Armored scale insects have long been known to harbor endosymbionts, but until recently the endosymbionts have remained unidentified. Using DNA data I have identified the primary endosymbiont as a bacterium from the phylum Bacteriodetes. To accomplish this, I amplified DNA sequences from two genes of the endosymbiont and use these sequences (16S and 23S, 2105 total base pairs), along with previously published sequences from the armored scale hosts (elongation factor 1α and 28S rDNA) to investigate phylogenetic congruence between the two clades. The Bayesian tree for the bacteria is roughly congruent with that of the hosts, with 67% of nodes identical. The high level of congruence between the topologies indicates that these Bacteroidetes are the primary endosymbionts of armored scale insects. To investigate the phylogenetic affinities of these endosymbionts, I aligned some of their 16S rDNA sequences with other known Bacteroidetes endosymbionts and with other similar sequences identified by BLAST searches. I found these endosymbionts to be closely related to bacteria associated with eriococcid and margarodid scale insects, and enodosymbionts of cockroaches and auchenorrynchan insects, and proposed the name " Candidatus Uzinura diaspidicola " for the primary endosymbionts of armored scale insects. I have also investigated secondary endosymbionts found in the armored scale insect species Aspidiotus nerii. A host manipulating endosymbiont similar to Wolbachia, Cardinium was found in Aspidiotus nerii and there is circumstantial evidence that Cardinium is responsible for inducing parthenogenesis in Aspidiotus nerii. Using PCR and sequencing of 16S rDNA, we have tested 593 individuals and obtained positive PCR results in 67 individuals of 34 populations representing 20. A phylogenetic analysis including all known insect-associated Cardinium 16S sequences shows patterns of horizontal transmission of this endosymbiont among insects. Lastly, I tested molecular evolution rates of Uninura from sexual and parthenogenetic populations of Aspidiotus nerii using five loci: two host nuclear loci, one mtDNA loci, and two symbiont loci. I hypothesized that the symbionts and mitochondria would evolve at similar rates, however, though the endosymbionts are vertically transmitted and experience similar bottlenecks to the mitochondria, the data suggest that they evolve at similar or slower rates to the nuclear DNA.