While ant colonies serve as host to a diverse array of myrmecophiles, few parasitoids are able to exploit this vast resource. A notable exception is the wasp family Eucharitidae, which is the only family of insects known to exclusively parasitize ants. Eucharitidae (Chalcidoidea) contains 54 genera and over 500 described species. Worldwide, eucharitids attack five subfamilies across the ant phylogeny. To evaluate phylogenetic relationships, a molecular dataset was used (18S, 28S-D2, 28S-D3-5, COI, and COII). Eucharitid diversification is examined, including timing of key evolutionary events, biogeographic patterns, and potential cophylogeny with ant hosts. Eucharitidae arose approximately 50 mya after their hosts, during the time when the major ant lineages were already diversifying. Even after multiple dispersal events to the New World and extensive speciation within biogeographic regions, eucharitids remain parasitic on the same ant subfamilies as their Old World relatives, suggesting host conservatism despite access to a diverse novel ant fauna.

A speciose group of eucharitids is the `Kapala clade', which is a group of genera found mainly in the Neotropics. It includes some of the most morphologically bizarre members within the family. Their monophyly is supported in both morphological and molecular analyses. However, relationships among genera within the clade are not so easily resolved. Within this clade, Kapala Cameron is one of the most diverse and easily recognizable genera, with species distributed across the Neotropical region and one disjunct species found in the Afrotropical region. Divergence dating and estimated mutation rates suggest that the age of this species, K. ivorensis, is >1 my, thus predating a human introduction. A morphological matrix of 52 characters was analyzed both separately and combined with molecular data. There was support for the Kapala clade and the described genera sampled, with three major clades being recovered. The genus Kapala was rendered paraphyletic by two distinct clades of other kapaline genera. A redescription is provided for the type species, Kapala furcata, and for Kapala cuprea Cameron 1913. Three new species are described, K. deltalis sp. nov., K. parafurcata sp. nov., and K. quasimodo sp. nov.

Phylogenetic studies on two different groups of Chalcidoidea were conducted. The subfamily Pteromalinae (Hymenoptera: Pteromalidae) was analyzed using the 28S D2-D5, Cytochrome b and Cytochrome Oxidase II gene regions, with focus on discovering the nearest relatives of the genus Nasonia. No support was found for monophyly of Pteromalinae, perhaps in part because of the low 28S molecular divergence within the group. Both maximum likelihood and Bayesian results indicate that Nasonia renders Trichomalopsis paraphyletic, and that Trichomalopsis sarcophagae is the sister group of the genus Nasonia. Nasonia and T. sarcophagae are both infected by Wolbachia bacteria, suggesting a history of Wolbachia infection that could be older than Nasonia itself. Other clades of Pteromalini that were 100% infected with Wolbachia among sampled species include the Australasian genus Pseudanogmus and Coelopisthia + Diglochis. Support was found for a monophyletic assemblage of taxa historically placed in Miscogasterinae, and the subfamilies Cratominae, Miscogasterinae and Panstenoninae are synonymized under Pteromalinae n. syn. The genus Diconocara is transferred from Pteromalini to Miscogasterini n. stat. based on both molecular and morphological data.

A combined molecular and morphological phylogeny of the family Eulophidae is presented with focus on relationships within the subfamily Entedoninae. The 28S D2-D5 and CO1 gene regions were examined in partitioned maximum likelihood and Bayesian analyses, and an additional Bayes analysis was conducted to oberserve the effect of historically recognized morphological characters on the results. Eulophidae was strongly supported as monophyletic only when the genus Trisecodes was excluded. The subfamilies Eulophinae, Entiinae (=Euderinae) and Tetrastichinae were consistently monophyletic, but monophyly of Entedoninae was supported only in the combined morphological and molecular analysis. The lack of support from the molecular data was likely due to the form of the 3e´ subregion of the 28S D2 rDNA in the nominal subgenus of Closterocerus. The tribe Euderomphalini was excluded from a monophyletic Entiinae, suggesting that it should be retained in Entedoninae. Opheliminae n. stat. is raised from unplaced tribe to subfamily status, and a sister group relationship of Opheliminae + Entiinae was strongly supported. The genera Neochrysocharis n. stat. and Asecodes n. stat. were removed from synonymy with Closterocerus because molecular data corroborate their morphological differences. Closterocerus (Achrysocharis) germanicus was transferred to the genus Chrysonotomyia n. comb. based on molecular and morphological characters.

