Author's School

Graduate School of Arts & Sciences

Author's Department/Program

Biology and Biomedical Sciences: Evolution, Ecology and Population Biology


English (en)

Date of Award

Summer 8-12-2013

Degree Type


Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Kenneth M. Olsen


Agricultural weeds have evolved to compete aggressively with domesticated plants in agricultural environments. Although the evolution of invasiveness has been studied extensively in natural ecosystems, few comparable studies have been conducted using agricultural weeds. In this dissertation, I used the genus Amaranthus to examine agricultural weed evolution over different evolutionary time scales, ranging from fitness measurements within a single species to a genus-wide, macroevolutionary analysis.

To explore the recent evolution of agricultural invasiveness, I studied a native Midwestern species, A. tuberculatus: waterhemp), which has become an aggressive agricultural weed only within the last several decades. I used microsatellite markers to investigate the present-day population structure of A. tuberculatus. To assess intraspecific variation in agricultural adaptation, I conducted a common garden study measuring the relative fitness of plants from across the species range in experimental soybean plots. I discovered two genetic subpopulations. The 20th century invasion of Midwestern agricultural fields was due to the eastward migration of the "western" genetic subpopulation, which has high competitive fitness in soybean fields and which may have been preadapted to the agricultural environment.

Waterhemp has rapidly evolved resistance to multiple classes of herbicides. The role of native Midwestern riverbank populations in this process is unknown. I screened agricultural and riverbank populations of A. tuberculatus in Ohio for a common agricultural resistance mutation, using a combination of herbicide resistance phenotyping, PCR genotyping, and gene sequencing. I found that the most common agricultural mutation was indeed present in riverbank populations, suggesting that these native populations may serve as a reservoir of resistance alleles.

Finally, I constructed a phylogeny for the genus Amaranthus to investigate traits associated with the evolution of weediness. Amaranthus is a worldwide genus of 70 species, with no previous generic phylogeny. I included 58 species and two outgroups, sequenced at four nuclear genes and two chloroplast regions, in my molecular phylogenetic analyses using maximum parsimony, maximum likelihood, and Bayesian inference. Weediness exhibits no phylogenetic signal in Amaranthus; however, using non-phylogenetic statistical tests, I found associations of weediness with several morphological and ecological traits in the genus.


Permanent URL: