top of page

The first step to understand how spatial sorting can influence metapopulation dynamics is to determine which ecologically important traits are genetically and phenotypically correlated with dispersal. To accomplish this goal, I artificially spatially sorted individuals by creating an environment where females sensitive to competition will disperse long distances to locate resources without competition. I artificially selected for beetles that did not disperse or dispersed long distances in response to this environmental manipulation. I used these selected lines to evaluate the phenotypic and genetic correlation of multiple life-history traits with dispersal behavior.


I found that immediately after artificial selection, individuals from the dispersal selected lines moved farther than individuals from the non-disperser lines when placed in an environment with competition, suggesting that we did select for density-dependent dispersal in our experiment. I compared two other measures of dispersal which were not directly selected in the experiment: general activity and dispersal tendency. I found there were no significant differences in general activity or dispersal tendency among the short and long-distance disperser populations. It may be that the dispersal phenotype is triggered by the environment because I observed dispersal differences among populations when there was competition, but for the activity and dispersal tendency behavioral tests, there was no competition in the environment.

For both fecundity and body mass, there was a strong positive genetic correlation with dispersal in the disperser populations, however, for the non-disperser populations, there is no evidence of a genetic correlation between either trait and dispersal. Additionally, I found evidence that the populations evolved for dispersal also evolved larger bodies and a decreased lifetime fecundity. I found no evidence that 24-hr fecundity, egg size, lifespan, larval development time or larval competitive ability evolved differently between the disperser and non-disperser populations.

When beetles spatially sort in fragmented landscapes, as simulated in our experiment, indirectly sorted traits (e.g., body mass and fecundity) can interact with the local environment, ultimately leading to evolution of the population. I propose fecundity as a life-history trait that, through its phenotypic and genetic correlation with dispersal, is a likely route for spatial sorting - a consequence of individual differences in dispersal - to alter the population dynamics of fragmented populations.


Dispersal response to artificial selection, measured as the farthest dish where a female laid at least one egg across 7 generations of selection. Results for individual females put through the arrays are shown as jittered, unfilled points. Population means are indicated by black symbols and lines. Generation 7 is skipped in the figure because this was a rest month where no selection was conducted. Horizontal lines show the average dish where a female laid at least one egg in the control population at the start of the experiment (dashed line) and end of the experiment (dotted line). 

bottom of page