The posterior surface of the head was examined in several families of Chalcidoidea and interpreted according to theories of head capsule evolution as proposed by Snodgrass. Most chalcidoids have only a hypostomal bridge, but some species in the families Chalcididae, Eurytomidae, Pteromalidae, and Torymidae have postgenal bridges that may have been independently derived. Species of small-bodied parasitic wasps with a reduced head capsule, such as many Aphelinidae, Mymaridae, and Trichogrammatidae, lack important landmarks and cannot be easily interpreted without making inferences from related species. Several features provide potentially useful phylogenetic information, such as the presence of a postgenal bridge, extent of the hypostomal carina, extent of secondary posterior tentorial pits, and form of the mesal lamellae extending from the foramen magnum to the oral cavity. However, in many cases these characters present problems of homology that require a larger phylogenetic context to answer.

The subfamily Oraseminae (Hymenoptera: Chalcidoidea: Eucharitidae) is a circumtropical group of parasitoid wasps that specialize on ants in the subfamily Myrmicinae (Hymenoptera: Formicidae). Multiple phylogenetic inference methods, including parsimony, maximum likelihood, Bayesian, and coalescence, with the largest to date sampling of taxa for this subfamily allow us to elucidate their dated biogeography, ancestral host ant genus, diversification rates, and provide valuable data for revising the constituent genera and species. Using molecular datasets comprised of Illumina sequences from Anchored Hybrid Enrichment probe capture, Sanger sequences from paired primers, and a combination of the previous two datasets, we show that Oraseminae has likely dispersed from the Old World into the New World across a tropical Beringian Land Bridge approximately 24–33 Ma. This dispersal, which seemed to have occurred around the same time but in the opposite direction of their ancestral host ants in the genus Pheidole, led to an increased diversification rate in the exclusively New World genus, Orasema. Including the species described herein, Orasema comprises 94 described species organized into 16 species groups with several unplaced taxa. There are 22 new species described and 7 redescriptions with comprehensive distribution maps, plant and ant host association data, and digitally imaged photo plates, which represents the largest revision of this genus to date. Within the superfamily Chalcidoidea, eucharitid wasps, along with their sister family, Perilampidae, have a unique morphology for their first-instar larvae called planidia, which are hypermetamorphic, highly mobile, and strongly sclerotized. Relationships to this planidial clade have historically been difficult to establish. A morphological phylogenetic analysis of the planidial larvae that includes the newly discovered, less-derived larva in the species Eutrichosoma mirabile Ashmead (Pteromalidae) has provided evidence of a close relationship to the Eucharitidae and Perilampidae, allowing us to make more informed predictions on the origin and natural history of planidial larvae in Chalcidoidea.

The tribe Cirrospilini (Hymenoptera: Chalcidodea: Eulophidae) is revised in a combined molecular and morphological phylogenetic context, using ribosomal (18S, 28S, ITS2) and mitochondrial (COI) data, and scoring specimens for morphological characters. Twenty genera are now recognized in the tribe. Colpoclypeus Lucchese and Trichospilus Ferrière are included in Cirrospilini, while Cirrospiloidelleus Girault is removed. Melittobiopsis Timberlake is also removed from Cirrospilini, and raised to Melittobiopsini Perry n. tribe., sister to Eulophini Ashmead. Cirrospilus Westwood is revised and is split into 6 morphologically distinct genera: Atoposoma Masi stat. rev., Burkseus Perry, Cirrospilus str. s., Gyrolasella Girault stat. rev., Pseudozagramma Perry n. gen., and Vagus Perry n. gen. The new synonymy of Semielacher Bouček under Cirrospilus is proposed. The first worldwide key to all genera is presented, as well as distribution maps, and extensive specimen, host, and host plant records. Zagrammosoma Ashmead is molecularly revised using ribosomal (28S, ITS2) and mitochondrial (COI) data, with 24 species now recognized, including descriptions of 9 new species. A worldwide key, distribution maps, and image plates of all species are provided.

