Population Structure and Non-nestmate Aggression within Black Carpenter Ant, Camponotus pennsylvanicus

Adam Salyer, Purdue University


Black carpenter ants, Camponotus pennsylvanicus (DeGeer), are an important urban destructive pest native to North America. Little is known about the dispersal, ecology, or behavior of the black carpenter ant due to its cryptic nesting habitats and nocturnal ecology. Knowledge of C. pennsylvanicus population structure and intraspecific competition is needed to predict potential range expansion, dispersal, and establishment of future black carpenter ant populations in the context of climate change. This research investigated population structure and potential prevalence of increased aggression toward neighboring colonies, known as the “nasty-neighbor” effect, within C. pennsylvanicus. Within this research, I test the hypothesis that black carpenter ant dispersal is slow and therefore neighboring colonies are more closely related than distant colonies. Ninety ant colonies were collected throughout nine Indiana and four surrounding state counties. Molecular characterization included Illumina high throughput sequencing of two mtDNA genes (CO1 and CO2) and two rDNA loci (ITS1 and ITS2) within 40 colonies and scoring six microsatellite loci within 90 colonies on a Sanger sequencer. Maximum likelihood analysis of both trees revealed no discernable population structure or gene flow. Illumina and microsatellite scored datasets combined to generate a consensus tree with higher bootstrap support but failed to reveal nested spatial clustering. These three analyses provide strong evidence that black carpenter ant population range extends throughout the entirety of the Midwest. Conspecific outgroups were always recovered together, however maximum likelihood analysis failed to cluster together outgroup specimens in all three analyses. Black carpenter ant intercolony aggression was examined to reveal any support for significantly increased or decreased aggression in correlation with conspecific genetic or geographic distance. Fifteen colonies were collected and subjected to paired nestmate discrimination assay trials. A Mantel test determined genetic and geographic distance factors do not share collinearity and thus these variables were included as fixed effects in model analyses. Data obtained utilizing the 1 – 4 integer scale included maximum aggression score, relative aggression level proportion per trial, and aggression interaction index. These varied forms of data were examined for any correlation of aggression to genetic or spatial distance within ANOVA, cumulative link mixed model, linear mixed-effect model, and generalized linear mixed model analyses. ANOVA results revealed significant differences of mean aggression proportion expressed as well as significantly different mean aggression proportion received. None of the three model analyses found any significant relationship between aggression and genetic or geographic distance. Intercolony crossings nearly always revealed some form of aggression, confirming standard dyad field assays are effective ways to delineate colony relationships.




Bennett, Purdue University.

Subject Area

Entomology|Genetics|Behavioral Sciences

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server