The natural dispersal of Tribolium castaneum Herbst (Coleoptera: Tenebrionidae) has been emulated in the laboratory for more than 50 years, using a simple dispersal apparatus. This has typically comprised of a starting container (initial resource or patch) connected by tubing, which contains thread for the animals to climb into a tube and hence to an end container. That is, beetles move to a new viable resource or patch from an inter-patch zone or non-viable habitat. We modified this basic apparatus design to test the effect of tubing length and tubing insertion angle on the dispersal rate and proportion of successful dispersers. We expected that the proportion of successful dispersers would be repeatable within each apparatus design, and that increasing tubing length and steepness of the insertion angle would reduce dispersal rate and success across apparatus designs. Dispersal increased linearly through time, similarly so for both males and females. The design with the most vertical tubing insertion angle had a lower proportion of successful dispersers. Tubing length also had a negative relationship with dispersal success (as judged by insects reaching the end container), but a significant reduction in dispersal success was only apparent between the shortest and longest tubing between containers. We suggest that locating and climbing the vertical section of string before they can enter the tubing between containers restricts dispersal and that at higher densities, insects exhibit greater inclination to climb. This type of apparatus has flexible design tolerances and further potential to study the dispersal of other small insect species that primarily use pedestrian locomotion.

Individuals vary in their ability to disperse. Much of this variation can be described by covarying phenotypic traits that are related to dispersal (constituting the ‘dispersal phenotype’ or ‘dispersal syndrome’), but the nature of the associations among these traits is not well understood. Unravelling the associations among traits that potentially constitute the dispersal phenotype provides a foundation for understanding evolutionary trade-offs due to variation in dispersal. Here, we tested five predictions pertaining to the relationships among physiological, morphological and movement traits that are associated with dispersal, using a species with a long history as a laboratory model for studying ecological phenomena, red flour beetles (Tribolium castaneum). We identified a dominant axis of movement ability that describes variation in dispersal-related movement traits. Individuals that scored positively on this axis moved at higher speed, travelled longer distances, had lower movement intermittency and dispersed quicker to a specified area. Relative leg length, but not body size nor routine metabolic rate related positively with movement ability, indicating a likely mechanistic relationship between increased stride length and movement ability. Our data suggest that the dispersal phenotype may be more strongly linked to morphological traits than physiological ones. We demonstrate that associations among many functional traits do not necessarily conform to a priori expectations, and predict that the substantial intraspecific variation in trait values may be important for selection. Movement is a complex behavioural trait, but it has a mechanistic basis in locomotor morphology that warrants further exploration.