Plant-insect interactions have long been studied due to their ecological and economic importance. However, little is known about how flowers avoid being damaged, despite their importance for plant reproduction. Flowers interact with mutualists, such as beneficial pollinators, but also with antagonists that damage flowers or reduce plant reproduction. When multiple insects use flowers concurrently, there is the potential for competition. Such flower-insect interactions can be mediated by floral traits, including chemical defenses and nectar rewards, but the role of these traits is largely unknown. The goal of this research is to determine how floral chemical defenses and nectar rewards mediate mutualist and antagonist insect interactions with flowers. To do this, the researcher will perform two experiments: (1) develop a model of how mutualist and antagonist insects, floral defenses, and nectar rewards interact with each other and affect plant reproduction; (2) manipulate pollinators and antagonists to determine how they affect each other, interact with floral defenses and rewards, and influence plant reproduction. This research will comprehensively examine interactions between floral competitors, floral defensive and attractive traits, and effects on plant reproduction.
Understanding insect interactions with flowers can be imperative for agriculture and conservation. Pollinators are estimated to provide $100 billion USD in services each year, and insects that damage flowers can devastate crops, vector diseases, cause invasions, or act as biological control agents. Understanding how floral competitors interact with one another and their host plants is critical for managing crop yield and species diversity. Additionally, this research has and will continue to provide many opportunities for mentoring undergraduates, including students from under-represented minorities. The results from this work will provide protocols for chemical defense analysis in an ecological model system. The researcher will also communicate results widely by publishing in peer reviewed journals and presenting at national conferences
The goal of this research was to determine now floral defenses and rewards mediate floral mutualist and antagonist interactions. This was accomplished using two large experiments and three smaller ones. Florivory Manipulation We applied florivory treatments (control, intermediate florivory, or intense florivory) to plants and measured effects on flower traits, insect visitation, and plant reproduction. We found that florivory had no effect on floral chemical defenses or floral attractiveness, suggesting that plants do not change flower traits in response to floral damage. However, florivory significantly decreased leaf herbivory, but significantly increased subsequent natural florivory. This may suggest that florivory can induce chemical defenses in leaves, or that generalist feeders may move from leaves to flowers after flower damage, either because flowers become more attractive or leaves become less so. Additionally, we found that florivory decreased the amount and proportion of outcrossing reproduction, suggesting that I. capensis may tolerate florivory by producing fewer of their conspicuous outcrossing flowers. Interestingly, the strongest effects on plant reproduction occurred when there was moderate (not intense) levels of florivory, suggesting a damage threshold above which plants no longer respond to florivores. Manipulation of Three Flower Insects We manipulated three flower-using insects: florivores (30% flower tissue removed), nectar robbers (piercing the corolla and removing nectar), and pollinators (using a paintbrush to hand pollinate). We used a fully factorial 2x2x2 design with every possible combination of treatments, leading to eight different treatment combinations. As in our previous experiment, we found that florivory significantly decreased outcrossing reproduction but not selfed reproduction, leading to a shift towards relying more on selfing when florivory is common. Nectar robbers and nectar thieves (insects that take nectar but are too small to pollinate) preferentially visited undamaged plants. Conversely, florivores preferred plants in the florivory and/or nectar robbing treatments, suggesting they prefer plants with flower damage. There were also many significant multi-way interactions between the three treatments, indicating that the effect of one insect visitor on a plant will depend on what other insects are present. Mapping Vasculature To map vascular connections in I. capensis, we applied Rhodamine-B dye to cut leaf petioles. The dye travels through the vasculature of the plant, showing what parts of the plant are directly connected. We found that dye moved much farther and more frequently up than down the stem, and that leaves on the opposite side of the plant never showed dye. This suggests strong vascular transport towards the plant apex and very few horizontal vascular connections (Figure 1). Induced Floral Defenses We treated one flower on each plant (no damage, 30% tissue removed, or 60% tissue removed) and measured defenses in flowers at four time points and from three locations. Floral condensed tannins showed no change after flower damage. However, floral anthocyanins were induced, but only after moderate (not severe) floral damage, and only locally (on the same branch as the originally damaged flower). This suggests that anthocyanins may be induced by flower damage, but that signals to induce defenses may not travel very far, and that there is a damage threshold above which no defenses are induced. Mechanical versus Insect Floral Damage We damaged flowers using scissors or florivorous beetles and compared the effects to control plants. We found no difference between mechanical and insect damage in terms of floral defenses, plant growth, subsequent pollinator visitation, or plant reproduction. This suggests that using scissors to damage flowers is an adequate treatment to replicate natural florivory. Additionally, because there was no evidence of induced defenses after either kind of damage, flowers may rely more on constitutive defenses or tolerance mechanisms than induced resistance to combat florivores. Outcomes and Broader Impacts We mentored sixteen undergraduate students, five of whom were from underrepresented minorities, and two who completed undergraduate honor’s thesis projects. Additionally, students from a senior-level plant ecology course were able to visit the field experiments, collect data, and make and test their own hypotheses. Data from this research has been presented both locally and nationally at scientific conferences. Also, this research was featured on The Academic Minute, a radio program produced by WAMC Northeast Public Radio broadcast throughout the northeast United States. The program is archived for anyone to listen to on their website. We are also in the process of preparing at least three publications based on our data. All insects collected as part of this project were donated to the University of Massachusetts insect collection for research and teaching. We are publishing our methods for extracting anthocyanins and condensed tannins from I. capensis. This will allow future researchers to incorporate chemical ecology into what is already a popular model system for studying plant physiology and reproductive ecology. Seeds that have been collected as part of this research have been supplied to other researchers, both at the University of Massachusetts and elsewhere.
