Search

The relationship between plants and insects is one of the greatest evolutionary stories in the history of life on earth. Their importance in global terrestrial ecosystem functioning is self evident, as both represent the most abundant life on the planet. While plant-insect interactions have received much attention and are easily manipulated in experimental investigations, there have been few broad-scale phylogenetic studies for circumscribed herbivorous groups. As a result, the evolutionary role of interspecific interactions in promoting herbivorous insect diversification, at both the global and local scales, remains unclear. Remarkably, one of the largest gaps in our evolutionary and ecological understanding includes the sap-feeding insects in the Auchenorrhyncha suborder (Hemiptera), which contain some of the largest, terrestrially dominant host-plant restricted insect groups known (e.g., Cicadas, planthoppers, and leafhoppers). The evolutionary success of Auchenorrhyncha is due, at least in part, to ancient associations with a consortium obligate bacterial endosymbionts that have persisted for over 260 million years. However, like their insect hosts, the diversity and evolutionary relationships of endosymbiont associations remain relatively unknown for most aucchenorrhynchan groups.The leafhoppers (Cicadomorpha: Cicadellidae) remain one of the largest, yet poorly understood insect families. Of the 22,000 currently described species, thousands remain to be described with an overall unknown diversity (some estimates suggest as many as 90% of tropical Cicadomorpha remain to be described). This is surprising, since they offer excellent models to understand ecological and biogeographic mechanisms of species diversification due to their strict host-plant specificity, limited dispersal, and high rates of local endemism. The cicadellid subfamily, Deltocephalinae, represents the largest leafhopper groups, yet their patterns of species diversification, host-plant use, and endosymbiont associations remain almost entirely unknown. This study used the Hawaiian Archipelago as a model system to investigate the roles of ecology, biogeography, and endosymbiont interactions in the diversification of the native Hawaiian leafhopper genus, Nesophrosyne (Cicadellidae: Deltocephalinae). The Hawaiian Islands offer a tractable natural laboratory to circumscribe and study plant-insect evolution due to their isolated, discrete and replicated nature, and high levels of endemism. Nesophrosyne represents one of the most diverse and ecologically dominant herbivore radiations on Hawai`i, but has eluded scientific attention for over 60 years. Species are obligate phloem feeders and are highly host-plant specific. Moreover, Nesophrosyne exhibits the quintessential characteristics of an adaptive radiation, including dramatic morphological adaptations to the endemic Hawaiian flora and adaptive diversification across the archipelago to fill habitat types from coastal to sub-alpine regions. The specific goals of this study were to 1) update the current taxonomic status of Nesophrosyne, 2) determine the diversity and phylogenetic relationships of species in the genus, 3) infer the roles of ecology and geology in the adaptive radiation, historical biogeography, and species diversification dynamics of Nesophrosyne, and 4) to reconstruct the global relationships of the dual obligate bacterial endosymbionts of Nesophrosyne and their rates of evolution.In the first chapter, the taxonomic history and status of Nesophrosyne was reviewed. The genus was redescribed, and the subgenus Nesoreias was synonymized with Nesophrosyne. Eight new species associated with the widespread host-plant species, Broussaisia arguta (Hydrangeaceae), were described. Results reveal morphologically cryptic diversity according to individual Hawaiian Islands and volcanoes within this group. A model usage of morphological and molecular characters was developed for future delimitation of species in Nesophrosyne. The second chapter reconstructed a comprehensive phylogeny for Nesophrosyne in order to determine the origins, species diversity, and host-plant use of the native Hawaiian leafhoppers. Results support a monophyletic Nesophrosyne, originating from the Western Pacific basin, with a sister-group relationship to the genus Orosius. Nesophrosyne species are characterized by high levels of morphologically cryptic diversity and local endemicity, comprising > 200 species. Species demonstrate four dominant patterns of host-plant specialization that shape species diversity: 1) diversification through host switching; 2) specialization on widespread hosts with allopatric speciation; 3) repeated, independent shifts to the same hosts; and, 4) absence or low abundance on some hosts, suggesting herbivore interactions may limit ecological opportunity.The third chapter inferred the roles of ecology and geology in the adaptive radiation, historical biogeography, and species diversification dynamics of Nesophrosyne. The molecular age of Nesophrosyne indicates a split from Orosius 4.5 million years ago (Ma), with a basal divergence on Hawai`i 3.2 Ma. The genus originated on Kaua`i and subsequently colonized younger islands as they formed. Ancestral host-plant reconstructions reveal that the plant families, Urticaceae and Rubiaceae, played important roles in the early diversification of Nesophrosyne. Results indicate that island geography have imposed significant barriers to continued gene flow, leading to extensive allopatric speciation and intra-island diversification. Finally, Nesophrosyne diversification dynamics show an initial burst in speciation rates, with a subsequent diversity-dependent decline, corresponding to island formation. Finally, chapter four examined the global relationships of Nesophrosyne's dual obligate, bacteriome restricted bacterial endosymbionts, `Candidatus Sulcia muelleri' and a novel β-proteobacterium in the `Ca. Nasuia' genus. A global bacterial phylogeny was reconstructed, revealing a shared origin for the β-proteobacterial lineages throughout Deltocephalinae genera, and potentially throughout Auchenorrhyncha. The bacteriome association and transovarial transmission of Nesophrosyne's endosymbionts were confirmed using Fluorescent in situ Hybridization techniques. Finally, inference of absolute molecular rates demonstrates highly elevated rates of molecular evolution - the fastest so far recorded. We propose a second species in the genus Nasuia to describe the novel β-proteobacterium in Nesophrosyne.Hawai`i has long been held as a model system to understand adaptive radiation and evolutionary biology, however my study is one of the first to test these patterns directly for a hyper-diverse endemic insect radiation, and for the suborder Auchenorrhyncha. The presented results illustrate that, in diverse herbivorous groups, multiple evolutionary processes play fundamental roles in species diversification, including associations with bacterial endosymbionts, host-plant specialization, insect-insect interactions, and the geologic formation of islands. These results develop both an understanding of how ecological and geological controls shape adaptive diversification in insects, and a general model for contextualizing species diversification in herbivorous insects.

