Bombyliidae or bee flies are a large family of ~5,000 species, these flies have a cosmopolitan distribution and are ubiquitous in arid and semi-arid regions. Bee flies are particularly important as pollinators of many desert wildflowers in Africa, including the unusual Namibian gymnosperm Welwitschia. Studies have shown that bee flies are the keystone pollinators of flowers in the Greater Cape Floristic Region. Larval bee flies are parasitoids of other arthropods, attacking a wide variety of hosts, such as grasshoppers, wasps, and even the Tsetse fly. Despite their importance as pollinators and potential use as biological control agents, many groups within bee flies are taxonomically and ecologically poorly known. Research on their natural and evolutionary history is greatly needed to advance our meager understanding of this group. In this dissertation three aspects of bee fly evolutionary biology and ecology were studied: First, the microbial ecology of bee flies was investigated. Microbial communities play an important role in the life of their hosts, influencing digestion, development, behavior, and even speciation. The microbiome of bee flies was sequenced, and the composition of their microbial communities was identified. Bee flies were found to have a primarily transient microbiome, with some core resident microbes found across all samples. The microbes found in bee flies play an important role in pollen and nectar feeding; additionally, bacteria associated with the processing of chitin were also identified.
For the second chapter of my dissertation, the phylogenetic relationships of the genus Lordotus were investigated. Prevalent in arid habitats, Lordotus is restricted to the western United States and Mexico, with most species diversity in the Southwest. To date, 29 species are described, all recorded in California. Historically Lordotus species have been challenging to delimit, due to their highly variable morphology. Previous authors have attempted to organize Lordotus into three species groups, grouping species based on the characteristics of the antennal segments and wing vestiture. Using ultraconserved elements (UCEs), I generated the first phylogenetic hypothesis for Lordotus. A UCE dataset consisting of 936 loci from 76 individuals representing all species of the genus was generated using a combination of flies preserved in ethanol and historic, pinned museum specimens The three species groups suggested by previous authors were found to be paraphyletic. These results provide a much-needed foundation for future, detailed revision of Lordotus.
The third chapter of my dissertation generated low-coverage genomes for seven bee flies, across different subfamilies. The Dipteran genomes of fruit flies and mosquitoes have long been studied, however, genomes for other groups of flies remain lacking. The assembled bee fly genomes represent a valuable resource, allowing researchers to study a wide range of topics. One such topic of interest is the identification and evolution of heat shock proteins (HSPs). These are conserved proteins found in virtually all organisms. HSPs play an important role in protecting and stabilizing DNA during periods of heat stress. Desert-adapted organisms have been shown to have higher amounts of HSPs, however, these studies have been limited to select groups. The genomes were used to identify twelve HSPs found in bee flies, with some desert species exhibiting higher amounts of HSPs. The identified HSPs will help future researchers gain a better understanding of the evolution and adaptations of desert flies.