the relationship between male and female plant fitness and have found that they range from negatively to only weakly positively correlated (Bertin 1982, Ennos and Dodson 1987, Schlichting and Devlin 1989, Broyles and Wyatt 1990, Devlin and Ellstrand 1990). The effects of herbivory on both male and female plant fitness have yet to be documented.

The importance of estimating both components of total fitness arises because plants may shift allocation from female to male components of fitness, or vice versa, in response to damage (e.g., Whitham and Mopper 1985, Allison 1990, Snyder 1993, Frazee and Marquis 1994). Thus, our understanding of what happens to female plant fitness as a result of herbivory or pollination may not accurately reflect total plant fitness. By considering total plant fitness, the interrelationships between herbivores and pollinators as mediated through their host plants may become more apparent. Fig. 1 depicts various direct and indirect means through which herbivory might affect both male and female plant fitness. Note that the diagram is not inclusive and omits many feedback loops that might occur (e.g., damage to fruits might affect subsequent numbers of flowers produced, etc.). I have highlighted with bold arrows the subset of these relationships that will be emphasized in this discussion.

Example: effects of herbivory on pollen production and quality

To date, no study has investigated the effects of herbivory on male plant fitness, though several recent studies have examined effects on a number of potential correlates with male fitness. First, pollen production per flower has been shown to decrease as a result of herbivore damage to plants (Frazee and Marquis 1994, Quesada et al. 1995, Strauss et al. 1996; K. Lehtila and S. Strauss, unpublished data). Second, there is growing evidence that pollen quality is also affected by foliar damage; pollen from damaged plants is less successful at siring seeds when it competes with pollen from undamaged plants (see Delph et al. 1997 for review). Third, shifts in allocation to male and female function can also be caused by herbivory. For example, in response to herbivory, numbers of male strobili increased relative to female strobili in several species (Whitham and Moppet 1985, Allison 1990). Thus, three important components of male function and reproductive success have been shown to be affected by herbivory.

FOLIAR HERBIVORY CAN AFFECT BOTH MALE AND FEMALE PLANT FITNESS INDIRECTLY

The effects of herbivores on the use of plants by pollinators may provide a fourth, indirect pathway through which herbivory can affect both male and female plant fitness. Plant responses to herbivory may cause changes in floral attraction and reward characters that in turn affect plant use by pollinators. Whereas many studies have shown effects of herbivory on flower production, there have been relatively few studies that examine the response of individual floral attributes to herbivory. This lack of information may stem from the suggestion that variation in floral traits should have such a great impact on plant fitness that these traits should be highly conserved. For example Harper (1977: 195) states "The size and form of these units is usually tightly canalized and changes only fractionally over widely varied environmental conditions." In addition, if seed production was considered the bottom line for plant fitness, then there was little incentive for investigators to put time and resources into examining floral characters that were one step removed from fitness measures. However, floral characters affect both male and female plant fitness and have been shown to respond to herbivory.