Degree Name

Master of Science (MS)

Semester of Degree Completion

Spring 2019

Thesis Director

Zhiwei Liu

Thesis Committee Member

Ann H. Fritz

Thesis Committee Member

Gordon C. Tucker


Gall making insects form a special feeding guild of phytophagous animals, and by manipulating host plant tissue differentiation, are able to avoid plant chemical defenses and thus have no need for counter defense mechanisms. Host plant selection is crucial to the evolution of these insects because successful gall formation is largely dependent on host plant ability to respond to stimuli. In Illinois and neighboring states, Antistrophus gall wasps associated with the rosin weed (S. integrifolium) and the cup plant (S. perfoliatum) are morphologically indistinguishable and thus have been treated as belonging to single species. However, the wasps from the host plant species display strong host preference to the host plant species from which they are reared as well as other life history differences, suggesting that they may represent two distinct, although cryptic, species. In order to test the competing hypotheses regarding the identity of these wasps, I investigated whether there exists genetic discontinuity between wasp populations defined by host plant species based on molecular data. Wasps associated with either host plant species, mostly from sympatric populations, were collected from Illinois, Wisconsin, Indiana, Missouri and Iowa and extracted total DNA were PCR amplified and sequenced for three DNA loci, including two mitochondrial genes, (Cytb and COI) and one nuclear gene (28S D2). Phylogenetic reconstruction for all samples based on the Cytb and COI genes did not recover any monophyletic lineage consisting exclusively of wasps sampled from either of the two host-plant species but revealed a clear pattern of lineage bias toward host plant species in the tree resolved from Cytb gene. Examination of haplotype diversity revealed five haplotypes among our Cytb (n= 60) and COI (n= 59) sequences, respectively and three haplotypes for the 28S gene (N= 65). Further analyses of the genealogical relationship among the haplotypes using network method uncovered a distinct host affiliation pattern for Cytb, but not for the COI gene, similar to the results of the phylogenetic analyses. For the more conservative 28S gene, the wasps associated with the two host plant species can be readily separated based on a single indel event at position 181, except for one single sample (R28), which was associated with rosin weed, but without the extra T at 181 position found in other wasps associated with rosinweed. Except for this indel and an additional transitional mutation for R28, there was no variation throughout the entire 524 bp length of the 28S gene. These results do not provide unambiguous support for the cryptic species hypothesis, but revealed significant discontinuity between the gall wasp gene pools associated with the two host plant species, which may indicate incipient host plant mediated speciation.