Myriophora is a near globally distributed species-rich genus whose members are parasitoids of toxic, chemically defended millipedes. Prior to this study, only eight species were described, little was known about how they interacted with millipedes, and no phylogenetic hypotheses of intrageneric relationships had been proposed, resulting in a wide range of available directions to pursue understanding the natural history of Myriophora. The approach taken here is primarily from a systematics standpoint of understanding species level diversity, distribution, and relationships, but with the intent of layering on data from field observations and experiments to not only increase basic biological knowledge of Myriophora but to also then inform and enhance the primary systematics goals. To make feasible taxonomic headway under the time constraint of a Ph.D. project, the taxonomic revision focused only on the New World taxa, resulting in the description of 57 new species, covering a geographic area from Argentina to Canada. Several experiments were carried out aimed at determining how Myriophora locate millipedes. After concluding that the secretions are used as the long-range cue for host-finding, individual components of the benzoquinone blend were tested for attractiveness, and it was found that of the two predominate quinone components in the defensive blend of juliform millipedes, 2-methoxy-3-methyl-1,4-benzoquinone alone attracts adult flies of both sexes, and that a combination of 2-methoxy-3-methyl-1,4-benzoquinone and 2-methyl-1,4-benzoquinone increased attractiveness. Finally, to understand these findings in an evolutionary context, morphological and molecular character data were used to estimate the phylogeny of Myriophora. We find that the ancestral area of Myriophora is the New World and a single dispersal event to the Old World likely occurred via a Beringian connection before subsequently spreading to the Afrotropical Region. The ancestral host reconstructed for Myriophora is the benzoquinone-producing Juliformia, and this association has been conserved in the Old World clade. Species of Myriophora associated with the cyanide-producing millipedes are restricted to a single New World clade.

Aphelinidae Thompson (Hymenoptera: Chalcidoidea) are a biologically diverse family of parasitoids wasps, many of which are used in classical and augmentative biological control of economically impactful pest species, including whiteflies, scale insects, and aphids (Hemiptera: Sternorrhyncha). Historically, the systematics of Aphelinidae formed a Gordian knot: an intractable problem that could not be solved by available methods, due to extreme morphological homoplasy. Next-generation molecular methods have provided the necessary tools to approach Aphelinidae in a meaningful way. Here we develop a robust phylogenetic hypothesis for Aphelinidae and its internal relationship for an exonic anchored hybrid enrichment data set (186 taxa, 740 loci), explored as both nucleotide and amino acids (chapter 2). We find the family, all subfamilies, three tribes, and several major clades to be monophyletic. We propose Coccidae (Hemiptera: Sternorrhyncha) as the ancestral host of the family, and identify five major host shifts within the family. Finally, we propose an early cretaceous origin for the family, placing it congruently in evolutionary time with other major insect radiation events. Using our recovered hypothesis, we revise the classification of Aphelinidae at the family, subfamily, tribe, and generic levels (chapter 3), and we provide a comprehensive review the family and dichotomous key to tribes (chapter 1). Finally, we begin the revisionary taxonomy of monophyletic species-groups of Encarsia Förster, the most speciose genus in the family with 477 described species and an estimated species diversity of ~4000. Here we treat 79 species to revise the Encarsia strenua species-group, and establish 3 additional species groups (E. japonica species-group, E. pulliclava species-group, and E. quercicola species-group). The work presented here establishes the first comprehensive phylogenetic hypothesis for Aphelinidae, provides a modernized cladistic classification, and puts the family of a firm foundation for additional phylogenomics and revisionary taxonomy.