Because many aspects of aquatic and wetland plant biological control are still poorly understood, this dissertation covered work at many scales. At the scale of the field of aquatic and wetland plant biological control as a whole, meta-analysis of three decades of data showed that biological control of these plants has been generally successful, but also that experimental design may influence study results. Similarly, at a relatively smaller scale, experimental design may confound one’s ability to judge control efficacy using the milfoil weevil (Euhrychiopsis lecontei Dietz) to control Eurasian watermilfoil (Myriophyllum spicatum L.). In this system, the amount of time between surveys influenced final plant densities, especially when data were collected after mid-August when plants may be senescing. At an even smaller scale still, biological control agent behavior may influence plant control efficacy. To examine this, behavioral studies were performed on E. lecontei to better understand plant finding behavior. Understanding plant finding by E. lecontei should be directly related to understanding control efficacy, as the weevils overwinter on land and must find plants in lakes each spring. Consequently, weevils cannot control plants over multiple seasons if they cannot find them. Vision was shown to be very important to E. lecontei, even to the point that they may be able to visually differentiate plant species to select the proper host (M. spicatum). These behavioral data, along with other literature, were used to build a conceptual model for plant finding by E. lecontei that may also be extended to aquatic phytophagous insects in general. Finally, based on the work performed on visual cue use by E. lecontei and the last few decades of phytophagous insect visual ecology literature, many historical assumptions about vision being unimportant to phytophagous insects were discussed and refuted.

The density responses of insect herbivores to shifts in habitat area may depend on the degree of host-plant specialization and the composition of the matrix. I conducted a study in an experimental field comprised of plots of red clover (Trifolium pratense) that differed in size, fragmentation, and surrounding matrix. Densities of three insect herbivores were measured within the plots during two field seasons. All three species showed positive density-area relationships. The two generalist species, the meadow spittlebug (Philaenus spumarius) and the Agallian leafhopper (Agallia constricta), attained higher densities in red clover surrounded by orchard grass (Dactylis glomerata). The specialist species, the clover leafhopper (Ceratagallia agricola), attained higher densities in red clover surrounded by bare ground. The results of this study suggest that both patch area and matrix habitat strongly affect the densities of insect herbivores, and species responses largely depend upon host-plant specialization and the availability of secondary host plants.